Notices
Results 1 to 95 of 95

Thread: Controlled delayed quantum erasure - where is the causality?

  1. #1 Controlled delayed quantum erasure - where is the causality? 
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    In quantum eraser experiments, getting information about one entangled photon decides if the second photon behaves classically or quantum (interfere). Optical lengths for these photons chooses time order of these events, so we can delay the "decision" to happen after what it decides about. But in "standard version" of such delayed choice quantum erasure this decision is made randomly by physics.
    I've just found much stronger version - in which we can control this decision affecting earlier events.
    Here is a decade old Phys. Rev. A paper about its successful realization and here is simple explanation:

    We produce two entangled photons - first spin up, second spin down or oppositely.
    Photon s comes through double slit on which there are installed two different quarter wave plates changing polarization to circular in two different ways.
    Finally there are two possibilities:
    x y R L
    y x L R
    where columns are: linear polarization of p, initial linear polarization of s, circular polarization of s after going through slit 1, circular polarization of s after going through slit 2.
    So if we know only the final circular polarization of s, we still don't know which slit was chosen, so we should get interference. But if we additionally know if p is x or y, we would know which slit was chosen and so interference pattern would disappear.
    So let us add polarizer on p path - depending on its rotation we can or cannot get required information - rotating it 45deg we choose between classical and interfering behavior of s ... but depending on optical lengths, this choice can be made later ...

    Why we cannot send information back in time this way?
    For example placing s detector in the first interference minimum - while brightness of laser is constant, rotating p polarizer should affect the average number of counts of s detector.
    What for? For example to construct computer with time loop using many such single bit channels - immediately solving NP hard problems like finding satisfying cryptokeys (used to decrypt doesn't produce noise):

    Physics from QFT to GRT is Lagrangian mechanics - finds action optimizing history of field configuration - e.g. closing hypothetical causal time-loops, like solving the problem we gave it.
    Ok, the problem is when there is no satisfying input - time paradoxes, so physics would have to lie to break a weakest link of such reason-result loop.
    Could it lie? I think it could - there is plenty of thermodynamical degrees of freedom which seems random for us, but if we could create additional constrains like causal time loops, physics could use these degrees of freedom to break a weakest link of such loop.

    What is wrong with this picture?


    Last edited by Jarek Duda; March 6th, 2012 at 06:02 AM.
    Reply With Quote  
     

  2.  
     

  3. #2  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    Just a quick stab in the dark, but based on my understanding of the Kim, Scully et al. experiment, you need to use the coincidence counter to recover the interference pattern at the signal photon detector, which can only be done after the choice has been made for their entangled partners and they have been subsequently detected and that information sent to the coincidence counter. I may well be wrong, but it sounds to me like you are using a method to try to guess whether the signal photons are part of an interference pattern or not, by rotating the polariser.

    But the signal and partner photons are not just entangled between themselves, they are also somewhat entangled with the state of the quarter wave plate and the polariser. Wouldn't that make a difference?


    Reply With Quote  
     

  4. #3  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    Yes, the coincidence counter seems to be important here ...
    But what exactly is it doing? It is only determining the moments in which the measurements was important.
    So let us just take long enough time interval (like a few seconds) in which the p polarizer had fixed angle - now the exact moments found by the coincidence counter doesn't longer seem to be important ... and still the optical lengths time difference could be much larger than used time interval.

    So we just fix detector s in e.g. the first interference minimum, fix constant laser brightness - and now by just rotating p polarizer, we should affect the counts per second of the s detector ... not true?
    Reply With Quote  
     

  5. #4  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    Quote Originally Posted by Jarek Duda View Post
    Yes, the coincidence counter seems to be important here ...
    But what exactly is it doing? It is only determining the moments in which the measurements was important.
    It is the only way to do it. The coincidence counter is used to "vet" the data as it is collected, to relate the position of the hits at each detector with those of their entangled partner on the other detector, using the two different light travel times since emission. It does this to eliminate the noise from non entangled hits on the detectors and is the only way to recover the interference pattern from the laboratory noise. It receives a hit from the signal detector and knows how long to wait till it should receive an entangled hit on the partner detector. If there is a hit on the other detector in the alloted "time bucket" it is "satisfying return". Hence the interference pattern is only "found" after the entangled partners have hit each detector.

    Quote Originally Posted by Jarek Duda View Post
    So let us just take long enough time interval (like a few seconds) in which the p polarizer had fixed angle - now the exact moments found by the coincidence counter doesn't longer seem to be important ... and still the optical lengths time difference could be much larger than used time interval.
    I'm not sure the coincidence counter works quite in the way you think it does. It is still the only way to eliminate the noise. This isn't just about the setting the difference in elapsed time since emission of both detectors, it is about which of those "time buckets" correlate, and which don't, which is a random sequence of events.
    Reply With Quote  
     

  6. #5  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    [QUOTE=SpeedFreek;311697]It does this to eliminate the noise from non entangled hits on the detectors (..)[\QUOTE]
    If I understand it correctly, selecting only entangled photons is simple task - for example because they have larger wavelength than the pumping light: Spontaneous parametric down-conversion - Wikipedia, the free encyclopedia
    I think the only purpose of the coincidence counter is to eliminate entangled photons which was caught by only one detector?
    More precisely - the observed pattern could be affected only by cases that s was caught, while p photon came through polarizer but wasn't caught by the detector - what (interference?) pattern is expected for these s photons?
    If these cases cannot completely compensate the effect from coincidence photons (for example just don't depend on rotation of p polarizer), still rotating the polarizer would affect the pattern of s.
    Reply With Quote  
     

  7. #6  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    No, the purpose of the coincidence counter is to provide us with the data set from which we can recover the interference pattern.

    Coincidence counting (physics) - Wikipedia, the free encyclopedia
    Reply With Quote  
     

  8. #7  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    There are 3 cases the coincidence counter allows us to distinguish between here:
    1) both detectors catch photon - f^2 probability
    2) only s detector - f*(1-f) probability
    3) only p detector - f*(1-f) probability
    where f is detector effectiveness - they catch f of photons.
    The s pattern with coincidence counter is made of 1) cases, without coincidence counter is made of 1) and 2) cases.
    Assume that distinguishing these cases, the s pattern is correspondingly rho1 and rho2 (shouldn't depend on f?)
    So without the coincidence counter, the density would be
    f^2*rho1 + f(1-f)*rho2
    rho1 depends on rotation of p polarizer. Can this sum never longer depends on it? (for any f!)
    Can rho2 always compensate dependence on rotation of p polarizer?

    If not, without coincidence counter the effect might be weaker, but it still will be there and so can be used.
    Reply With Quote  
     

  9. #8  
    Time Lord
    Join Date
    Mar 2007
    Posts
    8,142
    Quote Originally Posted by SpeedFreek View Post
    Just a quick stab in the dark, but based on my understanding of the Kim, Scully et al. experiment, you need to use the coincidence counter to recover the interference pattern at the signal photon detector, which can only be done after the choice has been made for their entangled partners and they have been subsequently detected and that information sent to the coincidence counter. I may well be wrong, but it sounds to me like you are using a method to try to guess whether the signal photons are part of an interference pattern or not, by rotating the polariser.

    But the signal and partner photons are not just entangled between themselves, they are also somewhat entangled with the state of the quarter wave plate and the polariser. Wouldn't that make a difference?
    I don't see how the use of a coincidence counter invalidates the experiment's findings. Clearly some means must be used to determine whether the interference pattern or lack of interference pattern matches the detection or non-detection of "which path" information. Without that, then.... by very definition we could have no results.

    If this kind of experiment isn't good enough to demonstrate the effect, then what kind of experiment would be? Would you prefer an experiment where we don't compare the photons that form or don't form 2 slit interference patterns with their entangled "twins" to see if "which path" information had been recorded? We just speculate that they must have matched and don't confirm it?

    Naturally, that confirmation will have to take place after all the results are in. How else would you carry out the confirmation? Confirm it before the results in?
    Some clocks are only right twice a day, but they are still right when they are right.
    Reply With Quote  
     

  10. #9  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    Quote Originally Posted by kojax View Post
    Clearly some means must be used to determine whether the interference pattern or lack of interference pattern matches the detection or non-detection of "which path" information. Without that, then.... by very definition we could have no results.
    To determine if the interference pattern occurred, it should be enough to place s detector in the first interference minimum - intensity should depend on rotation of p polarizer ... at least for incident photons, but the rest of them (rho2) can only weaken the effect a bit.

    If this kind of experiment isn't good enough to demonstrate the effect, then what kind of experiment would be?
    I see also another looking practical possibility I cannot disprove - CPT analogue of laser - lasar (stimulated absorption).
    To see that it seems doable, imagine free electron laser - we enforce electron to move on sinus-like curve, emitting photons ... which finally e.g. are absorbed by some target.
    Physics is CPT invariant, so let us imagine such tranformation of this picture - excited target emit photons, which fly to the lasar and finally are absorbed by positron going in reverse direction.
    So such free electron laser should also work as lasar - but to make it work, it has to (anti-)hit target which is already excited to given specific wavelength - it doesn't occur often.
    Imagine we constantly excite the target to required energy (e.g. is sodium lamp) and it is surrounded in all but to the lasar directions by detectors - they usually get the produced light, but if we turn the lasar on, more energy should goin that direction and so we should see a disturbance in energy balance in the lamp-detectors system ... before turning the lasar on by the optic length.
    What if we make CPT transformation of free electron laser?
    What's wrong also with this picture?
    Reply With Quote  
     

  11. #10  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    I think there are some misunderstandings here. I was not at all saying the coincidence counter affects the experiment, I am saying it is absolutely necessary to use it to correlate the entangled hits and recover the interference pattern, as there is no other way to do it. This is why all delayed erasure experiments have to use one. A visible interference pattern doesn't actually appear at the s detector, it has to be recovered from the data by correlating the erased p photons hits on the p detector with their entangled partners at the s detector.

    Some pertinent quotes from the paper itself:
    We have shown that interference can be destroyed, by marking the path of the interfering photon, and recovered, by making an appropriate measurement on the other entangled photon.
    The possibility of delayed erasure generated a discussion with respect to its legitimacy, with the argument that it would be possible, in this way, to alter the past. This argument is founded on an erroneous interpretation of the formalism of quantum mechanics.
    Reply With Quote  
     

  12. #11  
    Time Lord
    Join Date
    Mar 2007
    Posts
    8,142
    Quote Originally Posted by SpeedFreek View Post
    I think there are some misunderstandings here. I was not at all saying the coincidence counter affects the experiment, I am saying it is absolutely necessary to use it to correlate the entangled hits and recover the interference pattern, as there is no other way to do it. This is why all delayed erasure experiments have to use one. A visible interference pattern doesn't actually appear at the s detector, it has to be recovered from the data by correlating the erased p photons hits on the p detector with their entangled partners at the s detector.
    If the experimenter wanted to, they could take all the registered hits that corresponded with an interference pattern, look at them all, and then afterward take all the registered hits that corresponded with observing "which path" information, and then after looking through these, they could compare the time stamps last, to see if they match.

    I mean, nothing stops them from putting the two data sets on two different computers and then choosing what order to observe everything in.

    Quote Originally Posted by SpeedFreek View Post

    The possibility of delayed erasure generated a discussion with respect to its legitimacy, with the argument that it would be possible, in this way, to alter the past. This argument is founded on an erroneous interpretation of the formalism of quantum mechanics.

    I suspect it's because of that word "alter". You can't take a known past event and then alter it. The wave function has already collapsed from your perspective. Collapsed wave functions can't be made into uncertainties again. However, a wave function that is allowed to remain in a superposed state for a long time without collapsing might be used as a means to send information back to the starting point. Well..... maybe...... That's what makes the Delayed Choice Quantum Eraser so interesting. It's basically taking an effect we expect to see on the quantum level and bringing it to the macro level.

    If you could send information back in time, it would be like Schrodinger's Cat. It's because the past is in a superposed state from your perspective. That means you haven't experienced it yet, so there is no contradiction of causality if you change it. If any part of the system is collapsed, then probably the whole system collapses, and you aren't able to send anything.
    Last edited by kojax; March 6th, 2012 at 05:16 AM.
    Some clocks are only right twice a day, but they are still right when they are right.
    Reply With Quote  
     

  13. #12  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    Quote Originally Posted by SpeedFreek View Post
    I think there are some misunderstandings here. I was not at all saying the coincidence counter affects the experiment, I am saying it is absolutely necessary to use it to correlate the entangled hits and recover the interference pattern, as there is no other way to do it. This is why all delayed erasure experiments have to use one. A visible interference pattern doesn't actually appear at the s detector, it has to be recovered from the data by correlating the erased p photons hits on the p detector with their entangled partners at the s detector.
    For clearer picture, let us maybe look at Mach-Zehnder configuration:

    Now the whole interference pattern is get by two detectors making the situation simpler:
    the path is unknown - everything goes to Ds2,
    the path is known - half goes to Ds1.
    So if anything goes to Ds1, means interference was destroyed - means the path is well defined.
    The path is not known if the polarizer is set to fast axis of one of quater wave plate, is unknown if it is between them (45deg).

    We would like to look at Ds1 and Ds2 to (statistically) distinguish rotation of the polarizer - we already know that it's possible using the coincidence counter.
    Looking at Ds1 and Ds2 without the coincidence counter would mean that additionally we see photons that didn't hit Dp - what statistics they have?
    If they hit Ds and not Dp, one would say that they remain indistinguishable, so all of them should go to Ds2, not true?

    So even without using the coincidence counter, observing photons in Ds1 means the polarizer is not in the 45deg rotation, not true?
    Attached Images
    Last edited by Jarek Duda; March 6th, 2012 at 10:05 AM.
    Reply With Quote  
     

  14. #13  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    Quote Originally Posted by kojax View Post
    If the experimenter wanted to, they could take all the registered hits that corresponded with an interference pattern, look at them all, and then afterward take all the registered hits that corresponded with observing "which path" information, and then after looking through these, they could compare the time stamps last, to see if they match.

    I mean, nothing stops them from putting the two data sets on two different computers and then choosing what order to observe everything in.
    I am not sure precisely what you mean here. When there is no interference pattern displayed at the detector, but just a bell curve, you cannot tell whether any of those hits are part of an interference pattern or not, as there are hits in the minimum areas of an interference pattern. The hits that correspond to observing "which path" information form a bell curve...

    There is no way to tell, by looking at a bell curve, which photons in it might form part of a hidden interference pattern.

    If that is not what you meant, then I must be missing your point.


    Quote Originally Posted by kojax View Post
    Quote Originally Posted by SpeedFreek View Post
    The possibility of delayed erasure generated a discussion with respect to its legitimacy, with the argument that it would be possible, in this way, to alter the past. This argument is founded on an erroneous interpretation of the formalism of quantum mechanics.

    I suspect it's because of that word "alter". You can't take a known past event and then alter it. The wave function has already collapsed from your perspective. Collapsed wave functions can't be made into uncertainties again. However, a wave function that is allowed to remain in a superposed state for a long time without collapsing might be used as a means to send information back to the starting point. Well..... maybe...... That's what makes the Delayed Choice Quantum Eraser so interesting. It's basically taking an effect we expect to see on the quantum level and bringing it to the macro level.

    If you could send information back in time, it would be like Schrodinger's Cat. It's because the past is in a superposed state from your perspective. That means you haven't experienced it yet, so there is no contradiction of causality if you change it. If any part of the system is collapsed, then probably the whole system collapses, and you aren't able to send anything.
    It's not about the word "alter", it is about information not being able to travel backwards in time, at all. No information travels backwards in time in the delayed choice quantum eraser, and no way has been found to use that quantum mechanism to send information back in time - without the ability to correlate between each end of the experiment (which has to happen at a maximum of the speed of light), all you can send back is "noise".
    Reply With Quote  
     

  15. #14  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    Quote Originally Posted by Jarek Duda View Post
    For clearer picture, let us maybe look at Mach-Zehnder configuration
    Okay, let's look at a recent experiment using a Mach-Zehnder interferometer:
    Quantum Eraser - Photon Quantum Mechanics

    I wonder why they chose to use a coincidence counter, if they didn't have to?
    Reply With Quote  
     

  16. #15  
    Time Lord
    Join Date
    Mar 2007
    Posts
    8,142
    Quote Originally Posted by SpeedFreek View Post
    Quote Originally Posted by kojax View Post
    If the experimenter wanted to, they could take all the registered hits that corresponded with an interference pattern, look at them all, and then afterward take all the registered hits that corresponded with observing "which path" information, and then after looking through these, they could compare the time stamps last, to see if they match.

    I mean, nothing stops them from putting the two data sets on two different computers and then choosing what order to observe everything in.
    I am not sure precisely what you mean here. When there is no interference pattern displayed at the detector, but just a bell curve, you cannot tell whether any of those hits are part of an interference pattern or not, as there are hits in the minimum areas of an interference pattern. The hits that correspond to observing "which path" information form a bell curve...

    There is no way to tell, by looking at a bell curve, which photons in it might form part of a hidden interference pattern.

    If that is not what you meant, then I must be missing your point.
    I would think you could simply put a detector at a location where photons would be very unlikely to go unless they're part of an interference pattern, or put one in a place where if there was yes an interference pattern the photons would not go.

    Then every hit that detector registers is overwhelmingly likely to be coming from the photon we expect. We still won't know whether it's entangled or not until we compare the time stamps, but we can observe the pattern first, and the question of entanglement second if we want. (It's quite a bit more tedius, but might resolve some issues.)
    Some clocks are only right twice a day, but they are still right when they are right.
    Reply With Quote  
     

  17. #16  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    To cut to the chase, as it were:

    If we could set up any delayed choice experiment so that an interference pattern is displayed at the signal detector, before the information is erased from the entangled partner particles, we would be able to see into the future. This would cause certain paradoxes, one of which is - now we have an interference pattern at the s detector, what if we now choose not to erase the which path information?

    Even after the which path information has been erased, the interference pattern doesn't suddenly and miraculously leap out of the data collected at the s detector - it is only found when you make your measurement of the p photon and correlate the two to see which photons from the s detector were entangled with erased partners.
    Reply With Quote  
     

  18. #17  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    Quote Originally Posted by kojax View Post
    I would think you could simply put a detector at a location where photons would be very unlikely to go unless they're part of an interference pattern
    There is no such place.

    Quote Originally Posted by kojax View Post
    or put one in a place where if there was yes an interference pattern the photons would not go.
    Again, there is no such place.

    Anywhere in an interference pattern could also be part of a bell curve. It is not about the place itself, it is about the number of hits at that place.

    If all you have is a bell curve, then any photon in it could also be part of a hidden interference pattern.

    How do you know if you have the right photon, or not?
    Reply With Quote  
     

  19. #18  
    Time Lord
    Join Date
    Mar 2007
    Posts
    8,142
    Quote Originally Posted by SpeedFreek View Post

    Anywhere in an interference pattern could also be part of a bell curve. It is not about the place itself, it is about the number of hits at that place.

    If all you have is a bell curve, then any photon in it could also be part of a hidden interference pattern.

    How do you know if you have the right photon, or not?
    Yes, but you could look at a large number of photons hitting that place and conclude that most of them are from an interference pattern or most of them are not from an interference pattern. There may not be any location where not one photon ever hits, but there should be multiple places where very very very few photons from the wrong group are hitting.

    That's good enough to work with. If you had a more perfect entangled photon generator, and instead of using a half silvered mirror to randomly measure or not measure "which path" information for the entangled twin, you had a device that could deliberately choose, then you'd notice that when you deliberately chose one direction the intensity of photons at that detector would drop substantially when you chose one direction, and increase substantially when you chose the other. You don't need it to go from perfect zero to perfect maximum.

    Quote Originally Posted by SpeedFreek View Post
    To cut to the chase, as it were:

    If we could set up any delayed choice experiment so that an interference pattern is displayed at the signal detector, before the information is erased from the entangled partner particles, we would be able to see into the future. This would cause certain paradoxes, one of which is - now we have an interference pattern at the s detector, what if we now choose not to erase the which path information?
    It doesn't create any paradoxes so long as all information concerned is contained in the system we are attempting to analyze. If the person changing the past doesn't know what that past was, then they can't choose to contradict it. If they don't choose to contradict it, then there's no reason to expect it would self contradict by accident. Whatever the future actor causes to happen is what did happen.

    We're taking a quantum effect and expanding it into the macro world. When two or more events are superposed over each other, any decision about the outcome is simultaneous with detection from that effect's perspective. If I had a working version of a device that could send information back 3 hours in time, and I decided to look up a lottery draw that was going to happen then and send the data back to the present, one of two things would happen.

    1) - My future self comes through and I see the numbers now.

    2) - My future self decides to change his mind and I don't see any numbers now.

    The possibility that involves a contradiction (me seeing the numbers and then deciding not to send them) is excluded from the realm of possible outcomes. The whole series of events is one event.
    Some clocks are only right twice a day, but they are still right when they are right.
    Reply With Quote  
     

  20. #19  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    Quote Originally Posted by kojax View Post
    Quote Originally Posted by SpeedFreek View Post
    How do you know if you have the right photon, or not?
    Yes, but you could look at a large number of photons hitting that place and conclude that most of them are from an interference pattern or most of them are not from an interference pattern. There may not be any location where not one photon ever hits, but there should be multiple places where very very very few photons from the wrong group are hitting.
    In the places where very very few from the wrong group are hitting, very very many of the right group are hitting. Such is the nature of photons either interfering or not interfering with themselves.

    Quote Originally Posted by kojax View Post
    It doesn't create any paradoxes so long as all information concerned is contained in the system we are attempting to analyze. If the person changing the past doesn't know what that past was, then they can't choose to contradict it. If they don't choose to contradict it, then there's no reason to expect it would self contradict by accident. Whatever the future actor causes to happen is what did happen.
    The only way the information can remain in the system and not cause paradoxes is if the observer in the past cannot know what choice the future actor makes. Especially if they are going to be that future actor. This is the situation with the delayed choice quantum eraser.

    Quote Originally Posted by kojax View Post
    We're taking a quantum effect and expanding it into the macro world. When two or more events are superposed over each other, any decision about the outcome is simultaneous with detection from that effect's perspective. If I had a working version of a device that could send information back 3 hours in time, and I decided to look up a lottery draw that was going to happen then and send the data back to the present, one of two things would happen.

    1) - My future self comes through and I see the numbers now.

    2) - My future self decides to change his mind and I don't see any numbers now.

    The possibility that involves a contradiction (me seeing the numbers and then deciding not to send them) is excluded from the realm of possible outcomes. The whole series of events is one event.
    Very nice!

    So, we can conclude then, that in a delayed choice experiment, if we see no interference pattern at the S detector, nothing can make us press the erase button in the future. Or, if we see an interference pattern, nothing can prevent us from pressing it.

    Does that sound right, to you? I mean, I have heard of people taking the observer affects experiment effect too literally, but here we have experiment affects observer, don't we?
    Reply With Quote  
     

  21. #20  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    Quote Originally Posted by SpeedFreek View Post
    Quote Originally Posted by Jarek Duda View Post
    For clearer picture, let us maybe look at Mach-Zehnder configuration
    Okay, let's look at a recent experiment using a Mach-Zehnder interferometer:
    Quantum Eraser - Photon Quantum Mechanics
    I wonder why they chose to use a coincidence counter, if they didn't have to?
    What does it prove? How do you conclude that without the coincidence counter the dependence from rotation of the polarizer would disappear?
    Without CC the only difference is that we would also include photons which entangled partner didn't hit the second detector (Dp) - what statistics they have?
    If we don't know their path, they should interfere ... not depending on the rotation of polarizer.
    So finally without the CC, we would see both coinciding photons depending on polarizer rotation and which didn't hit Dp - not depending on rotation of polarizer - sum of depending and not depending has to depend - without coincidence counter the dependence would be weaker, but still nonzero.

    If we could set up any delayed choice experiment so that an interference pattern is displayed at the signal detector, before the information is erased from the entangled partner particles, we would be able to see into the future. This would cause certain paradoxes, one of which is - now we have an interference pattern at the s detector, what if we now choose not to erase the which path information?
    Not true - you assume we can make perfect causal loops, while nobody claims that.
    Biology, electronics and especially such hypothetical backtime information channels are based on thermodynamics - degrees of freedom which now behave randomly for us.
    But if we would be able to make constrains with causal time loop, physics could make these thermodynamical degrees of freedom less random ... for example to break the weakest link of the loop but making it lie - e.g. the backtime channel.

    Physics from QFT to GRT is Lagrangian mechanics - finding optimizing action history of field configuration.
    It doesn't contradict using constrains with causal loops, as long they are imperfect - leave physics a place to lie, breaking such loop.
    Hypothetical wormholes would allow to construct a perfect loop, like a cannon which shoots to itself if and only if in doesn't do it - it could be so large that thermodynamics couldn't prevent it ... it's why I don't believe in wormholes.
    Reply With Quote  
     

  22. #21  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    Quote Originally Posted by Jarek Duda View Post
    What does it prove? How do you conclude that without the coincidence counter the dependence from rotation of the polarizer would disappear?
    You seem to be completely missing the point too. I am not , I repeat not, saying the dependence disappears without the CC!! I am saying it is still there, but you cannot measure it without the CC! There is no way to know which photons have had their path information "erased" at the s detector, BEFORE their entangled partners have been erased. I really is that simple.

    Why not take your proposal to someone who can actually try it out, or someone who has already performed these kind of experiments, and ask them why they HAVE TO USE A CC to recover the interference pattern from the data?
    Reply With Quote  
     

  23. #22  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    Quote Originally Posted by SpeedFreek View Post
    You seem to be completely missing the point too. I am not , I repeat not, saying the dependence disappears without the CC!! I am saying it is still there, but you cannot measure it without the CC! There is no way to know which photons have had their path information "erased" at the s detector, BEFORE their entangled partners have been erased. I really is that simple.
    I agree - you don't know which exactly photons had their path information "erased", but you can use the statistics - rotating polarizer doesn't affect the number of not coinciding photons, affects the number of coinciding ones - so finally without CC counter the total number of photons per time unit is still affected by polarizer rotation - not true?

    Statistics is one of many links of hypothetical causal loops that physics would always be able to eventually break on statistical level - making such channels always imperfect.
    Last edited by Jarek Duda; March 7th, 2012 at 02:27 PM.
    Reply With Quote  
     

  24. #23  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    Quote Originally Posted by Jarek Duda View Post
    I agree - you don't know which exactly photons had their path information "erased", but you can use the statistics - rotating polarizer doesn't affect the number of not coinciding photons, affects the number of coinciding ones - so finally without CC counter the total number of photons per time unit is still affected by polarizer rotation - not true?
    The total number of photons hitting the s detector remain the same, however you rotate the polarizer between the BBO and the p detector. This is about whether they have had their which path information erased or not, which has no effect on the signal photons probability of hitting the s detector - does it?

    Last edited by SpeedFreek; March 7th, 2012 at 06:57 PM. Reason: No point saying the same thing twice.
    Reply With Quote  
     

  25. #24  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    There are two "types" of photons hitting detector s - those which entangled partner hit detector p (coinciding) and those which didn't (not coinciding), not true?
    While rotating polarizer on p path, the amount of photons from the first group is affected, from the second is not affected, so the total number of photons hitting detector s is affected by rotation of p polarizer, not true?

    --------------------------------------------------------
    For action optimizing Lagrangian mechanics like QFT or GRT, we should think of particles as their trajectories.
    For example linear polarizer means that trajectory is fixed in one of two possibilities - like for string, angular tension makes that surrounding feels this tension - surrounding in both directions.
    So using the scheme of time-loop computer from the first post is like asking for fixing all polarizations to minimize total tension - action ... but sometimes (like paradoxes) such minimization means that physics had to lie to break this causal loop - for example spoiling statistics we used in hypothetical backtime channel.

    Imposing imperfect causal loop constrain to lagrangian mechanics seems to be against intuition, while "standard" quantum computers also do similar stuff (wider explaination).
    It is said that their strength is the reversible calculations, while we can also make them classically like
    (x,y,z)<->(x,y,z xor f(x,y))
    the problem with trying to reverse such calculations are the auxiliary bits - we don't know how to initialize them...
    Here is the real strength of quantum computers - that the measurement is kind of attaching trajectories in the future - for example "selecting" arguments leading to the same value of function in Shor algorithm:

    Last edited by Jarek Duda; March 8th, 2012 at 05:50 PM.
    Reply With Quote  
     

  26. #25  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    Quote Originally Posted by Jarek Duda View Post
    There are two "types" of photons hitting detector s - those which entangled partner hit detector p (coinciding) and those which didn't (not coinciding), not true?
    Agreed. The photons hitting the s detector include the photons whose entangled partner did not hit the p detector. You said it yourself!

    Quote Originally Posted by Jarek Duda View Post
    While rotating polarizer on p path, the amount of photons from the first group is affected, from the second is not affected, so the total number of photons hitting detector s is affected by rotation of p polarizer, not true?
    As you just said yourself, rotating the polarizer has an effect on the total amount of photons hitting the p detector, but it does not have an effect on the total amount of photons hitting the s detector. The photons hitting the s detector include the group whose entangled partner did not hit the p detector, so rotating the polarizer, which may cause a different amount of partner photons from hitting the p detector, has no effect on the total number of hits at the s detector.

    Do you see what I mean yet? Are you beginning to understand why we have to use the coincidence counter in the first place?
    Reply With Quote  
     

  27. #26  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    Quote Originally Posted by SpeedFreek View Post
    The photons hitting the s detector include the group whose entangled partner did not hit the p detector, so rotating the polarizer, which may cause a different amount of partner photons from hitting the p detector, has no effect on the total number of hits at the s detector.
    As many post above, denote rho1 the density of photons hitting s, which entangled partner hit p and by rho2 - density of photons hitting s, which entangled partner did not hit p.
    So rho1 depends on angle of polarizer, while rho2 doesn't.
    Without coincidence counter we observe some linear combination of them: let say a*rho1+b*rho2. Does it depend on the angle of polarizer?
    Reply With Quote  
     

  28. #27  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    Quote Originally Posted by Jarek Duda View Post
    Quote Originally Posted by SpeedFreek View Post
    The photons hitting the s detector include the group whose entangled partner did not hit the p detector, so rotating the polarizer, which may cause a different amount of partner photons from hitting the p detector, has no effect on the total number of hits at the s detector.
    As many post above, denote rho1 the density of photons hitting s, which entangled partner hit p and by rho2 - density of photons hitting s, which entangled partner did not hit p.
    So rho1 depends on angle of polarizer, while rho2 doesn't.
    Without coincidence counter we observe some linear combination of them: let say a*rho1+b*rho2. Does it depend on the angle of polarizer?
    No. Some of the most brilliant minds in quantum physics have been trying to find a way around so called "cosmic censorship" and how it applies in a delayed choice experiment for decades, and so far they have failed.

    What you are basically saying is that more photons hit the s detector when their entangled partners are in one state, than when their entangled partners are in the other state, when the truth is that it is the same photons hitting the s detector, whatever the state of their entangled partners. Any photon density change only occurs at the p detector, due to the polarizer.

    Don't you think they might have thought of this already? I mean, it would be very simple to test it wouldn't it?
    Reply With Quote  
     

  29. #28  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    Quote Originally Posted by SpeedFreek View Post
    What you are basically saying is that more photons hit the s detector when their entangled partners are in one state, than when their entangled partners are in the other state, when the truth is that it is the same photons hitting the s detector, whatever the state of their entangled partners. Any photon density change only occurs at the p detector, due to the polarizer.
    I'm not sure what you are saying that I'm saying but I'm trying to tell something really simple. Let me repeat some above post:
    There are 3 cases the coincidence counter allows us to distinguish between here:
    1) both detectors catch photon - f^2 probability
    2) only s detector - f*(1-f) probability
    3) only p detector - f*(1-f) probability
    where f is detector effectiveness - they catch f of photons.
    The s pattern with coincidence counter is made of 1) cases, without coincidence counter is made of 1) and 2) cases.
    Assume that distinguishing these cases, the s pattern is correspondingly rho1 and rho2
    So without the coincidence counter, the density observed by s would be
    f^2*rho1 + f(1-f)*rho2
    Not true?
    Reply With Quote  
     

  30. #29  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    Quote Originally Posted by Jarek Duda View Post
    Quote Originally Posted by SpeedFreek View Post
    What you are basically saying is that more photons hit the s detector when their entangled partners are in one state, than when their entangled partners are in the other state, when the truth is that it is the same photons hitting the s detector, whatever the state of their entangled partners. Any photon density change only occurs at the p detector, due to the polarizer.
    I'm not sure what you are saying that I'm saying but I'm trying to tell something really simple.
    Yes, I know what you are saying is really simple. I would imagine it is the same thought that most of us have when we first hear about the delayed choice quantum eraser (I know I did!), but unfortunately it just doesn't work. If it did work like that, we would be able to send information backwards through time, which would be very interesting indeed.

    Quote Originally Posted by Jarek Duda View Post
    Let me repeat some above post:
    There are 3 cases the coincidence counter allows us to distinguish between here:
    1) both detectors catch photon - f^2 probability
    2) only s detector - f*(1-f) probability
    3) only p detector - f*(1-f) probability
    where f is detector effectiveness - they catch f of photons.
    The s pattern with coincidence counter is made of 1) cases, without coincidence counter is made of 1) and 2) cases.
    Assume that distinguishing these cases, the s pattern is correspondingly rho1 and rho2
    So without the coincidence counter, the density observed by s would be
    f^2*rho1 + f(1-f)*rho2
    Not true?
    Not true. An s photon is detected at the s detector. The coincidence counter then waits to see if its entangled p photon hits the p detector. If yes, we have a coincidence count, and if no, there is no coincidence. The overall density of photons hitting s is the same, regardless of whether their partner has been erased or not, or is detected or not. So, without the coincidence counter there is no way to know whether the s photon is part of a bell curve or part of an interference pattern.

    Without a coincidence counter, the photons detected at the s detector will either be part of group 1, or group 2. If you change the angle of the polarizer, the photons hitting the s detector will still either be part of group 1 or group 2. The total amount of photons will be the same. You seem to be trying to count the same photons twice.

    No photons are added to, or subtracted from, the stream of photons leaving each QWP and reaching the s detector. The density doesn't change.
    Reply With Quote  
     

  31. #30  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    Quote Originally Posted by SpeedFreek View Post
    Yes, I know what you are saying is really simple. I would imagine it is the same thought that most of us have when we first hear about the delayed choice quantum eraser (I know I did!), but unfortunately it just doesn't work. If it did work like that, we would be able to send information backwards through time, which would be very interesting indeed.
    In Lagrangian mechanics fixing ends determines the trajectory - as action minimizing. This picture is time symmetric. Shifting the beginning modify future path, shifting the end modify past path.
    QFT and GRT are also Lagrangian mechanics - finds action optimizing history of field configuration.
    Why do you think it contradicts imperfect constrains as causal time-loops?

    Without a coincidence counter, the photons detected at the s detector will either be part of group 1, or group 2. If you change the angle of the polarizer, the photons hitting the s detector will still either be part of group 1 or group 2. The total amount of photons will be the same. You seem to be trying to count the same photons twice.
    The only purpose of coincidence counter is to distinguish photons from group 1 and 2 - for example allowing to measure statistics while restricting to the group 1.
    Without CC, we could only see the total number of photons from both groups - sum of number of those from group 1 and from group 2. The first number depends on polarizer rotation, the second not - so their sum depends.
    Reply With Quote  
     

  32. #31  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    I think we must be having a miscommunication issue here, and perhaps I am misunderstanding you. I am guessing English is not your first language, and although your English is very good, it is a little unusual in places and difficult to understand.

    I fully accept that shifting the beginning modifies the future path and shifting the end modifies the past path. I am not denying any of the findings of the delayed choice quantum eraser. Yes, "spooky action at a distance" (as Einstein put it) works backwards through time. But we can't use it to transmit information faster than c, or backwards through time.

    The issue is that, although we can consider the future backward modification of the past path as valid, due to the analysis of the measurements made at both detectors, we have to do it this way. In reality future backwards modification can only ever be validated after seeing the end result of the future path, which is why it cannot be used to send information backwards through time, which was your original question.

    I assure you, physicists have been checking for this for decades. There is no way to determine whether the which path information has been erased until the p photons are detected. There is no way to determine whether there is a hidden interference pattern at the s detector, or not, before the p photons have been detected.

    Once we detect the p photons and correlate them with the partners, it seems that the interference pattern was hidden in the data from the s detector all along.

    It seems as if the universe is conspiring against us, as whenever we come across a situation that might violate causality, the laws of physics manage to prevent any such violation. We cannot use quantum entanglement to transmit information faster than light, or backwards through time, without having to correlate our findings with the other end of the future path.

    Here are the results from another delayed choice experiment, and before you say it, it makes no difference which set-up you use, it would be the same situation with any delayed choice setup:





    Here, D0 is the signal detector.
    The quantum eraser experiment :: Strange Paths

    Now then, this isn't to say that we cannot use quantum entanglement to compute encryption keys and send a message to a recipient where the decryption key doesn't exist until after the encrypted message has arrived, of course. Or to compute very difficult problems in a very quick time frame. We can do this in theory anyway. But it has nothing to do with the "backwards through time" part of the delayed choice experiment.
    Reply With Quote  
     

  33. #32  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    Quote Originally Posted by SpeedFreek View Post
    But we can't use it to transmit information faster than c, or backwards through time.
    I really understand that there is great belief and intuition for that, but I'm far from being convinced that there is something more behind than faith ... and I don't see any real counterarguments from you. I don't see why Lagrangian mechanics would forbid something like that. What I see is confirmation bias.
    The realization you've linked is again the version from wikipedia - in opposite to the one from this discussion, it doesn't allow to control the erasure.
    Ok, let us maybe try to discus another hypothetical possibility I've already written.

    Do you believe in the the basis of e.g. QFT: CPT symmetry conservation?
    If so, why we cannot construct CPT analogue of laser - lasar (stimulated absorption)?
    To see that it seems doable, imagine free electron laser - we enforce electron to move on sinus-like curve, emitting photons ... which finally e.g. are absorbed by some target.
    If physics is CPT invariant, let us imagine such transformation of this picture - excited target emit photons, which fly to the lasar and finally are absorbed by positron going in reverse direction.
    So such free electron laser should also work as lasar - but to make it work, it has to (anti-)hit target which is already excited to given specific wavelength - it doesn't occur often.
    Imagine we constantly excite the target to required energy (e.g. is sodium lamp) and it is surrounded in all but to the lasar directions by detectors - they usually get the produced light, but if we turn the lasar on, more energy should going that direction and so we should see a disturbance in energy balance in the lamp-detectors system ... before turning the lasar on by the optic length.
    What's wrong also with this picture?
    Reply With Quote  
     

  34. #33  
    Moderator Moderator Markus Hanke's Avatar
    Join Date
    Nov 2011
    Location
    Ireland
    Posts
    7,168
    I quote from the Wikipedia article for "Delayed Choice Quantum Eraser" :

    The apparatus under discussion here could not communicate information in a retro-causal manner because it takes another signal, one which must arrive via a process that can go no faster than the speed of light, to sort the superimposed data in the signal photons into four streams that reflect the states of the idler photons at their four distinct detection screens.
    Also, retro-causal quantum relationships would violate the Eberhard-Ross theorem :

    Foundations of Physics Letters, Volume 2, Number 2 - SpringerLink

    and I don't see any real counterarguments from you. I don't see why Lagrangian mechanics would forbid something like that. What I see is confirmation bias.
    I think that the Eberhard theorem is a pretty strong counterargument; it works off the assumption that relativistic QFT is a valid description of the processes taking place in the quantum eraser, and proves that FTL or retro-causal processes cannot occur within such a setup.
    Reply With Quote  
     

  35. #34  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    The first argument is kind of proving by assuming the thesis.
    Quote Originally Posted by Markus Hanke View Post
    I think that the Eberhard theorem is a pretty strong counterargument; it works off the assumption that relativistic QFT is a valid description of the processes taking place in the quantum eraser, and proves that FTL or retro-causal processes cannot occur within such a setup.
    Unfortunately I(we?) don't have access to this journal - if you put your faith in it, please explain us the idea behind this proof? E.g. that we can be sure its assumptions aren't too strong?
    Personally, I have a problem with putting my faith in general theorems of physics (in opposite to mathematics) - they assume we really do understand the situation ... do we? E.g. do we have the theory of everything?
    So it would be even better if you could just explain what is wrong with the concrete realizations this thread is about? Or CPT analogue of laser?
    Having a theorem it should be simple to use it to e.g. contradict a concrete situation?
    Reply With Quote  
     

  36. #35  
    Moderator Moderator Markus Hanke's Avatar
    Join Date
    Nov 2011
    Location
    Ireland
    Posts
    7,168
    Quote Originally Posted by Jarek Duda View Post
    Unfortunately I(we?) don't have access to this journal - if you put your faith in it, please explain us the idea behind this proof? E.g. that we can be sure its assumptions aren't too strong?
    Personally, I have a problem with putting my faith in general theorems of physics (in opposite to mathematics) - they assume we really do understand the situation ... do we? E.g. do we have the theory of everything?
    So it would be even better if you could just explain what is wrong with the concrete realizations this thread is about? Or CPT analogue of laser?
    Having a theorem it should be simple to use it to e.g. contradict a concrete situation?
    Unfortunately I can't find the public domain version of this - I know there is one somewhere. I will keep looking and will post the link here when I come across it.
    The basic idea behind the theorem is simple ( the maths are not, however ! ) - information in entangled systems is encoded in the correlation between measurements on the constituents, not the constituents themselves. Thus the individual measurements must be correlated, which is then subject to normal relativistic limits so far as information exchange is concerned.
    I can't present the maths here, as much of that is quite beyond me I'm afraid, but the above is the general idea. The only assumption made is that relativistic QED is a valid model of the photon interactions/dynamics within the apparatus.

    I'm surprised that there is no Wikipedia entry for the Eberhard theorem...
    Reply With Quote  
     

  37. #36  
    Moderator Moderator Markus Hanke's Avatar
    Join Date
    Nov 2011
    Location
    Ireland
    Posts
    7,168
    When searching for the Eberhard theorem paper I came across a thread on another forum which considers the same questions :

    Why does quantum entanglement not allow ftl communication

    Interesting read
    Reply With Quote  
     

  38. #37  
    Moderator Moderator Markus Hanke's Avatar
    Join Date
    Nov 2011
    Location
    Ireland
    Posts
    7,168
    Personally, I have a problem with putting my faith in general theorems of physics
    Well, so far as I am concerned I am not aware of any repeatable experiments in this area that have actually demonstrated FTL information exchange, or retro-causality. Correct me if I am wrong. This lack of evidence to the contrary seems to corroberate the theory, at least until proven otherwise.
    Reply With Quote  
     

  39. #38  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    Quote Originally Posted by Markus Hanke View Post
    . Thus the individual measurements must be correlated, which is then subject to normal relativistic limits so far as information exchange is concerned.
    By relativistic limit you mean that anything cannot travel faster than light?
    I completely agree ... that direct propagation is restricted by c.
    But it's past->future thinking our intuition is based on ... while SRT we are talking about is time symmetric - ok, propagation cannot go faster than light, but it "goes" in both time directions ... or more proper understanding (of Lagrangan mechanics) is that physics just finds action optimizing history of field configuration.
    It's like living in four-dimensional jello - physics finds equilibrium: tension(action) minimizing configuration - affecting situation in a point would affect configuration of its past and future.
    Well, so far as I am concerned I am not aware of any repeatable experiments in this area that have actually demonstrated FTL information exchange, or retro-causality. Correct me if I am wrong. This lack of evidence to the contrary seems to corroberate the theory, at least until proven otherwise.
    Probably you are right, but a century ago they haven't demonstrated many todays experiments - that it disprove anything?
    I'm not saying that it is simple or even possible, but only that I'm far from being convinced that it's impossible.
    Reply With Quote  
     

  40. #39  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    Quote Originally Posted by Jarek Duda View Post
    ... and I don't see any real counterarguments from you. I don't see why Lagrangian mechanics would forbid something like that. What I see is confirmation bias.
    I have explained the practical reason why we cannot transmit information backwards through time, or faster than c, in your delayed choice experiment.

    The density of photons collected at the s detector does not change when the polarizer is rotated - in the experiment you used as an example.
    Last edited by SpeedFreek; March 10th, 2012 at 07:18 AM.
    Reply With Quote  
     

  41. #40  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    Quote Originally Posted by SpeedFreek View Post
    The density of photons collected at the s detector does not change when the polarizer is rotated - in the experiment YOU used as an example.
    The density of photons hitting s which entangled partner didn't hit p - and I agree with that.
    But what we see without CC is total density - summed with the number of coinciding photons, which change when polarizer is rotated - so total number/density also changes.
    Let us finish this subject because it doesn't lead anywhere.
    So what about CPT analogue of laser?
    Reply With Quote  
     

  42. #41  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    Quote Originally Posted by Jarek Duda View Post
    Quote Originally Posted by SpeedFreek View Post
    The density of photons collected at the s detector does not change when the polarizer is rotated - in the experiment YOU used as an example.
    The density of photons hitting s which entangled partner didn't hit p - and I agree with that.
    But what we see without CC is total density - summed with the number of coinciding photons, which change when polarizer is rotated - so total number/density also changes.
    Let us finish this subject because it doesn't lead anywhere.
    It won't lead anywhere whilst you continue to labour under the misconception that the total number/density also changes. It doesn't.
    Reply With Quote  
     

  43. #42  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    If there would be at average e.g. 5 incidents/second of (s hitting p not hitting) and 3-5/second of (s and p hitting) depending on polarizer rotation, not using the coincidence counter we would see at average 8-10/second - depending on polarizer rotation.
    And you indeed rather cannot talk me out of this "misconception" ... so maybe better just explain the problem with creating CPT analogue of laser?
    Reply With Quote  
     

  44. #43  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    Quote Originally Posted by Jarek Duda View Post
    If there would be at average e.g. 5 incidents/second of (s hitting p not hitting) and 3-5/second of (s and p hitting) depending on polarizer rotation, not using the coincidence counter we would see at average 8-10/second - depending on polarizer rotation.
    There seems to be a fundamental error here. You seem to think that if p doesn't hit, there are more total incidents of s hitting the s detector than when p does hit. Why?

    Quote Originally Posted by Jarek Duda View Post
    And you indeed rather cannot talk me out of this "misconception" ... so maybe better just explain the problem with creating CPT analogue of laser?
    There seems little point chopping and changing the experiment whilst you continue to think that there is an interference pattern at the s detector whilst the which path information still exists within the whole system. No?
    Reply With Quote  
     

  45. #44  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    Quote Originally Posted by SpeedFreek View Post
    There seems to be a fundamental error here. You seem to think that if p doesn't hit, there are more total incidents of s hitting the s detector than when p does hit. Why?
    No, they are kind of independent - I give up ... once again:
    There are 3 cases the coincidence counter allows us to distinguish between here:
    1) both detectors catch photon - f^2 probability
    2) only s detector - f*(1-f) probability
    3) only p detector - f*(1-f) probability
    where f is detector effectiveness - they catch f of photons.
    The s pattern with coincidence counter is made of 1) cases, without coincidence counter is made of 1) and 2) cases.
    Assume that distinguishing these cases, the s pattern is correspondingly rho1 and rho2
    So without the coincidence counter, the density observed by s would be f^2*rho1 + f(1-f)*rho2

    If efficiency of detectors would grow to 100%, every time s would hit, p would also hit - the coincidence counter would have completely no meaning.
    For smaller efficiencies it only allows us to ignore all but 1) cases - enhancing the effect.
    Reply With Quote  
     

  46. #45  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    Quote Originally Posted by Jarek Duda View Post
    Quote Originally Posted by SpeedFreek View Post
    There seems to be a fundamental error here. You seem to think that if p doesn't hit, there are more total incidents of s hitting the s detector than when p does hit. Why?
    No, they are kind of independent - I give up ... once again:
    There are 3 cases the coincidence counter allows us to distinguish between here:
    1) both detectors catch photon - f^2 probability
    2) only s detector - f*(1-f) probability
    3) only p detector - f*(1-f) probability
    where f is detector effectiveness - they catch f of photons.
    The s pattern with coincidence counter is made of 1) cases, without coincidence counter is made of 1) and 2) cases.
    Assume that distinguishing these cases, the s pattern is correspondingly rho1 and rho2
    So without the coincidence counter, the density observed by s would be f^2*rho1 + f(1-f)*rho2
    This is exactly where you keep going wrong. You keep mixing up the probability of coincidence detection with the probability of a photon simply hitting the s detector which never changes however efficient your detector.

    As long as you keep adding 1) and 2) together, you are counting the same s photon twice, as, in the absence of a coincidence counter, 1) is a subset of 2). These are the probabilities of coincidence counting the s photon, but they are not the probabilities of the s photon actually hitting the s detector, which is all we are concerned with in the absence of coincidence counting.

    Quote Originally Posted by Jarek Duda View Post
    If efficiency of detectors would grow to 100%, every time s would hit, p would also hit - the coincidence counter would have completely no meaning.
    For smaller efficiencies it only allows us to ignore all but 1) cases - enhancing the effect.
    Take a look at the data from an s detector. At any time, and with any delayed choice rotation of the polarizer, it is a bell curve with the same density. Simple fact. I think we might have noticed, otherwise. You seem to be saying that with 100% efficient detectors we would see an interference pattern, but why would we not see it with less than 100% efficient detectors?

    Detector efficiency is totally irrelevant to this.

    If we assume we detect all entangled s photons, they will still form a bell curve of the same density, whatever the rotation of the polarizer.
    Last edited by SpeedFreek; March 10th, 2012 at 09:54 AM.
    Reply With Quote  
     

  47. #46  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    Detector efficiency is totally irrelevant to this.
    Ok, so let us focus on theoretical case of 100% efficient detectors.
    In this case if a photon hit detector s, its entangled partner with 100% probability has hit p detector - not true?
    All observed photons were coinciding ones - what do we need coincidence counter in this case??
    Reply With Quote  
     

  48. #47  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    Quote Originally Posted by Jarek Duda View Post
    Detector efficiency is totally irrelevant to this.
    Ok, so let us focus on theoretical case of 100% efficient detectors.
    In this case if a photon hit detector s, its entangled partner with 100% probability has hit p detector - not true?
    Not true.

    We can completely remove the p detector from the system, leaving all the p photons undetected. This has no effect on whether the s photon hits the s detector. We just have a bell curve of hits at the s detector, with nothing to correlate them with. Do you think the s photons disappear or scatter or something?
    Reply With Quote  
     

  49. #48  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    I've meant both p and s detectors have 100% efficiency - if there is produced entangled pair, both detectors will spot a photon - all cases are coinciding.
    Reply With Quote  
     

  50. #49  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    We end up with the set of fringes and antifringes that correspond to the two erased paths, superimposed over each other, making a bell curve. Or we end up with the bell curves for both unerased paths, superimposed over each other, making an identical bell curve to the one for the erased paths.
    Reply With Quote  
     

  51. #50  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    Yes, depending on performing the erasure - polarizer angle. Coincidence counter completely is useless in this 100% efficiency case.
    Reply With Quote  
     

  52. #51  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    I don't understand why you are saying that. Again you are mixing up the likelihood of the coincidence counter detecting a entangled pair, with the likelihood of an s photon hitting the s detector. I can't see how these two probabilities are supposed to be linked.
    Reply With Quote  
     

  53. #52  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    By the fact that measuring photon p denotes if rotation of the polarize controls the quantum erasure - now it applies to all s photons we get - no coincidence counter is needed.
    Reply With Quote  
     

  54. #53  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    Look, let's just dispense with the p detector completely, shall we? If, as you say, there is no need to know which photons at the p detector correspond with their partners at the s detector, why do we need the p detector at all?

    All we need to know is that, whilst we measure a bell curve at the s detector, there might be a hidden interference pattern in it. How do we find out if there is one or not, without calculating which photons at the p detector correspond with their partners at the s detector? Which photons at the s detector should we be looking for the interference pattern in? They were all polarized by the quarter wave plates on their journey, and their partners don't hit the polarizer until after they have been detected, so the s photons for each path form two bell curves, superimposed over each other, due to the which path information still being contained within the system when they were detected.
    Reply With Quote  
     

  55. #54  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    Quote Originally Posted by SpeedFreek View Post
    Look, let's just dispense with the p detector completely, shall we? If, as you say, there is no need to know which photons at the p detector correspond with their partners at the s detector, why do we need the p detector at all?
    Quantum erasure without erasure???? Without it we get practically standard interference. It is the essence of this experiment - allows to remotely erase the which-path information accordingly to our choice (polarizer rotation). I give up. Here is simple description ... A Double Slit Quantum Eraser Experiment
    Reply With Quote  
     

  56. #55  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    Quote Originally Posted by Jarek Duda View Post
    Quote Originally Posted by SpeedFreek View Post
    Look, let's just dispense with the p detector completely, shall we? If, as you say, there is no need to know which photons at the p detector correspond with their partners at the s detector, why do we need the p detector at all?
    Quantum erasure without erasure???? Without it we get practically standard interference. It is the essence of this experiment - allows to remotely erase the which-path information accordingly to our choice (polarizer rotation). I give up. Here is simple description ... A Double Slit Quantum Eraser Experiment
    Heheh, no, I give up! You really don't sound like you understand the experiment at all.

    The p detector is not what erases the which path information. The erasure is performed at the polarizer. Once the marker has been removed from the p photon at the polarizer, the which path information no longer exists in the system. The data collected at the s detector was collected whilst the which path information still existed in the system.

    So, what is the p detector for, then?

    It is used purely and solely to work out which p photons correspond with which s photons. We need to do this to find if there is an interference pattern hidden in the s detector data, as in a delayed choice experiment all we ever see at the s detector is a bell curve with the same density. And the only way to know if a particlular p photon corresponds to a particular s photon is to...

    ...use a coincidence counter. We need to know which hits to superimpose over each other. We need to count them in at the s detector (whilst the which path information is contained in the system) and then count them in at the p detector (after the which path information has been erased from the system).

    We cannot just set the experiment running to see what the "lag" is between the s and p detectors, without somehow superimposing the data from the p detector with that of the s detector.

    So there can be no faster than light or backwards through time transmission of information, unless we use the p detector to find it, which means we can only ever find it after the fact.
    Last edited by SpeedFreek; March 11th, 2012 at 10:07 AM. Reason: reworded 2nd last paragraph to generalise for all experiments
    Reply With Quote  
     

  57. #56  
    Time Lord
    Join Date
    Mar 2007
    Posts
    8,142
    Quote Originally Posted by SpeedFreek View Post

    Quote Originally Posted by kojax View Post
    It doesn't create any paradoxes so long as all information concerned is contained in the system we are attempting to analyze. If the person changing the past doesn't know what that past was, then they can't choose to contradict it. If they don't choose to contradict it, then there's no reason to expect it would self contradict by accident. Whatever the future actor causes to happen is what did happen.
    The only way the information can remain in the system and not cause paradoxes is if the observer in the past cannot know what choice the future actor makes. Especially if they are going to be that future actor. This is the situation with the delayed choice quantum eraser.
    The alternative would be for the past observer to be unable to alter their future choices. Specificaly.... to be unable to alter their future choices for some reason other than not knowing them (because not knowing them is the default reason we can't "alter" the future - we don't know what we'd be changing it from). That's how it would work if you were able to send information into the past also. It would have to be a part of the past you didn't know yet, so you're not "changing" it, since you never knew what it was in the first place.

    In a Schrodinger's Cat type set up, that's what it would mean to be able to decide whether the cat had been dead or alive the whole time the box was closed (but before opening the box to look.)


    Quote Originally Posted by kojax View Post
    We're taking a quantum effect and expanding it into the macro world. When two or more events are superposed over each other, any decision about the outcome is simultaneous with detection from that effect's perspective. If I had a working version of a device that could send information back 3 hours in time, and I decided to look up a lottery draw that was going to happen then and send the data back to the present, one of two things would happen.

    1) - My future self comes through and I see the numbers now.

    2) - My future self decides to change his mind and I don't see any numbers now.

    The possibility that involves a contradiction (me seeing the numbers and then deciding not to send them) is excluded from the realm of possible outcomes. The whole series of events is one event.
    Very nice!

    So, we can conclude then, that in a delayed choice experiment, if we see no interference pattern at the S detector, nothing can make us press the erase button in the future. Or, if we see an interference pattern, nothing can prevent us from pressing it.

    Does that sound right, to you? I mean, I have heard of people taking the observer affects experiment effect too literally, but here we have experiment affects observer, don't we?
    Yes, and this implies that the experiment is most likely to be successful if it is fully automated, so the entity making the decisions has no free will.

    I know this makes time travel appear to be useless, but you could use it to know tomorrow's lottery numbers, and even buy a ticket, so long as the only part of the future you decide to glimpse is the draw itself (which you don't intend to try and change), and you avoid looking at what your future self is going to do (a decision which you are still free to change because you didn't look at it.)

    When conspiracy theorists talk about the (supposedly many) put options that were in place for 911, to sell/buy stocks that related to the attack, I'm tempted to wonder if maybe some stock traders have already come into possession of such a machine. The beauty of put options is that their presence doesn't alter the course of events in the stock market until those events have already transpired (at least usually). So if you foresaw a stock changing price dramatically, you could set up a put option now, which wouldn't take effect until after the change was complete (at which point the prophecy is fully fulfilled and you're free to resume normal behavior). The point is: if this tech works it's not useless. It's not as awesome as the kind of time Machine HG Wells proposed, or the one in Back to the Future, but it would still be pretty useful.

    Just don't try to use it to change anything. If you see a bus full of children driving off a cliff tomorrow. Too bad. You can't stop it from happening. I suppose you could set up a net at the bottom of the cliff, or have ambulances in the area standing by, but that bus is going to go over the cliff no matter what.
    Some clocks are only right twice a day, but they are still right when they are right.
    Reply With Quote  
     

  58. #57  
    Moderator Moderator Markus Hanke's Avatar
    Join Date
    Nov 2011
    Location
    Ireland
    Posts
    7,168
    Quote Originally Posted by SpeedFreek View Post
    It is used purely and solely to work out which p photons correspond with which s photons. We need to do this to find if there is an interference pattern hidden in the s detector data, as in a delayed choice experiment all we ever see at the s detector is a bell curve with the same density. And the only way to know if a particlular p photon corresponds to a particular s photon is to use a coincidence counter.
    Precisely SpeedFreek. This is in accordance with the aforementioned Eberhard theorem - in order to extract useful information from the system, you need to correlate the data from both parts of the system, i.e. from both detectors in this particular case. This process is by necessity restricted to the speed of light. As for the overall experimental setup, that means that retro-causality is thus not possible.
    Also, if it was indeed possible, surely someone would have noticed by now ??
    Reply With Quote  
     

  59. #58  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    Quote Originally Posted by SpeedFreek View Post
    The p detector is not what erases the which path information. The erasure is performed at the polarizer. Once the marker has been removed from the p photon at the polarizer, the which path information no longer exists in the system. The data collected at the s detector was collected whilst the which path information still existed in the system.
    So, what is the p detector for, then?
    Ok, I agree that what's important is the polarizer - it mounts the trajectory in one of two possibilities - and in lagrangian mechanics it has influence on action optimizing history of field configuration in both time directions - among others minimize "angular stress" of the trajectory.
    If it's true, case 2) (s hit, p not) has the same dependency on polarizer rotation as case 1) (both hit) - the dependency without coincidence counter is even stronger.
    It is used purely and solely to work out which p photons correspond with which s photons. We need to do this to find if there is an interference pattern hidden in the s detector data, as in a delayed choice experiment all we ever see at the s detector is a bell curve with the same density. And the only way to know if a particlular p photon corresponds to a particular s photon is to...
    But what for???? Isn't it just enough that they will hit sometimes?
    What we directly observe is the statistical pattern on s detectors - knowing only it, we know the rotation of polarizer - not true?

    Ok - let's try from another side - so just looking at s photon (without coincidence counter), what statistical pattern we would observe?
    Your answer(?):
    All we need to know is that, whilst we measure a bell curve at the s detector, there might be a hidden interference pattern in it. How do we find out if there is one or not, without calculating which photons at the p detector correspond with their partners at the s detector? Which photons at the s detector should we be looking for the interference pattern in? They were all polarized by the quarter wave plates on their journey, and their partners don't hit the polarizer until after they have been detected, so the s photons for each path form two bell curves, superimposed over each other, due to the which path information still being contained within the system when they were detected.
    If I properly understood you say it depends on if p has already hit the detector?
    But e.g. what does it mean if interval between them is space-like - we could change their order by changing reference frame?
    Reply With Quote  
     

  60. #59  
    Time Lord
    Join Date
    Mar 2007
    Posts
    8,142
    Quote Originally Posted by Markus Hanke View Post
    Quote Originally Posted by SpeedFreek View Post
    It is used purely and solely to work out which p photons correspond with which s photons. We need to do this to find if there is an interference pattern hidden in the s detector data, as in a delayed choice experiment all we ever see at the s detector is a bell curve with the same density. And the only way to know if a particlular p photon corresponds to a particular s photon is to use a coincidence counter.


    Precisely SpeedFreek. This is in accordance with the aforementioned Eberhard theorem - in order to extract useful information from the system, you need to correlate the data from both parts of the system, i.e. from both detectors in this particular case. This process is by necessity restricted to the speed of light. As for the overall experimental setup, that means that retro-causality is thus not possible.
    Also, if it was indeed possible, surely someone would have noticed by now ??

    The trouble with your reasoning is that we can guess, statistically, about how many photons any given detector ought to be picking up based on what the entangled partners are doing. With billions of photons emitted per nano-second, the law of large numbers kicks in, and those guesses become near- certainties. We don't actually need to compare the data at the end of the experiment, because it doesn't actually matter exactly which photons were paired or unpaired. Only the total quantities matter.


    If instead of a half silvered mirror, we used some kind of fiber optic transistor that we could control to make the decision about which photons' path information will be erased and which ones won't, then we could observe that during those moments when we were diverting the photons down a path where it would be erased, a certain number of photons were hitting a detector at a certain location within the interference pattern, and if that number is the number we expect, then that confirms the entangled partners were doing what we expect. When the transistor is set to divert them the other way, the frequency of hits would change according to that setting also.

    That would save us having to determine what random choice the half silvered mirror had made (though even with it, we can measure how close our statistic is to a 50% split).
    Some clocks are only right twice a day, but they are still right when they are right.
    Reply With Quote  
     

  61. #60  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    Quote Originally Posted by Jarek Duda View Post
    Quote Originally Posted by SpeedFreek View Post
    The p detector is not what erases the which path information. The erasure is performed at the polarizer. Once the marker has been removed from the p photon at the polarizer, the which path information no longer exists in the system. The data collected at the s detector was collected whilst the which path information still existed in the system.
    So, what is the p detector for, then?
    Ok, I agree that what's important is the polarizer - it mounts the trajectory in one of two possibilities - and in lagrangian mechanics it has influence on action optimizing history of field configuration in both time directions - among others minimize "angular stress" of the trajectory.
    I'm really not sure what all that is supposed to mean. The polarizer simply reverses the effects of the quarter wave plate on the entangled partner to the signal photon. It erases the "marker" we put on the signal photons, from their entangled partners.

    Quote Originally Posted by Jarek Duda View Post
    If it's true, case 2) (s hit, p not) has the same dependency on polarizer rotation as case 1) (both hit)
    Agreed, I think! Due to language problems, I'm no longer sure I am answering the same questions you are asking.

    Quote Originally Posted by Jarek Duda View Post
    the dependency without coincidence counter is even stronger.
    I still don't understand why you say this.

    Quote Originally Posted by Jarek Duda View Post
    It is used purely and solely to work out which p photons correspond with which s photons. We need to do this to find if there is an interference pattern hidden in the s detector data, as in a delayed choice experiment all we ever see at the s detector is a bell curve with the same density. And the only way to know if a particlular p photon corresponds to a particular s photon is to...
    But what for???? Isn't it just enough that they will hit sometimes?
    What we directly observe is the statistical pattern on s detectors - knowing only it, we know the rotation of polarizer - not true?
    Language issues: Are you talking about us directly observing the statistical pattern at the p detector or the s detector? In either case it is NOT TRUE. The pattern only emerges when we SUPERIMPOSE the data from both detectors.

    What we directly observe at the s detector is a bell curve, of the same density, whatever the rotation of the polarizer!

    What we directly observe at the p detector is the timing and polarizations of the p photons as they arrive - there is no bell curve or interference pattern as such there, because the p photons did not pass through any slits or splitter. We are not measuring the position of the p photon at the detector.

    So, we have timing and polarization information at one detector, and timing and positional information at the other. We need both, to find the interference pattern.

    The signal photons hit the s detector BEFORE their entangled partners hit the polarizer, so the which path information is still contained within the system when the s photons are detected, so all we see at the s detector is a bell curve.

    Quote Originally Posted by Jarek Duda View Post
    Ok - let's try from another side - so just looking at s photon (without coincidence counter), what statistical pattern we would observe?
    Once again - a bell curve!

    Quote Originally Posted by Jarek Duda View Post
    Quote Originally Posted by SpeedFreek View Post
    All we need to know is that, whilst we measure a bell curve at the s detector, there might be a hidden interference pattern in it. How do we find out if there is one or not, without calculating which photons at the p detector correspond with their partners at the s detector? Which photons at the s detector should we be looking for the interference pattern in? They were all polarized by the quarter wave plates on their journey, and their partners don't hit the polarizer until after they have been detected, so the s photons for each path form two bell curves, superimposed over each other, due to the which path information still being contained within the system when they were detected.
    If I properly understood you say it depends on if p has already hit the detector?
    Language problem here: What are you saying I say depends on if p has already hit the detector?

    The results at the s detector do not change when the p photons hit the p detector. What changes is our ability to now compare the s photons with their partner photons at the p detector to see if they form an interference pattern or not. We cannot do this if the p photons have not been detected.

    The fact is that we cannot calculate, through any method including statistical analysis of the s detector data alone, whether the partners of erased p photons hit the s detector. We need the data from the p detector to see where their partners hit the s detector, so we might find the interference pattern.

    From this thread so far it seems like you don't really appreciate the concepts behind the experiment. The delayed choice quantum eraser - the erasure is of which path information through a polarizer, and the delayed choice is positioning the polarizer so that the p photons do not pass through it until after their s partners have been detected.

    Perhaps we should go back to first principles?
    Last edited by SpeedFreek; March 12th, 2012 at 04:35 PM.
    Reply With Quote  
     

  62. #61  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    Quote Originally Posted by SpeedFreek View Post
    Quote Originally Posted by Jarek Duda View Post
    Ok, I agree that what's important is the polarizer - it mounts the trajectory in one of two possibilities - and in lagrangian mechanics it has influence on action optimizing history of field configuration in both time directions - among others minimize "angular stress" of the trajectory.
    I'm really not sure what all that is supposed to mean. The polarizer simply reverses the effects of the quarter wave plate on the entangled partner to the signal photon. It erases the "marker" we put on the signal photons, from their entangled partners.
    It can be viewed from different points of view - you are talking about quantum mechanical description and I complement it from the point of view of lagrangian mechanics: polarizer fixes polarization in one of two possibilities in given point of spacetime, what influence photon's trajectory.
    Polarizer is a constrain for physics while it searches for action optimizing history of field configuration - such constrain influence the solution toward both future and past.
    The signal photons hit the s detector BEFORE their entangled partners hit the polarizer, so the which path information is still contained within the system when the s photons are detected, so all we see at the s detector is a bell curve.
    So what you say is that while manipulating optical length of p photon (at erasure angle) and observing statistical pattern of s photons (without coincidence counter):
    if (length p < length s) then we observe interference pattern
    if (length p > length s) then we observe classical bell curves
    ?

    The first problem is as I've already written, these inequalities are not well defined:
    Let us take (length p = length s) case - in spacetime both measurements are in the same time, but different positions - interval between them is spatial.
    So accordingly to SRT, we can change frame of reference by making boost - accordingly to one observer there is interference, while to another there is classical behavior ...?

    The second problem is that: let both detectors just write times of photon hits.
    So a week later you could take listings and find which photons were coinciding ... and so they should be interfering (not depending on the distance) - not true?
    Reply With Quote  
     

  63. #62  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    Quote Originally Posted by Jarek Duda View Post
    It can be viewed from different points of view - you are talking about quantum mechanical description and I complement it from the point of view of lagrangian mechanics: polarizer fixes polarization in one of two possibilities in given point of spacetime, what influence photon's trajectory.
    Polarizer is a constrain for physics while it searches for action optimizing history of field configuration - such constrain influence the solution toward both future and past.
    I think, before complicating the situation by interpreting it through lagrangian mechanics, you would be wise to gain a better understanding of the experiment from a quantum mechanical point of view.

    Quote Originally Posted by Jarek Duda View Post
    The signal photons hit the s detector BEFORE their entangled partners hit the polarizer, so the which path information is still contained within the system when the s photons are detected, so all we see at the s detector is a bell curve.
    So what you say is that while manipulating optical length of p photon (at erasure angle) and observing statistical pattern of s photons (without coincidence counter):
    if (length p < length s) then we observe interference pattern
    if (length p > length s) then we observe classical bell curves
    In an idealised case, with no unentangled photons in the system to cause noise, and 100% efficent BBO, polarizer and detectors, yes (theoretically). But it has proved impossible in practice, so a coincidence counter is always used.

    Of course, if the p photons are erased at the polarizer before the s photons are detected, then we are not dealing with a delayed choice experiment any more, it is a straightforward (!) quantum eraser.

    Quote Originally Posted by Jarek Duda View Post
    The first problem is as I've already written, these inequalities are not well defined:
    Let us take (length p = length s) case - in spacetime both measurements are in the same time, but different positions - interval between them is spatial.
    So accordingly to SRT, we can change frame of reference by making boost - accordingly to one observer there is interference, while to another there is classical behavior ...?
    Oh my word, now you are bringing Special Relativity into it?! Please stop this, it will only muddy the waters even further!

    Firstly, as anyone who knows STR will understand, there is no such thing as absolute simultaneity, when events are separated by space.

    So, to which of your observers are the events of erasure at the polarizer and detection at the s detector simultaneous?

    To the other observer, it might be, or might not be a delayed choice, depending on their velocity relative to the experimental apparatus. Do you want to go on to define the frames involved? (please don't, it adds nothing to the discussion)

    In other words, there is no problem with the experiment and STR that the relativity of simultaneity cannot handle. Events that are simultaneous in one frame can happen at different times in another.

    Quote Originally Posted by Jarek Duda View Post
    The second problem is that: let both detectors just write times of photon hits.
    So a week later you could take listings and find which photons were coinciding ... and so they should be interfering (not depending on the distance) - not true?
    Well, if the s detector only records the timing of the hit, it doesn't know the position, does it? We need positional information from the s detector. You have to be careful with your language use, it makes these discussions even more confusing than they are already.

    And you do know that all delayed choice experiments are analysed after the experiment, don't you? It may well be a week later.
    Last edited by SpeedFreek; March 12th, 2012 at 05:43 PM.
    Reply With Quote  
     

  64. #63  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    Quote Originally Posted by SpeedFreek View Post
    I think, before complicating the situation by interpreting it through lagrangian mechanics, you would be wise to gain a better understanding of the experiment from a quantum mechanical point of view.
    It's not so simple - quantum mechanics is practical description, but this theory predicts only probabilities of experiments - is not deterministic, is not complete.
    In comparison, more fundamental theories like QFT or GRT are Lagrangian mechanics - are deterministic (by some Euler-Lagrange equation applied to unknown field configuration).
    To understand situation well we just have to look from both perspectives.
    To the other observer, it might be, or might not be a delayed choice, depending on their velocity relative to the experimental apparatus. Do you want to go on to define the frames involved? (please don't, it adds nothing to the discussion)
    Objectively there is observed a concrete statistical pattern - in opposite to your view, it doesn't depend on the choice of frame of reference of an observer.
    I would say that for any optical lengths, the information should be erased in this case.
    Well, if the s detector only records the timing of the hit, it doesn't know the position, does it?
    What is important is that we know the positions - taking recordings of hits of both detectors, we can easily say which one were coinciding - and so which one were erased.
    Only mounting hit recorders on detectors make that some photons should interfere (all if both detectors have 100% efficiency) - not depending on optical lengths.
    Reply With Quote  
     

  65. #64  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    Quote Originally Posted by Jarek Duda View Post
    To the other observer, it might be, or might not be a delayed choice, depending on their velocity relative to the experimental apparatus. Do you want to go on to define the frames involved? (please don't, it adds nothing to the discussion)
    Objectively there is observed a concrete statistical pattern - in opposite to your view, it doesn't depend on the choice of frame of reference of an observer.
    I would say that for any optical lengths, the information should be erased in this case.
    Well, I did say you would muddy the waters by bringing SR into it, didn't I? I completely misunderstood what you meant before, and made a mistake. (Either that, or you led me into it! I will say again that your usage of English is good, but sometimes hard to understand, especially with something as complicated as this subject)

    You are correct in relation to Special Relativity, of course, so what I said in the last post has to be wrong. Not that it changes the message I have been trying to get across all along.

    We cannot, even in a theoretically ideal situation, see an interference pattern at the s detector, before the p photons are detected and compared to work out which are entangled with s photons. I don't know what I was thinking, in the last post.

    Quote Originally Posted by Jarek Duda View Post
    Well, if the s detector only records the timing of the hit, it doesn't know the position, does it?
    What is important is that we know the positions - taking recordings of hits of both detectors, we can easily say which one were coinciding - and so which one were erased.
    Only mounting hit recorders on detectors make that some photons should interfere (all if both detectors have 100% efficiency) - not depending on optical lengths.
    Well, the results are the same, whichever way round you detect the photons at each detector.

    A bell curve showing no interference, even if both detectors have "100% efficiency". I have just realised that there is no such thing as 100% efficiency in this context. I'm stupid, I should have thought of that before. You and your lagrangians! This is quantum mechanics!

    I originally said the coincidence counter was required, in order to differentiate the entangled hits from the "noise". This is indeed correct, and always will be.

    I need to think hard about how to explain this whole thing better, before I continue. Either that or I should just stick to cosmology!
    Last edited by SpeedFreek; March 12th, 2012 at 08:34 PM.
    Reply With Quote  
     

  66. #65  
    Time Lord
    Join Date
    Mar 2007
    Posts
    8,142
    Quote Originally Posted by SpeedFreek View Post
    [

    A bell curve showing no interference, even if both detectors have "100% efficiency". I have just realised that there is no such thing as 100% efficiency in this context. I'm stupid, I should have thought of that before. You and your lagrangians! This is quantum mechanics!
    Even when path information is detected, you still get a single slit pattern, don't you? So really if we didn't do coincidence counting we'd see a 1 slit and 2 slit pattern overlapping each other.
    Some clocks are only right twice a day, but they are still right when they are right.
    Reply With Quote  
     

  67. #66  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    SpeedFreek,
    So finally you say that without coincidence counter we will always observe classical bell curves because the which-path information is (still?) in the system?
    But is there an access to this information? No, because the measurement will be made after it came through erasing polarizer.
    So not only accordingly to my damn lagrangian mechanics ... also quantum mechanics requires not only that the information is there, but is somehow measured ... what is impossible in this case because of the polarizer.

    You say that situation will be radically changed while using coincidence counter (... observed in experiments ...) - what exactly does this tool to find correspondence between ticks in detectors changes?
    Why it cannot be used post-factum by recording times of ticks? So maybe we don't even have to look at it, if the information is already stored somewhere ...

    I think you are too used to natural for our intuition past->future thinking. Fundamental physics, Lagrangian mechanics is very different - is time/CPT symmetric.
    So time asymmetry is not written in the equations ruling our physics, but it is a result of its solution we live in. Like Higg's potential is symmetric, while in reality this symmetry is naturally broken - on level of solutions not equations.
    Or think of statistical physics - canonical ensemble among possibilities - where is time asymmetry here?
    To understand that thermodynamics deeply is also time-symmetric, try to think of simple thought experiment: Thermodynamical thought experiment with time reversing loop?
    Don't just fight with it, but try to understand it instead
    Reply With Quote  
     

  68. #67  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    Okay, let's just back up a bit. I may have had some of my own misconceptions about certain aspects of this experiment, but the fact remains that we do always detect a bell curve at the s detector in a delayed choice experiment.

    I have been doing some research, and the reason for this is that the non-entangled particles in the system far far outweigh the entangled particles, so only an incredibly small proportion of the particles hitting the s detector will have had their which path information erased. The ratio of entangled to non-entangled particles is so incredibly small, that it has proven impossible to use statistical analysis.

    As far as I can tell, there is only ever a very low probability of spontaneous down conversion for a given photon sent through the BBO, which is why, out of the masses of hits on the detectors, we only ever have a few instances of entangled particles hitting the s detector. The polarizer itself also introduces additional uncertainty to the system.

    In other words, it seems to be theoretically impossible to send only entangled particles through the system - they will always be overwhelmed by non-entangled photons. It also seems to be theoretically impossible to erase all entangled photons that reach the polarizer. Therefore there is little point discussing "ideal" cases, 100% efficient detectors or what have you.

    So, when the s photons hit the s detector, all of the non-entangled photons still have their which path information, whereas only a very few have had that information erased.

    Okay so far?
    Last edited by SpeedFreek; March 13th, 2012 at 03:22 PM.
    Reply With Quote  
     

  69. #68  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    If it's your last line of defense, unfortunately I will have to sadden you - such pairs from SPDC are usually created in some angle from the original beam and additionally have twice smaller energy, so they could be also easily separated using just a prism - it is not a problem to produce only entangled photons (of much smaller intensity than the original source). Another view: if you would shoot through your delicate system with this powerful laser, you would probably just burn it ...
    Spontaneous parametric down-conversion - Wikipedia, the free encyclopedia
    ... but maybe you shouldn't see time/CPT symmetric lagrangian mechanics as your enemy, but rather as a powerful ally instead
    Reply With Quote  
     

  70. #69  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    Perhaps you might tell us why, in the conclusions of the papers where they perform these experiments, they always rule out retro-causal information transfer?
    Reply With Quote  
     

  71. #70  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    If you have only para-arguments left ... I don't know, but I would suspect confirmation bias - there is a strong belief in intuitive "fundamental asymmetry"against CPT conservation because e.g. many thinks it would contradict free will (I disagree - our minds are separate constructs) ... or are afraid of getting a label of being a freak scientist what could cost them the career in our society ... it is allowed to say that antiparticle is travelling back in time pariticle in Feynaman diagrams (Feynman-Stueckelberg interpretation), but taking it macroscopic physics is "controversial"...
    If you have run out of real arguments on this experiment, maybe you could finally explain what's wrong with constructing CPT analogue of laser I've mentioned a few times?
    Reply With Quote  
     

  72. #71  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    I still get the nagging feeling we are missing something obvious here. I must admit, whilst I understand the results of the experiment itself, I was only viewing it from one perspective. I must also admit that my memory is not what it used to be - when I look at posts I was making a few years back I am surprised by how much I have forgotten! Additionally, I must admit I am on far surer footing with cosmology than I am with quantum physics, so I should probably keep out of these kind of discussions in future.

    But it is the way it has been explained to me many times in the past - that there is no interference at the s detector due to the which path information not having been erased when they are detected. They have been polarized in different directions by the quarter wave plates, and their entangled partners have not reached the polarizer yet, so we know which path they took, and therefore there is no interference pattern. Are you saying this is not true?

    And once their entangled partners have been erased, we find that the s photons were indeed in an interference pattern after all, but we could not know that until we have detected their partners. The s photons haven't moved position on the detector, of course.

    I was always told that this was not due to any inefficiencies in the apparatus or method, and that there was no way around this problem. I have both read, and been involved in discussions with working physicists on different forums where this has come up in the past.

    I am not convinced what you are saying is correct, in other words. Although I still wonder if I am misunderstanding you. What I do know is that I am not really as qualified as I thought I was to discuss this, and I should stick to the subjects I know a little better in future. My first words in this thread were an apt description!

    I haven't got the foggiest idea about anything to do with CPT either.

    So now I have managed to completely confuse myself (with your help!), perhaps someone could help me and tell me what the hell is going on in this experiment?!

    Are you really saying the interference pattern is there, even though the s photons have been polarized at the quarter wave plates so we know which path they took? I mean, do we have the which path information or not? I know you will say this is why I should consider the problem from the lagrangian perspective, but here we are not talking so much about the results of the delayed choice quantum eraser as the whole "knowing which path means there is no interference pattern" paradigm. Or that if the which path informations exists in the universe, there is no interference pattern.

    If that is the case you seem to be trying to overturn more than just the backwards through time information transfer problem of the delayed choice experiment.
    Reply With Quote  
     

  73. #72  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    Quote Originally Posted by SpeedFreek View Post
    But it is the way it has been explained to me many times in the past - that there is no interference at the s detector due to the which path information not having been erased when they are detected. They have been polarized in different directions by the quarter wave plates, and their entangled partners have not reached the polarizer yet, so we know which path they took, and therefore there is no interference pattern. Are you saying this is not true?
    It's more subtle - the which-path information is in the system, but is deeply hidden - we don't know the path. The only way (approved by QM) to get this information out is a measurement ... but in this system any kind of measurement can be made only after photon passes through polarizer erasing the information - this measurement decides if the path of photons is well defined (classical) or not (quantum).

    If you prefer to discuss time/CPT symmetry from cosmological point of view, looking at the Big Bang moment it should be time/CPT symmetric - suggesting it was Big Bounce in fact.
    So how would entropy behavior look like before? In BB everything was well localized - it denotes it had extremely low entropy.
    So before Big Bounce it was loosing this localization - increasing entropy ... so didn't BB entropy minimum created two entropy gradients - our 2nd law toward our future and opposite 2nd law before?
    What about 2nd law of thermodynamics in Cyclic Universe Model?
    Reply With Quote  
     

  74. #73  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    Quote Originally Posted by Jarek Duda View Post
    Quote Originally Posted by SpeedFreek View Post
    But it is the way it has been explained to me many times in the past - that there is no interference at the s detector due to the which path information not having been erased when they are detected. They have been polarized in different directions by the quarter wave plates, and their entangled partners have not reached the polarizer yet, so we know which path they took, and therefore there is no interference pattern. Are you saying this is not true?
    It's more subtle - the which-path information is in the system, but is deeply hidden - we don't know the path. The only way (approved by QM) to get this information out is a measurement ... but in this system any kind of measurement can be made only after photon passes through polarizer erasing the information - this measurement decides if the path of photons is well defined (classical) or not (quantum).
    We measure the s photons when they hit the s detector. If the which path information is in the system, they form a bell curve. These are the results of the classic single/dual slit experiment. So, what about the correspondence principle? Does it not apply here?

    Quote Originally Posted by Jarek Duda View Post
    If you prefer to discuss time/CPT symmetry from cosmological point of view, looking at the Big Bang moment it should be time/CPT symmetric - suggesting it was Big Bounce in fact.
    So how would entropy behavior look like before? In BB everything was well localized - it denotes it had extremely low entropy.
    So before Big Bounce it was loosing this localization - increasing entropy ... so didn't BB entropy minimum created two entropy gradients - our 2nd law toward our future and opposite 2nd law before?
    What about 2nd law of thermodynamics in Cyclic Universe Model?
    So far, the only theory I know of that predicts a big-bounce is loop quantum gravity, and there are quite a few problems with it. Anything dealing with "before" the Big-Bang is speculative, at best. So no, I do not prefer to discuss it in this context.
    Reply With Quote  
     

  75. #74  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    Quote Originally Posted by SpeedFreek View Post
    We measure the s photons when they hit the s detector. If the which path information is in the system, they form a bell curve
    Being in the system is not enough. It has to be not in superposition, but well defined - measurement is required.
    Look at the realizations - there is a reason this experiment is called delayed choice quantum erasure - it is not enough that the information is in the system, but it has to be measured to make that the way of s is well defined (classical behavior).
    So far, the only theory I know of that predicts a big-bounce is loop quantum gravity, and there are quite a few problems with it. Anything dealing with "before" the Big-Bang is speculative, at best. So no, I do not prefer to discuss it in this context.
    Maybe loop quantum gravity sounds sexy and they can produce tons of papers about it ... but maybe let's look first at the situation from the perspective of less marvelous theories like time/CPT conserving Lagrangian mechanics ;p (ok, LQG is also one of them)
    If there is no reason to tell that in some points time/CPT conservation is violated, it shouldn't be also violated in the BB moment - making it Bounce ... not true?
    Reply With Quote  
     

  76. #75  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    Seeing as our best theory falls apart and predicts a singularity at the BB, we cannot even talk about the BB "moment", can we.

    Also, if the universe were created at the Big Bang, who is to say that the physical laws that apply within the universe have to apply to the process that created it?
    Reply With Quote  
     

  77. #76  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    Indeed trying to create asymmetric point of manifold with nothing in one direction and everything in opposite requires a lot of mathematical acrobatics
    From the other side, natural for lagrangian mechanics trivial and boring: gravitational collapse ... BAAM ... expansion (because of momentum conservation) is so simple that they couldn't produce tons of papers about it required from modern scientist (instead of naive idealistic "finding the truth" of obsolete ones).
    Reply With Quote  
     

  78. #77  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    You mentioned conservation.. well the Big-Bounce idea would be more convincing if we had any evidence that the expansion of the universe were decelerating towards a Big Crunch, but instead we find it to have ceased in its deceleration and to now be accelerating, so it doesn't seem like there is going to be another Big-Bounce, does it?

    So where did all that extra energy come from?
    Reply With Quote  
     

  79. #78  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    First of all, I was talking about our Big Bang - that it's being Bounce is much less mathematically problematic - for example that symmetrically the Universe were collapsing from infinity and now it will grow to infinity ... and there is a question of 2nd law of thermodynamics in such scenario?
    But Cyclic Universe scenario is indeed much ... nicer(? ) for our intuition ... and of course I have also argument for that: energy conservation - gravity is attracting (1/r^2), while what is repelling is called dark energy. Standard models assume its density (cosmological constant) is constant, while energy conservation suggests that what is conserved is the total amount - so repelling should decrease with the volume (1/r^3) and so gravity should finally win ...
    Reply With Quote  
     

  80. #79  
    Time Lord
    Join Date
    Mar 2007
    Posts
    8,142
    Quote Originally Posted by SpeedFreek View Post
    Perhaps you might tell us why, in the conclusions of the papers where they perform these experiments, they always rule out retro-causal information transfer?
    I'm pretty sure they do that because they have to if they want to get published. You can't question longstanding beliefs unless you have overwhelming evidence. If an experiment were to challenge GR, for example, you would not want to say so anywhere in the paper. You'd bend over backwards saying again and again how fundamentally true GR is and then.... hope somebody else happens to put all the pieces together from your data, and see the conflict and point it out. That way it's their career on the line too, and not just yours. There is still a lot of politics in science, whether people are comfortable admitting it or not, and the sheer number of people who see what you're seeing has quite an effect on the likelihood of being allowed to spread your idea instead of seeing it summarily suppressed.

    It's better to get one's data out into public view first, whatever disclaimers you have to make to get it there.


    Unlike the example of GR (which will probably survive the test of time), forward causality is much less proven. It conforms the best with common sense, but if common sense is what we wanted, we'd discard relativity also and go back to Newtonian mechanics (after all, it's much simpler, and conforms very well with our everyday experience better than Relativity does.) Just imagine!! Forward and backward progress through time being a matter of perspective!!?!?!? Oh.... wait..... we've seen other things become matters of an observer's perspective......
    Some clocks are only right twice a day, but they are still right when they are right.
    Reply With Quote  
     

  81. #80  
    Time Lord
    Join Date
    Mar 2007
    Posts
    8,142
    If you want to wrap your head around backwards time causality, just remember that photons are quantized. That means either the whole photon appears in one place at one time, or it appears somewhere else at some time, but it doesn't spread out and appear in multiple places at once. Any machine you build on a QM principle will, by very nature, be taking the behavior of a photon and amplifying it big enough to see on a macro scale, but...... if we're seeing the effect and the machine is working right, then that machine's behavior and the behavior of the whole system that interacts with it will be like the photon it is attempting to emulate.

    If the time stream has the option to choose between you going back in time and killing your grandfather and then consequently never being born, vs. you not being able to get the damn machine to work, then...... just like a photon.... it will choose one and exactly one (never two, or three) of those options, and it will be the only option that was ever chosen. Trying to get two contradictory events to overlap is like trying to get two photons to strike a detector at a location where they would cancel one another. You just have to take the concept of a photon out to the absurd limit, and then it would be clear to you why backwards time causality won't cause contradictions. Trying to identify what mechanism causes the no-contradiction to occur is like trying to identify what mechanism causes a single photon passing through a double slit to interfere with itself. Insofar as we know..... it's quite possible there is no mechanism. It just is.
    Some clocks are only right twice a day, but they are still right when they are right.
    Reply With Quote  
     

  82. #81  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    Quote Originally Posted by kojax View Post
    Trying to identify what mechanism causes the no-contradiction to occur is like trying to identify what mechanism causes a single photon passing through a double slit to interfere with itself. Insofar as we know..... it's quite possible there is no mechanism. It just is.
    Actually, it is not so difficult - let us not complicate things with grandparents, especially that getting larger time delays seems to be extremely difficult to obtain in practice.
    Let's focus on simple hypothetical application (e.g. finding the cryptokey in a moment) using microscopic time differences:


    So we use a few dozens of such quantum erasure systems - the upper boxes mean choosing the angle of p polarizer and the lower: checking if pattern of s photons is classical or interfering.
    If there is no satisfying input (like cryptokey decrypting the message), optimizing action lagrangian mechanics would have to break this "causality time-loop constrain" created by us - making that some of looking random for us degrees of freedom (thermodynamical, statistical) would loose some of its freedom/randomness.

    So finally we can use the obtained "input"/"output"- if they are different means that physics has lied in the imperfect(!) back-time channel - e.g. bending the statistics a bit.
    If they are equal, physics could lie somewhere in the electronics checking the satisfiability (governed by thermodynamics).
    It suggests that such hypothetical backtime channels would have an important finite quantitative parameter: the strength. It can be increased by combining many of such channels in parallel way and should decrease while increasing time difference.
    Reply With Quote  
     

  83. #82  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    Quote Originally Posted by Jarek Duda View Post
    First of all, I was talking about our Big Bang - that it's being Bounce is much less mathematically problematic - for example that symmetrically the Universe were collapsing from infinity and now it will grow to infinity ... and there is a question of 2nd law of thermodynamics in such scenario?

    This just pushes all the difficult questions further back, rather than answering them. If energy is conserved, then the previous universe should not have collapsed. Hence my question about the source of the extra 70% represented by dark energy.

    Quote Originally Posted by Jarek Duda View Post
    But Cyclic Universe scenario is indeed much ... nicer(? ) for our intuition ... and of course I have also argument for that: energy conservation - gravity is attracting (1/r^2), while what is repelling is called dark energy. Standard models assume its density (cosmological constant) is constant, while energy conservation suggests that what is conserved is the total amount - so repelling should decrease with the volume (1/r^3) and so gravity should finally win ...
    A cosmological constant is constant for a given volume of space - if the space increases in volume, the repulsive effect is magnified, not reduced, as there is now extra space and the cosmological constant remains the same in any given volume of space. So the bigger the gap between the galactic clusters, the larger the repulsive effect.

    This is why, having discovered that the rate of expansion is accelerating, that we think the expansion of our universe will never come to a halt. We think that gravity lost the fight 6 billion years ago.
    Reply With Quote  
     

  84. #83  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    Quote Originally Posted by SpeedFreek View Post
    This just pushes all the difficult questions further back, rather than answering them. If energy is conserved, then the previous universe should not have collapsed. Hence my question about the source of the extra 70% represented by dark energy.
    Why do you think energy conservation prevents collapse?
    Imagine a dust cloud - without angular momentum and interaction/viscosity it could just collapse to a point, then expand back to the original radius and so on.
    Such hypothetical evolution before BB would be practically reversed ours - it's because lagrangian mechanics is time/CPT symmetric and the BB moment should be a total chaos - it's difficult to find a reason for an essential asymmetry there, so "both Universes" should evolve similarly in opposite directions - both interpreting BB moment as their Big Bang ... and so entropy should also grow symmetrically - with entropy gradient according to this total entropy minimum: out of the the Big Bounce moment.
    A cosmological constant is constant for a given volume of space - if the space increases in volume, the repulsive effect is magnified, not reduced, as there is now extra space and the cosmological constant remains the same in any given volume of space. So the bigger the gap between the galactic clusters, the larger the repulsive effect.
    You are talking about assumption which disagree with the most important principle of physics: energy conservation.
    If density of dark energy (cosmological constant) is indeed constant and the volume of Universe increases ... so where this huge amount of new energy comes from? alternative universes? ;p
    Reply With Quote  
     

  85. #84  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    Quote Originally Posted by Jarek Duda View Post
    Quote Originally Posted by SpeedFreek View Post
    This just pushes all the difficult questions further back, rather than answering them. If energy is conserved, then the previous universe should not have collapsed. Hence my question about the source of the extra 70% represented by dark energy.
    Why do you think energy conservation prevents collapse?
    Imagine a dust cloud - without angular momentum and interaction/viscosity it could just collapse to a point, then expand back to the original radius and so on.
    Such hypothetical evolution before BB would be practically reversed ours - it's because lagrangian mechanics is time/CPT symmetric and the BB moment should be a total chaos - it's difficult to find a reason for an essential asymmetry there, so "both Universes" should evolve similarly in opposite directions - both interpreting BB moment as their Big Bang ... and so entropy should also grow symmetrically - with entropy gradient according to this total entropy minimum: out of the the Big Bounce moment.
    We need omega > 1 for a universe to collapse, which, if we include your symmetry should produce a universe with omega >1.

    I think you are stretching the domain of applicability of lagrangian mechanics far beyond their limits, trying to use them to cross the BB singularity.

    Quote Originally Posted by Jarek Duda View Post
    A cosmological constant is constant for a given volume of space - if the space increases in volume, the repulsive effect is magnified, not reduced, as there is now extra space and the cosmological constant remains the same in any given volume of space. So the bigger the gap between the galactic clusters, the larger the repulsive effect.
    You are talking about assumption which disagree with the most important principle of physics: energy conservation.
    If density of dark energy (cosmological constant) is indeed constant and the volume of Universe increases ... so where this huge amount of new energy comes from? alternative universes? ;p
    I am describing the current cosmology. There is no new energy, but rather there is a ratio between the energy density of the cosmological constant, which never changes, and the critical density of the universe, which does change. This is what caused gravity to lose the fight to the cosmological constant around 6 billion years ago. Before that, gravity was slowing the rate of expansion, but since then the cosmological constant has been increasing it. The presence of the cosmological constant precludes our universe from collapsing, which is why we predict heat death as the ultimate fate of the universe.
    Reply With Quote  
     

  86. #85  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    SpeeDFreek, I really know the standard assumptions, but I cannot say that they satisfy me - like I put more faith in energy conservation (total energy conserved) than in guessed assumption of constance of (dark) energy density - because of eastetic reasons.
    About lagrangian mechanics used from QFT to GRT, it provides simple and universal principles like time/CPT conservation and so in opposite to fantastic alternative theories like inflation, string theory etc., it doesn't leave much freedom for confabulations.
    "In every point physics is time/CPT symmetric and so in BB moment" - it's just that simple.
    Like cloud of dust without angular momentum and interactions - collapsing then expanding - you could run it backward and it would generally look the same.

    The main problem with modern physics is that simple answers would make physicists unemployed - we live in tme of searching not finding, treating not curing...
    Last edited by Jarek Duda; March 16th, 2012 at 07:03 AM.
    Reply With Quote  
     

  87. #86  
    Quagma SpeedFreek's Avatar
    Join Date
    Jan 2011
    Location
    London, UK
    Posts
    2,786
    Well, the laws of physics do not preclude a smashed egg from leaping off the floor, putting itself back together again and landing gently on the worktop in pristine state, either.

    They just deem it highly unlikely.
    Reply With Quote  
     

  88. #87  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    It's because our natural reason-result chain provides us eggs only from our past: form our Big Bang, planet creation, evolution of life and finally the egg - it's a property of conrete solution we live in.
    But physics is time/CPT symmetric - if there would be "reversed initial conditions", analogue reason-result chain would go toward our past and create eggs which would also have tendency to increase entropy (break), but this time toward our past.

    You would tell: there cannot be made "reversed initial conditions", but let us for a moment assume (thought experiment) that our Universe will finally collapse.
    What entropy would have such Big Collapse? It would be very similar to Big bang - everything is completely localized (extremely low entropy), temperature erases everything - destroying all asymmetry.
    So shouldn't such hypothetical Big Collapse be similar to time reversed Big Bang?
    If so, it should start similar reason-result chain as our BB - for example leading to an egg, which would like to increase entropy - break .. this time toward our past.

    Another hypothetical way for "time reversed" egg is that your marvelous GRT doesn't forbid nonorientable spacetime, there are even such nonorientable constructions like slowly rotating black holes or wormholes. In the second post here are links to the papers: Thermodynamical thought experiment with time reversing loop?
    Time-nonorientability means that there is a loop, such that if one would travel through it, its past and future light cones would be literally switched (GRT doesn't distinguish past and future).
    So an "time reversed" egg from rocket going through such loop would also have tendency to increase entropy (break), but this time toward our past as you wanted.
    loop.jpg
    Last edited by Jarek Duda; March 16th, 2012 at 10:12 AM.
    Reply With Quote  
     

  89. #88  
    Time Lord
    Join Date
    Mar 2007
    Posts
    8,142
    Quote Originally Posted by SpeedFreek View Post
    Well, the laws of physics do not preclude a smashed egg from leaping off the floor, putting itself back together again and landing gently on the worktop in pristine state, either.

    They just deem it highly unlikely.
    It's an entropy problem, and like all entropy problems you can always reduce the entropy in one part of a system by causing more entropy somewhere else.

    A DCQE time communication device would send just a few bits of data back in time, at the cost creating a tremendous amount of forward time entropy elsewhere. Each microsecond jump back in time that the bits of information undergo involves emitting a large number of photons in order to notice the statistical abnormalities. Quite a lot of entropy is being created in order to turn back the clock on a very small part of the system.

    That's really all reverse time causality is. It's making entropy go in reverse.
    Some clocks are only right twice a day, but they are still right when they are right.
    Reply With Quote  
     

  90. #89  
    Suspended
    Join Date
    Aug 2009
    Location
    Israel
    Posts
    272
    Transmitting Information Time Machine

    The machine is designated for transmitting information back in time.
    According to the quantum rules and the uncertainty rules, you can cause a destruction of the wave function and influence the past. When we check that, we check the statistics with identical machines and close a circuit in order to transmit a message in negative time.

    The machine structure:
    The machine transmits a double photon every second fraction, at about 20 mega- hertz.
    In both sides there are poles.
    One pole is close; the second pole is far on the other end.
    If a photon passes in the near side, it transcripts mechanically the direction of the further pole, after the statistics with identical machines. We close a time circuit.

    Explanation:
    The photons have similar polarization because there were made together. Because of the uncertainty principle, the further pole determines the chance for a photon transition in the closer pole. Because one side is close and one side is far, there is a negative time difference. When the data stabilizes, the far pole determines the polarization that is similar to the photon that is close. A close pole is stabile, a far pole is changeable. The photons have a unique adjustment. According to the future polarization a photon comes out in the close pole. If a photon came out or didn’t come out means 0 or 1 logical. More statistics should be done with other machines. And with closing a time circuit , is how information goes back in time.
    If the photon comes out it determines mechanically the direction of the far pole.

    Time- table:
    From the further pole to the calcium atom about 30 meters of light are spared that are about 10 mega- hertz 100 nano second.
    About 1 giga hertz are spent on the statistics. About 1 nano second.
    On the pole about 30 mega hertz are spent. About 33 nano second.
    Another 66 mega hertz are spent on supplementary things. About 15 nano second
    About 50 negative nano second are left that are 20 negative mega hertz for closing a lasting circuit.

    Physical explanation:
    The causality doesn’t break; there is still a cause and a result. The time order is disrupted. The localization is broken. The changeable far pole determines a statistical polarization; the measurement determines the stabilization of data according to the uncertainty principle. If both poles determine different things there is a transition of 50%, but if the poles determine similar things, because of the influence of the pole on the polarization on both sides, there is a transition of 60%. It is a little bit like the EPR experiment in a circle.

    Notes:
    1. There is a possibility of overpowering the waste of time in distance by building a double turned over machine that is adjusted and parallel.
    2. The center of every doubled part is not in the middle.
    3. On every side you build 4 identical machines in order to do statistics.
    4. All together there are 8 identical machines.
    5. In order to do quick statistics of 50% you build a hardware of two logical portals: an “or” portal that is connected to 2 machines and a 2 “and” 2 portal. The statistics activates the identical double machines.
    6. The machine transmits one bit for 10 seconds back in time. (Because of the statistics the information gets lost with the negative time).
    7. It is a little bit like quantum games, only in a circle.

    Supplement parts that are not in the machine patent:
    1. Mechanical pole – because the electricity works too slowly. (We will wait for the invention). A disc of 6000 poles in a meter radios that spins about 50000 spins a second. With breaks that break it and dismantle it in a quick reaction in order to transmit a light bit to the sensor that is influenced statistically on the other end.
    2. Double photon shooter from the calcium atom. In the second level drop two adjusted photons are released in their spin. With two philters of wave lengths that determine contrasting exit directions.

    That is how we transmitted information back in time.

    Thank you,




    Transmitting Information Time Machine
    Reply With Quote  
     

  91. #90  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    kojax, I think you try to interpret thermodynamics as fundamental theory, while it's effective one - result not reason.
    Shannon's formula for entropy: H((p_i)_i) = -sum_i p_i lg(p_i) where p is probability ... while objectively there is some concrete scenario going on: p_i=0 or 1.
    Information/entropy is the proper approach to work with limited amount of subjective knowledge - we cannot use it to impose anything for fundamental physics: lagrangian mechanics just finding action optimizing history of field configuration...
    ... then on effective-thermodynamical level we can infer statistical properties.
    For example if many times there could be chosen one of 2 possibilities and there are completely no reasons to make some concrete assumptions about it, pure combinatorics says that we should assume that asymptotically almost certainly half of them will make the first choice. It's because p=0.5 case maximizes entropy, which is exponent coefficient in the number of all possibilities - subsets of different choices of probabilistic parameters asymptotically will be completely dominated:


    So while objectively there is a concrete e.g. atom configuration, we can assign to each point the number of atoms in a ball of small radius around this point - getting the density(/probability) function and other probabilistic/thermodynamical descriptions. Optimizing entropy/free energy for them, in phenomenological thermodynamics we get the most probable behavior.
    Now one can derive e.g. Boltzmann theorem that entropy is always growing ... but underlying fundamental physics is time/CPT symmetric: reversing time we would get analogous evolution and so Boltzmann theorem would also predict entropy growth, but this time in opposite direction.
    This theorem have to be understood that if there is an entropy minimum, there is statistical tendency to destroy this ordering - increase entropy. Entropy minimum creates entropy gradient ... and it works in both time directions.
    The entropy minimum "propelling" reason-result chains of our universe was our Big Bang - when everything had completely known position.

    Water Nosfim,
    if sending information back in time is possible, to get some we would need specialized receivers.
    And to send in future information back in time, obtaining this information we still will have to indeed want to send in back - physics has already made all hypothetical causal time-loops self consistent (eventually lying on some weakest links).
    This makes that only having possibility to send information back, it itself would affect the reality - for example after terrorist attack we could warn ourselves in the past and prevent it - so finally physics would probably made that it didn't happened (bending looking randomly for us thermodynamical dof) - to avoid the paradox.
    Here is some longer text about consequences of such hypothetical possibilities - I would gladly discuss about.
    But generally obtaining large time differences could become impractical and allowing physics to lie there (maybe using CPT analogue of laser?) ... but microscopic time difference would already allow for time-loop computers, e.g. making todays cryptography useless...
    Last edited by Jarek Duda; March 18th, 2012 at 06:17 AM.
    Reply With Quote  
     

  92. #91  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    There were/are some attempts to backtime channels by John Cramer ( John Cramers Retrocausal Experiments ), like:

    I don't see how it is supposed to work (or two other of his) - for me there is no erasure of quantum information by the sender here (?)
    I think it assumes that both entangled photons behaved the same on D-mirrors, ... but I completely don't see a base for such assumption(?)
    For me, to make something like that work, SPDC would need to be after splitting, like in File:Kim EtAl Quantum Eraser.svg - Wikipedia, the free encyclopedia ... but double SDPCs are very rare ...

    Do anybody understand why this or two other Cramer's configurations are supposed to work? Is some quantum erasure there?
    Reply With Quote  
     

  93. #92  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    In last week Nature Physics there is nice experiment about controlling in the future if photons in the past are entangled or not:

    Victor chooses later (QRNG) if Alice-Bob photons are correlated (left) or not (right):

    so obtaining |R>|R> means that more probably Victor has chosen entanglement, |R>|L> separation - there is nonzero mutual information, so once again I don't see a reason it couldn't be used to send information?
    Here is good informative article with link to the paper: Quantum decision affects results of measurements taken earlier in time

    ps. If someone is anxious about the "conflict" of fundamental time/CPT symmetry with our 2nd law-based intuition, it should be educative to look at very simple model: Kac ring - on a ring there are black and white balls which mutually shift one position each step. There are also some marked positions and when a ball goes through it, this ball switches color.
    Using natural statistical assumption ("Sto▀zahlansatz"): that if there is p such markings (proportionally), p of both black and white balls will change the color this step, we can easily prove that it should leads to equal number of black and white balls - maximizing the entropy ...
    ... from the other side, after two complete rotations all balls have to return to the initial color - from 'all balls white' fully ordered state, it would return back to it ... so the entropy would first increase to maximum and then symmetrically decrease back to minimum.
    Here is a good paper with simulations about it: http://www.maths.usyd.edu.au/u/gottw...s/kac-ring.pdf
    The lesson is that when on time/CPT symmetric fundamental physics we "prove" e.g. Boltzmann H theorem that entropy always grows ... we could take time symmetry transformation of this system and use the same "proof" to get that entropy always grows in backward direction - contradiction.
    The problem with such "profs" is that they always contain some very subtle uniformity assumption - generally called Sto▀zahlansatz. If underlying physics is time/CPT symmetric, we just cannot be sure that entropy will always grow - like for Kac ring and maybe our universe also ...
    Reply With Quote  
     

  94. #93  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    If someone does not belief in time/CPT symmetry of our physics:
    Reply With Quote  
     

  95. #94  
    Comet Dust Collector Moderator
    Join Date
    Mar 2011
    Location
    New Jersey, USA
    Posts
    2,847
    What does this have to do with the subject of this discussion?

    Meteor Wayne (as moderator)
    Reply With Quote  
     

  96. #95  
    Forum Junior
    Join Date
    Jul 2008
    Posts
    284
    The main issue of this discussion is understanding time/CPT symmetry of our physics (like in action optimizing Lagrangian mechanics used from QFT to GRT) - that on fundamental level there is no difference between past and future and so there is no reason to assume that causality goes only in one time direction (like in Wheeler's experiment or delayed quantum erasure).
    Unfortunately, in our society these fundamental symmetries are being replaced by asymmetry of our intuition and statistical properties of solution we live in (2nd law).

    This movie is a simple and beautiful reminder of this time symmetry, obstructed by some mental barrier in our society. And that in systems governed by time-symmetric equations (like for Kac ring I've mentioned before), claiming that disorder always grows is just wrong - required assumption ("Sto▀zahlansatz") just cannot be always true.
    Reply With Quote  
     

Similar Threads

  1. About causality
    By termina in forum Philosophy
    Replies: 3
    Last Post: January 23rd, 2012, 06:32 AM
  2. Replies: 1
    Last Post: December 12th, 2011, 06:07 PM
  3. Reverse time causality
    By kojax in forum Personal Theories & Alternative Ideas
    Replies: 2
    Last Post: September 1st, 2010, 11:33 PM
  4. Selective Erasure
    By gottspieler in forum Behavior and Psychology
    Replies: 8
    Last Post: March 27th, 2009, 01:26 PM
  5. Computer Controlled Telescope?
    By subodh in forum Astronomy & Cosmology
    Replies: 2
    Last Post: May 22nd, 2006, 12:46 PM
Bookmarks
Bookmarks
Posting Permissions
  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •