FTL by Down-converting (Revised) [Archive]

Horace Heffner

Oct28-04, 01:28 PM

<jabberwocky><div class="vbmenu_control"><a href="jabberwocky:;" onClick="newWindow=window.open('','usenetCode','toolbar=no, location=no,scrollbars=yes,resizable=yes,status=no ,width=650,height=400'); newWindow.document.write('<HTML><HEAD><TITLE>Usenet ASCII</TITLE></HEAD><BODY topmargin=0 leftmargin=0 BGCOLOR=#F1F1F1><table border=0 width=625><td bgcolor=midnightblue><font color=#F1F1F1>This Usenet message\'s original ASCII form: </font></td></tr><tr><td width=449><br><br><font face=courier><UL><PRE>\nFTL by Down-converting (Revised)\n\nA method is proposed here to achieve faster than light (FTL) communication\nby the use of down-converters. A down-converter splits a photon into two\nphotons each having half the energy of the original photon.\n\nSuppose we have a sender Alice, a receiver Bob, and an intermediary\nfacilitator Charlie. Charlie uses a beam splitter to create two beams of\nlaser light: L the left beam and R, the right beam. Charlie then\ndown-converts the L beam to create beams L1 and L2, and similarly creates\nbeams R1 and R2 from the beam R. Beams R2 and L2 are normal path or\n"signal" photons through the down-converter, while beams R1 and L1 are\ncalled "idler" photons. "Beam" here means a flow of individually\ndetectable photons sent in very short intervals so as to provide a useful\nrate of communication. Charlie directs beams L1 and R1 to Alice and beams\nR2 and L2 to Bob. The corresponding photons arrive at both Bob and Alice\nat nearly the same time, but here assume Alice receives hers first, but\njust barely before Bob.\n\nBob directs beams R2 and L2 such that they can create an interference\npattern in a set of detectors arranged so it is feasible to rapidly and\nwith high probability determine whether an interference pattern is present\nor not. The signal photon beams R2 and L2 can create such an interference\npattern because they are the two paths from a beam splitter.\n\nAlice can direct her idler beams L1 and R1 at will, in a co-linear\nfashion, to opposing sides of a half silvered mirror, but at an angle of\n45 degrees. Fig.1, which requires fixed font (e.g courrier) to view,\nshows this configuration. Half of L1 and half of R1 then goes to a\ndetector DL. Similarly, half of L1 and half of R1 then goes to a detector\nDR. The beams emerging from both sides of the mirror are thus fully mixed,\nand the which-path information for all photons is lost. In this case Bob\nmust see an interference pattern. If Alice then diverts her beams\ndirectly to detectors, the which-way information is then restored to 100\npercent available, and Bob must see a bimodal distribution.\n\n\nFull Mirror\nR1--->-------------\\\n|\n| Half Mirror\nL1---->------------\\----------------------DL\n|\n|\nDR\n\nFig. 1 - Alice\'s which-path scrambler\n\n\nBob will in fact see such an interference pattern provided the which-path\ninformation is lost for idler beams R1 and L1.[1] If Alice does place\ndetectors directly in both her idler beams, then this is equivalent to\nknowing which path each of Bob\'s photons have traveled, and thus Bob can\nobserve no interference pattern. This known-path-no-interference result\nhas been characteristic of numerous versions of the two slit or two path\ninterference experiments.[2] If Alice detects directly and sees an idler\nshe knows which path the corresponding signal photon took to Bob, and the\ninterference wavefunction instantly collapses. Bob, when his photons\narrive shortly after Alice\'s corresponding photons, knows the current\nstate of Alice\'s detectors by whether he sees an interference pattern or\nnot.\n\nSince Alice and Bob could be light years away from each other, and since\nAlice thus might have years from the time Charlie released the photons to\nmake the choice to detect or not detect her photons, faster than light\ncommunication from Alice to Bob is clearly a possible result. It might be\nsaid that the communication can not be verified for years, but such\nverification is in this case not necessary. Bob does not require\nverification or comparison to Alice\'s results to know the immediate state\nof Alice\'s detectors, or to immediately detect a change of state of those\ndetectors, with sufficient speed and reliability to establish a practical\ncommunication channel. Further, a similar channel can be established from\nBob to Alice, thus permitting immediate error detection and correction or\nretransmission.\n\nAssuming that beams adequate for fast communication can be generated and\nthe resulting interference detected sufficiently fast, achieving high data\nrate FTL communication at short range then primarily boils down to how\nfast Alice can switch from a detecting mode to a non-detecting mode. This\nmight be as simple as her redirecting beams R1 and/or L1, or by switching\non and off the information from her detectors. This experiment then, in\naddition to achieving FTL communication, may be useful for determining\nexactly of what an observation consists.\n\nAn experiment requiring the simplest possible message would involve\nsending a data bit (actually only a change of state) via a one-way FTL\ncommunication channel and returning it via a second one-way return FTL\ncommunication channel, and repeating this process to establish an\noscillation. A fiber pair from Charlie to Bob and Charlie to Alice could\nbe used, if desired, to create a single FTL communication channel. A\nsimilar set of fiber pairs would be used for the return channel. To\ndemonstrate FTL communication it is then necessary to transmit over a\nsufficient distance D that the oscillation frequency, f, is faster than\nthe oscillation frequency F = c/D that can be achieved by light. A 10 km\ncommunication link (each way) need only cycle faster than about 15 kHz to\nbreak the light speed barrier. Assuming a sample of 100 photons to be\nsufficient for determining interference, a photon transmission and\ndetection rate of 1.5 million photons per second is required. However, it\nis not known what precisely constitutes an observation. It may be that\nindividual photon detection is not even necessary, but rather mere beam\nintensity determination is sufficient.\n\nReferences:\n\n[1] Kim et al, Phys. Rev. Lett., Vol 84, no. 1, pp 1-5\n[2] Brian Green, *The Fabric of the Cosmos*, (New York, Alfred A Knopf,\n2004), pp 193-197\n\nRegards,\n\nHorace Heffner\n</UL></PRE></font></td></tr></table></BODY><HTML>');"> <IMG SRC=/images/buttons/ip.gif BORDER=0 ALIGN=CENTER ALT="View this Usenet post in original ASCII form">  View this Usenet post in original ASCII form </a></div><P></jabberwocky>FTL by Down-converting (Revised)

A method is proposed here to achieve faster than light (FTL) communication
by the use of down-converters. A down-converter splits a photon into two
photons each having half the energy of the original photon.

Suppose we have a sender Alice, a receiver Bob, and an intermediary
facilitator Charlie. Charlie uses a beam splitter to create two beams of
laser light: L the left beam and R, the right beam. Charlie then
down-converts the L beam to create beams L1 and L2, and similarly creates
beams R1 and R2 from the beam R. Beams R2 and L2 are normal path or
"signal" photons through the down-converter, while beams R1 and L1 are
called "idler" photons. "Beam" here means a flow of individually
detectable photons sent in very short intervals so as to provide a useful
rate of communication. Charlie directs beams L1 and R1 to Alice and beams
R2 and L2 to Bob. The corresponding photons arrive at both Bob and Alice
at nearly the same time, but here assume Alice receives hers first, but
just barely before Bob.

Bob directs beams R2 and L2 such that they can create an interference
pattern in a set of detectors arranged so it is feasible to rapidly and
with high probability determine whether an interference pattern is present
or not. The signal photon beams R2 and L2 can create such an interference
pattern because they are the two paths from a beam splitter.

Alice can direct her idler beams L1 and R1 at will, in a co-linear
fashion, to opposing sides of a half silvered mirror, but at an angle of
45 degrees. Fig.1, which requires fixed font (e.g courrier) to view,
shows this configuration. Half of L1 and half of R1 then goes to a
detector DL. Similarly, half of L1 and half of R1 then goes to a detector
DR. The beams emerging from both sides of the mirror are thus fully mixed,
and the which-path information for all photons is lost. In this case Bob
must see an interference pattern. If Alice then diverts her beams
directly to detectors, the which-way information is then restored to 100
percent available, and Bob must see a bimodal distribution.

Full Mirror
R1--->-------------\
|
| Half Mirror
L1---->------------\----------------------DL
|
|
DR

Fig. 1 - Alice's which-path scrambler

Bob will in fact see such an interference pattern provided the which-path
information is lost for idler beams R1 and L1.[1] If Alice does place
detectors directly in both her idler beams, then this is equivalent to
knowing which path each of Bob's photons have traveled, and thus Bob can
observe no interference pattern. This known-path-no-interference result
has been characteristic of numerous versions of the two slit or two path
interference experiments.[2] If Alice detects directly and sees an idler
she knows which path the corresponding signal photon took to Bob, and the
interference wavefunction instantly collapses. Bob, when his photons
arrive shortly after Alice's corresponding photons, knows the current
state of Alice's detectors by whether he sees an interference pattern or
not.

Since Alice and Bob could be light years away from each other, and since
Alice thus might have years from the time Charlie released the photons to
make the choice to detect or not detect her photons, faster than light
communication from Alice to Bob is clearly a possible result. It might be
said that the communication can not be verified for years, but such
verification is in this case not necessary. Bob does not require
verification or comparison to Alice's results to know the immediate state
of Alice's detectors, or to immediately detect a change of state of those
detectors, with sufficient speed and reliability to establish a practical
communication channel. Further, a similar channel can be established from
Bob to Alice, thus permitting immediate error detection and correction or
retransmission.

Assuming that beams adequate for fast communication can be generated and
the resulting interference detected sufficiently fast, achieving high data
rate FTL communication at short range then primarily boils down to how
fast Alice can switch from a detecting mode to a non-detecting mode. This
might be as simple as her redirecting beams R1 and/or L1, or by switching
on and off the information from her detectors. This experiment then, in
addition to achieving FTL communication, may be useful for determining
exactly of what an observation consists.

An experiment requiring the simplest possible message would involve
sending a data bit (actually only a change of state) via a one-way FTL
communication channel and returning it via a second one-way return FTL
communication channel, and repeating this process to establish an
oscillation. A fiber pair from Charlie to Bob and Charlie to Alice could
be used, if desired, to create a single FTL communication channel. A
similar set of fiber pairs would be used for the return channel. To
demonstrate FTL communication it is then necessary to transmit over a
sufficient distance D that the oscillation frequency, f, is faster than
the oscillation frequency F = c/D that can be achieved by light. A 10 km
communication link (each way) need only cycle faster than about 15 kHz to
break the light speed barrier. Assuming a sample of 100 photons to be
sufficient for determining interference, a photon transmission and
detection rate of 1.5 million photons per second is required. However, it
is not known what precisely constitutes an observation. It may be that
individual photon detection is not even necessary, but rather mere beam
intensity determination is sufficient.

References:

[1] Kim et al, Phys. Rev. Lett., Vol 84, no. 1, pp 1-5
[2] Brian Green, *The Fabric of the Cosmos*, (New York, Alfred A Knopf,
2004), pp 193-197

Regards,

Horace Heffner

Horace Heffner

Oct30-04, 08:10 AM

<jabberwocky><div class="vbmenu_control"><a href="jabberwocky:;" onClick="newWindow=window.open('','usenetCode','toolbar=no, location=no,scrollbars=yes,resizable=yes,status=no ,width=650,height=400'); newWindow.document.write('<HTML><HEAD><TITLE>Usenet ASCII</TITLE></HEAD><BODY topmargin=0 leftmargin=0 BGCOLOR=#F1F1F1><table border=0 width=625><td bgcolor=midnightblue><font color=#F1F1F1>This Usenet message\'s original ASCII form: </font></td></tr><tr><td width=449><br><br><font face=courier><UL><PRE>\n\nThe following discussion of the subject faster than light (FTL)\ncommunication method, FTL by down-converting (Revised), is intended to\nshed some light on the design assumptions and places where I may have\nflawed thinking.\n\nA 1-1 photon-idler coincidence counting is vital to the Kim et al [1]\nexperiment, on which the FTL by down-converting method is based, because\nthat experiment is set up so that the paths of only about 50 percent of\nthe idlers is known. In the Kim experiment Bob can not see an\ninterference pattern at all, but when the photons are considered on an\nindividual basis, the signal photons corresponding to the idlers (via\nrecorded data) which are "observed" do not make an interference pattern,\nwhile those whose histories are "erased" by the configuration in Fig. 1 do\nmake an interference pattern. It is a fundamental assumption of the\nmethod that by knowing or not knowing the histories of 100 percent of the\nparticles Bob must clearly be able to distinguish a bimodal or\ninterference pattern accordingly, without knowledge of the timing or the\nneed for coincidence counting. If Bob dutifully records his\ndistributions, and 100 percent of the photons are affected by a quantum\nwavefunction that generates interference, then he must be able to see that\ninterference pattern because there are no photons left to mask it out.\nThere is no need for a later comparison with Alice, and thus no need for\nBob to record at all, other than to discriminate between the two patterns.\n\nThe significance of the Kim et al experiment is that *individual idler\nphoton histories can be wiped out* by scrambling that idler\'s history with\nother histories by injecting the idler into the *beam* from the opposed\npath. (There doesn\'t even have to be a particle-particle interaction, it\nis merely the fact that the origin of the mother photon becomes uncertain\nthat does the trick. Truly incredible!) Fig. 1 shows how an idler from\none beam can be (was) mixed into another idler beam, scrambled, so as to\nlose its history.\n\n\n\nFull Mirror\nR1--->-------------\\\n|\n| Half Mirror\nL1---->------------\\----------------------DL\n|\n|\nDR\n\nFig. 1 - Alice\'s which-path scrambler\n\n\nNote that ascii figures require a fixed font, like Courier. Microsoft\nOutlook users may need to select "fixed" in the "textsize" submenu of the\n"view" menu.\n\nThe fact that detector DR and DL can not determine whether an idler came\nfrom R1 or L1 in Fig. 1 *erases its history* and permits the\ncorresponding signal photon to experience the quantum wavefunction for\ninterference. DL detects half its particles from R1 and half from L1, as\ndoes detector DR. When the corresponding signal photon pattern is tallied\nfor all the photons detected by DL and DR, an interference pattern is\nobserved. The history of *every* photon passing through the scrambler\nthus must be erased, even those which pass straight through the\nhalf-mirror, i.e. Alice\'s beam splitter. The scrambler in Fig. 1 can be\nrepeated in series if necessary to compensate for imperfect beam splitting\nratios, imperfect beam overlap, and other problems.\n\nThe suggested method is a bit unusual because instead of utilizing an\ninstantaneous wavefunction collapse to achieve FTL it relies heavily on an\ninstantaneous *wavefunction resurrection.* By erasing the history that\ndestroys the wavefunction, the wavefunction is instantly resurrected, and\nthus the interference is projected at faster than the speed of light,\ninstantaneously, to Bob\'s location.\n\n\nFig. 2 and the following discussion sums up the critical issue for the\nproposed experiment design in a nutshell.\n\n\nDL\n|\n| Alice in\n| "erase history"\n^ mode\nR1 |\n/------------->----------//----->------->DR\n| |\n^ ^ L1\n| |\n... to Alice ... to Alice\n| |\n| |\nR1 | laser |\n| | | Legend:\n^ v ^\n| | | / - mirror\n| R | | // - splitter\n/----X----<----// Charlie | X - down-shifter\n| | | o - pattern detectors\n| v L | DL,DR - photon detectors\nv | L1 |\nR2 | X------>--------/\n| |\n...to Bob ... to Bob\n| |\n| |\nv v\no| | L2\no |\no------<------/\no\nBob\'s Pattern\n\n\nFig. 2 - Experiment with Alice in 100 % "erase history" mode.\n\n\nThe issue of whether the suggested experimental arrangement is capable of\ninformation transfer or not boils down to whether Bob sees an interference\npattern in the configuration diagramed in Fig. 2 or not (assuming path\nlengths are all adjusted properly). If he does, then the experiment must\nin fact be capable of transmitting information FTL, because it is well\nknown that which-way knowledge, for example obtained by eliminating\nAlice\'s splitter, and thus directing R1 to DR and L1 to DL, eliminates\nBob\'s interference pattern. Alice thus has numerous physical means\navailable to switch from interference mode to bimodal mode. The\n(experimental) reason Bob should see the interference pattern is that\nevery "history erased" photon entering Bob\'s detectors in the Kim et al\nexperiment contributed to the interference pattern. Since 100 percent of\nthe photons are "history erased" by Alice in Fig. 2, Bob should clearly\nsee the interference pattern. This is the fundamental assumption of the\ndesign and the most likely conceptual error.\n\nCausality and relativity arguments that the suggested method cannot\nachieve FTL communications are not germane, despite the fact they are\nlikely correct, because relativity and its related causality problems are\nthe very things tested by the proposed method. Though FTL communication\nmay indeed be impossible, it is hoped the subject method can reveal\nsomething about the nature of quantum events. Corrections to errors in\nthe underlying assumptions would be appreciated.\n\nReferences:\n\n[1] Kim et al, Phys. Rev. Lett., Vol 84, no. 1, pp 1-5\n\nRegards,\n\nHorace Heffner\n</UL></PRE></font></td></tr></table></BODY><HTML>');"> <IMG SRC=/images/buttons/ip.gif BORDER=0 ALIGN=CENTER ALT="View this Usenet post in original ASCII form">  View this Usenet post in original ASCII form </a></div><P></jabberwocky>The following discussion of the subject faster than light (FTL)
communication method, FTL by down-converting (Revised), is intended to
shed some light on the design assumptions and places where I may have
flawed thinking.

A 1-1[/itex] photon-idler coincidence counting is vital to the Kim et al [1]
experiment, on which the FTL by down-converting method is based, because
that experiment is set up so that the paths of only about 50 percent of
the idlers is known. In the Kim experiment Bob can not see an
interference pattern at all, but when the photons are considered on an
individual basis, the signal photons corresponding to the idlers (via
recorded data) which are "observed" do not make an interference pattern,
while those whose histories are "erased" by the configuration in Fig. 1 do
make an interference pattern. It is a fundamental assumption of the
method that by knowing or not knowing the histories of 100 percent of the
particles Bob must clearly be able to distinguish a bimodal or
interference pattern accordingly, without knowledge of the timing or the
need for coincidence counting. If Bob dutifully records his
distributions, and 100 percent of the photons are affected by a quantum
wavefunction that generates interference, then he must be able to see that
interference pattern because there are no photons left to mask it out.
There is no need for a later comparison with Alice, and thus no need for
Bob to record at all, other than to discriminate between the two patterns.

The significance of the Kim et al experiment is that *individual idler
photon histories can be wiped out* by scrambling that idler's history with
other histories by injecting the idler into the *beam* from the opposed
path. (There doesn't even have to be a particle-particle interaction, it
is merely the fact that the origin of the mother photon becomes uncertain
that does the trick. Truly incredible!) Fig. 1 shows how an idler from
one beam can be (was) mixed into another idler beam, scrambled, so as to
lose its history.

Full Mirror
R1--->-------------\
|
| Half Mirror
L1---->------------\----------------------DL
|
|
DR

Fig. 1 - Alice's which-path scrambler

Note that ascii figures require a fixed font, like Courier. Microsoft
Outlook users may need to select "fixed" in the "textsize" submenu of the
"view" menu.

The fact that detector DR and DL can not determine whether an idler came
from R1 or L1 in Fig. 1 *erases its history* and permits the
corresponding signal photon to experience the quantum wavefunction for
interference. DL detects half its particles from R1 and half from L1, as
does detector DR. When the corresponding signal photon pattern is tallied
for all the photons detected by DL and DR, an interference pattern is
observed. The history of *every* photon passing through the scrambler
thus must be erased, even those which pass straight through the
half-mirror, i.e. Alice's beam splitter. The scrambler in Fig. 1 can be
repeated in series if necessary to compensate for imperfect beam splitting
ratios, imperfect beam overlap, and other problems.

The suggested method is a bit unusual because instead of utilizing an
instantaneous wavefunction collapse to achieve FTL it relies heavily on an
instantaneous *wavefunction resurrection.* By erasing the history that
destroys the wavefunction, the wavefunction is instantly resurrected, and
thus the interference is projected at faster than the speed of light,
instantaneously, to Bob's location.

Fig. 2 and the following discussion sums up the critical issue for the
proposed experiment design in a nutshell.

DL
|
| Alice in
| "erase history"
^ mode
R1 |
/------------->----------//----->------->DR
| |^ ^ L1| |
... to Alice ... to Alice
| || |R1 | laser |
| | | Legend:
^ v ^| | | / - mirror
| R | | // - splitter
/----X----<----// Charlie | X - down-shifter
| | | o - pattern detectors
| v L | DL,DR - photon detectors
v | L1 |
R2 | X------>--------/
| |
...to Bob ... to Bob
| || |v vo| | L2
o |
o------<------/
o
Bob's Pattern

Fig. 2 - Experiment with Alice in 100 % "erase history" mode.

The issue of whether the suggested experimental arrangement is capable of
information transfer or not boils down to whether Bob sees an interference
pattern in the configuration diagramed in Fig. 2 or not (assuming path
lengths are all adjusted properly). If he does, then the experiment must
in fact be capable of transmitting information FTL, because it is well
known that which-way knowledge, for example obtained by eliminating
Alice's splitter, and thus directing R1 to DR and L1 to DL, eliminates
Bob's interference pattern. Alice thus has numerous physical means
available to switch from interference mode to bimodal mode. The
(experimental) reason Bob should see the interference pattern is that
every "history erased" photon entering Bob's detectors in the Kim et al
experiment contributed to the interference pattern. Since 100 percent of
the photons are "history erased" by Alice in Fig. 2, Bob should clearly
see the interference pattern. This is the fundamental assumption of the
design and the most likely conceptual error.

Causality and relativity arguments that the suggested method cannot
achieve FTL communications are not germane, despite the fact they are
likely correct, because relativity and its related causality problems are
the very things tested by the proposed method. Though FTL communication
may indeed be impossible, it is hoped the subject method can reveal
something about the nature of quantum events. Corrections to errors in
the underlying assumptions would be appreciated.

References:

[1] Kim et al, Phys. Rev. Lett., Vol 84, no. 1, [itex]pp 1-5

Regards,

Horace Heffner

Ralph Hartley

Nov2-04, 12:38 PM

<jabberwocky><div class="vbmenu_control"><a href="jabberwocky:;" onClick="newWindow=window.open('','usenetCode','toolbar=no, location=no,scrollbars=yes,resizable=yes,status=no ,width=650,height=400'); newWindow.document.write('<HTML><HEAD><TITLE>Usenet ASCII</TITLE></HEAD><BODY topmargin=0 leftmargin=0 BGCOLOR=#F1F1F1><table border=0 width=625><td bgcolor=midnightblue><font color=#F1F1F1>This Usenet message\'s original ASCII form: </font></td></tr><tr><td width=449><br><br><font face=courier><UL><PRE>\n\n\n\n\nHorace Heffner wrote:\n> The following discussion of the subject faster than light (FTL)\n> communication method, FTL by down-converting (Revised), is intended to\n> shed some light on the design assumptions and places where I may have\n> flawed thinking.\n\nOK but in exchange for my response I demand the right to assign homework.\nJust be glad I don\'t have a license to throw fireballs:-).\n\n> Fig. 1 shows how an idler from\n> one beam can be (was) mixed into another idler beam, scrambled, so as to\n> lose its history.\n\nThe laws of quantum mechanics do not permit information to be destroyed. It\ncan be scrambled, making it impractical to recover, but that isn\'t the same\n(and Fig 1 doesn\'t do that anyway).\n\n> Full Mirror\n> R1--->-------------\\\n> |\n> | Half Mirror\n> L1---->------------\\----------------------DL\n> |\n> |\n> DR\n>\n> Fig. 1 - Alice\'s which-path scrambler\n\nIt looks like this destroys information, but it does not.\n\nHomework: Does *Alice* see interference when she uses this device?\n\n> The suggested method is a bit unusual\n\nActually, it isn\'t even a bit unusual. It\'s in the same class as the "barn\ndoor" paradox in special relativity (and about equally instructive).\n\n> Fig. 2 - Experiment with Alice in 100 % "erase history" mode.\n\nHomework: Explain how this is just a different form of the Aspect experiment.\n\nDo the *math* and see what Quantum Mechanics says will happen.\n\nIf you don\'t know how to do the math, *learn*. It isn\'t that hard.\n\n> The issue of whether the suggested experimental arrangement is capable of\n> information transfer or not boils down to whether Bob sees an interference\n> pattern in the configuration diagramed in Fig. 2 or not\n\nHomework: Does he? Does Alice?\n\n(Hint: the apparatus is essentially the same if Bob and Alice trade places)\n\nDo Bob and Alice see interference when they compare notes?\n\nWhat pattern (if any) does Bob see if he only counts trials in which Alice\ndetects a photon at DL?\n\nHow is that related to what he sees on DR trials?\n\nWhen you understand the answer to the last question, you will understand\nsomething important! Something you didn\'t understand before.\n\n> Causality and relativity arguments that the suggested method cannot\n> achieve FTL communications are not germane, despite the fact they are\n> likely correct, because relativity and its related causality problems are\n> the very things tested by the proposed method.\n\nThey *are* germane, because it is a *theorem* that QM has those properties.\nSo any proposed FTL communication scheme has to violate the rules of QM\n*somewhere*. Just as any procedure to trisect an angle must violate the\nrules of straight edge and compass construction (or fail to work at all).\n\n> it is hoped the subject method can reveal something about the nature of quantum events.\n\nThat may be the case, but the effort will be wasted if you don\'t do your\nhomework.\n\nHomework (to be completed before you post your next version of an FTL\nscheme): Why does the next version not work? What would the actual results\nof the experiment be?\n\nUse the math. The math is your friend! The math is much easier to\nunderstand than most of your post, and it is always right.\n\nRalph Hartley\n</UL></PRE></font></td></tr></table></BODY><HTML>');"> <IMG SRC=/images/buttons/ip.gif BORDER=0 ALIGN=CENTER ALT="View this Usenet post in original ASCII form">  View this Usenet post in original ASCII form </a></div><P></jabberwocky>Horace Heffner wrote:
> The following discussion of the subject faster than light (FTL)
> communication method, FTL by down-converting (Revised), is intended to
> shed some light on the design assumptions and places where I may have
> flawed thinking.

OK but in exchange for my response I demand the right to assign homework.
Just be glad I don't have a license to throw fireballs:-).

> Fig. 1 shows how an idler from
> one beam can be (was) mixed into another idler beam, scrambled, so as to
> lose its history.

The laws of quantum mechanics do not permit information to be destroyed. It
can be scrambled, making it impractical to recover, but that isn't the same
(and Fig 1 doesn't do that anyway).

> Full Mirror
> R1--->-------------\
> |
> | Half Mirror
> L1---->------------\----------------------DL
> |
> |
> DR
>
> Fig. 1 - Alice's which-path scrambler

It looks like this destroys information, but it does not.

Homework: Does *Alice* see interference when she uses this device?

> The suggested method is a bit unusual

Actually, it isn't even a bit unusual. It's in the same class as the "barn
door" paradox in special relativity (and about equally instructive).

> Fig. 2 - Experiment with Alice in 100 % "erase history" mode.

Homework: Explain how this is just a different form of the Aspect experiment.

Do the *math* and see what Quantum Mechanics says will happen.

If you don't know how to do the math, *learn*. It isn't that hard.

> The issue of whether the suggested experimental arrangement is capable of
> information transfer or not boils down to whether Bob sees an interference
> pattern in the configuration diagramed in Fig. 2 or not

Homework: Does he? Does Alice?

(Hint: the apparatus is essentially the same if Bob and Alice trade places)

Do Bob and Alice see interference when they compare notes?

What pattern (if any) does Bob see if he only counts trials in which Alice
detects a photon at DL?

How is that related to what he sees on DR trials?

When you understand the answer to the last question, you will understand
something important! Something you didn't understand before.

> Causality and relativity arguments that the suggested method cannot
> achieve FTL communications are not germane, despite the fact they are
> likely correct, because relativity and its related causality problems are
> the very things tested by the proposed method.

They *are* germane, because it is a *theorem* that QM has those properties.
So any proposed FTL communication scheme has to violate the rules of QM
*somewhere*. Just as any procedure to trisect an angle must violate the
rules of straight edge and compass construction (or fail to work at all).

> it is hoped the subject method can reveal something about the nature of quantum events.

That may be the case, but the effort will be wasted if you don't do your
homework.

Homework (to be completed before you post your next version of an FTL
scheme): Why does the next version not work? What would the actual results
of the experiment be?

Use the math. The math is your friend! The math is much easier to
understand than most of your post, and it is always right.

Ralph Hartley

Bruce Zweig

Nov3-04, 09:57 AM

<jabberwocky><div class="vbmenu_control"><a href="jabberwocky:;" onClick="newWindow=window.open('','usenetCode','toolbar=no, location=no,scrollbars=yes,resizable=yes,status=no ,width=650,height=400'); newWindow.document.write('<HTML><HEAD><TITLE>Usenet ASCII</TITLE></HEAD><BODY topmargin=0 leftmargin=0 BGCOLOR=#F1F1F1><table border=0 width=625><td bgcolor=midnightblue><font color=#F1F1F1>This Usenet message\'s original ASCII form: </font></td></tr><tr><td width=449><br><br><font face=courier><UL><PRE>Your FTL will not work because you cannot get an interference pattern\nfrom the signal beams R2 and L2. I did a post called "Why quantum\nerasers cannot communicate backwards in time" which addresses this\nissue.\n\nGreene oversimplifies things a little too much in his noble efforts to\npopularize quantum physics, so he left out the part where the signal\nbeams and idler beams are coincidence counted. So far, no one ever\nhas gotten signal beams to interfere: you need to do a coincidence\ncount with the idler photons if you want to see the interference\npattern emerge, and this requires relativistic communication between\nthe detectors.\n\nA good paper on this subject is at\nhttp://arxiv.org/abs/quant-ph/0201036 by Chioa and Kwiat.\n\nA good forum to ask these kinds of questions is:\nhttp://www.physicsforums.com\n\nBruce\n\nhheffner@mtaonli ne.net (Horace Heffner) wrote in message news:<hheffner-2710041409430001@dialups-45.palmer.mtaonline.net>...\n> FTL by Down-converting (Revised)\n>\n> Beams R2 and L2 are normal path or\n> "signal" photons through the down-converter, while beams R1 and L1 are\n> called "idler" photons...\n>\n> Bob directs beams R2 and L2 such that they can create an interference\n> pattern in a set of detectors\n\n</UL></PRE></font></td></tr></table></BODY><HTML>');"> <IMG SRC=/images/buttons/ip.gif BORDER=0 ALIGN=CENTER ALT="View this Usenet post in original ASCII form">  View this Usenet post in original ASCII form </a></div><P></jabberwocky>Your FTL will not work because you cannot get an interference pattern
from the signal beams R2 and L2. I did a post called "Why quantum
erasers cannot communicate backwards in time" which addresses this
issue.

Greene oversimplifies things a little too much in his noble efforts to
popularize quantum physics, so he left out the part where the signal
beams and idler beams are coincidence counted. So far, no one ever
has gotten signal beams to interfere: you need to do a coincidence
count with the idler photons if you want to see the interference
pattern emerge, and this requires relativistic communication between
the detectors.

A good paper on this subject is at
http://arxiv.org/abs/http://www.arxiv.org/abs/quant-ph/0201036 by Chioa and Kwiat.

A good forum to ask these kinds of questions is:
http://www.physicsforums.com

Bruce

hheffner@mtaonline.net (Horace Heffner) wrote in message news:<hheffner-2710041409430001@dialups-45.palmer.mtaonline.net>...
> FTL by Down-converting (Revised)
>
> Beams R2 and L2 are normal path or
> "signal" photons through the down-converter, while beams R1 and L1 are
> called "idler" photons...
>
> Bob directs beams R2 and L2 such that they can create an interference
> pattern in a set of detectors

Mike Stay

Nov3-04, 09:59 AM

<jabberwocky><div class="vbmenu_control"><a href="jabberwocky:;" onClick="newWindow=window.open('','usenetCode','toolbar=no, location=no,scrollbars=yes,resizable=yes,status=no ,width=650,height=400'); newWindow.document.write('<HTML><HEAD><TITLE>Usenet ASCII</TITLE></HEAD><BODY topmargin=0 leftmargin=0 BGCOLOR=#F1F1F1><table border=0 width=625><td bgcolor=midnightblue><font color=#F1F1F1>This Usenet message\'s original ASCII form: </font></td></tr><tr><td width=449><br><br><font face=courier><UL><PRE>> Bob directs beams R2 and L2 such that they can create an interference\n> pattern in a set of detectors arranged so it is feasible to rapidly and\n> with high probability determine whether an interference pattern is present\n> or not. The signal photon beams R2 and L2 can create such an interference\n> pattern because they are the two paths from a beam splitter.\n\nNo, they\'re not. (R1 xor R2) and (L1 xor L2) are the two paths from\nthe beam splitter, and will show an interference pattern. All you get\nfrom R2 and L2 is a mixed state, and no pattern will be visible.\nYou\'re still mistaking averages for linear superpositions.\n\nMike\n\n</UL></PRE></font></td></tr></table></BODY><HTML>');"> <IMG SRC=/images/buttons/ip.gif BORDER=0 ALIGN=CENTER ALT="View this Usenet post in original ASCII form">  View this Usenet post in original ASCII form </a></div><P></jabberwocky>> Bob directs beams R2 and L2 such that they can create an interference
> pattern in a set of detectors arranged so it is feasible to rapidly and
> with high probability determine whether an interference pattern is present
> or not. The signal photon beams R2 and L2 can create such an interference
> pattern because they are the two paths from a beam splitter.

No, they're not. (R1 xor R2) and (L1 xor L2) are the two paths from
the beam splitter, and will show an interference pattern. All you get
from R2 and L2 is a mixed state, and no pattern will be visible.
You're still mistaking averages for linear superpositions.

Mike

Uncle Al

Nov3-04, 10:02 AM

<jabberwocky><div class="vbmenu_control"><a href="jabberwocky:;" onClick="newWindow=window.open('','usenetCode','toolbar=no, location=no,scrollbars=yes,resizable=yes,status=no ,width=650,height=400'); newWindow.document.write('<HTML><HEAD><TITLE>Usenet ASCII</TITLE></HEAD><BODY topmargin=0 leftmargin=0 BGCOLOR=#F1F1F1><table border=0 width=625><td bgcolor=midnightblue><font color=#F1F1F1>This Usenet message\'s original ASCII form: </font></td></tr><tr><td width=449><br><br><font face=courier><UL><PRE>Horace Heffner wrote:\n>\n> The following discussion of the subject faster than light (FTL)\n> communication method, FTL by down-converting (Revised), is intended to\n> shed some light on the design assumptions and places where I may have\n> flawed thinking.\n\nNo EPR experiment can transmit superluminal information, not even\nMorse code. You cannot square a circle or transect an angle by Greek\nrules, either.\n\n[snip]\n\n\n--\nUncle Al\nhttp://www.mazepath.com/uncleal/\n(Toxic URL! Unsafe for children and most mammals)\nhttp://www.mazepath.com/uncleal/qz.pdf\n\n</UL></PRE></font></td></tr></table></BODY><HTML>');"> <IMG SRC=/images/buttons/ip.gif BORDER=0 ALIGN=CENTER ALT="View this Usenet post in original ASCII form">  View this Usenet post in original ASCII form </a></div><P></jabberwocky>Horace Heffner wrote:
>
> The following discussion of the subject faster than light (FTL)
> communication method, FTL by down-converting (Revised), is intended to
> shed some light on the design assumptions and places where I may have
> flawed thinking.

No EPR experiment can transmit superluminal information, not even
Morse code. You cannot square a circle or transect an angle by Greek
rules, either.

[snip]

--
Uncle Al
http://www.mazepath.com/uncleal/
(Toxic URL! Unsafe for children and most mammals)
http://www.mazepath.com/uncleal/qz.pdf

Ilja Schmelzer

Nov3-04, 10:06 AM

<jabberwocky><div class="vbmenu_control"><a href="jabberwocky:;" onClick="newWindow=window.open('','usenetCode','toolbar=no, location=no,scrollbars=yes,resizable=yes,status=no ,width=650,height=400'); newWindow.document.write('<HTML><HEAD><TITLE>Usenet ASCII</TITLE></HEAD><BODY topmargin=0 leftmargin=0 BGCOLOR=#F1F1F1><table border=0 width=625><td bgcolor=midnightblue><font color=#F1F1F1>This Usenet message\'s original ASCII form: </font></td></tr><tr><td width=449><br><br><font face=courier><UL><PRE>"Horace Heffner" <hheffner@mtaonline.net> schrieb\n> FTL by Down-converting (Revised)\n\nDoesn\'t work.\n\n> Alice can direct her idler beams L1 and R1 at will, in a co-linear\n> fashion, to opposing sides of a half silvered mirror, but at an angle of\n> 45 degrees. Fig.1, which requires fixed font (e.g courrier) to view,\n> shows this configuration. Half of L1 and half of R1 then goes to a\n> detector DL. Similarly, half of L1 and half of R1 then goes to a detector\n> DR. The beams emerging from both sides of the mirror are thus fully mixed,\n> and the which-path information for all photons is lost. In this case Bob\n> must see an interference pattern. If Alice then diverts her beams\n> directly to detectors, the which-way information is then restored to 100\n> percent available, and Bob must see a bimodal distribution.\n\nI recommend you to compute the probabilities instead of using\nverbal descriptions like "must see an interference pattern".\nIt doesn\'t work. The point is that in constructions of your type\none does not see an interference pattern without additional\ninformation. One sees the sum of two interference patterns\nwhich does not look like an interference pattern.\nLater, with the information of what has been done at the other end,\none can decompose the visible result into two parts which look\nlike interference patterns, or two parts which don\'t look this way.\n\nBut you have to compute it yourself to see how this works.\n\nIlja\n\n</UL></PRE></font></td></tr></table></BODY><HTML>');"> <IMG SRC=/images/buttons/ip.gif BORDER=0 ALIGN=CENTER ALT="View this Usenet post in original ASCII form">  View this Usenet post in original ASCII form </a></div><P></jabberwocky>"Horace Heffner" <hheffner@mtaonline.net> schrieb
> FTL by Down-converting (Revised)

Doesn't work.

> Alice can direct her idler beams L1 and R1 at will, in a co-linear
> fashion, to opposing sides of a half silvered mirror, but at an angle of
> 45 degrees. Fig.1, which requires fixed font (e.g courrier) to view,
> shows this configuration. Half of L1 and half of R1 then goes to a
> detector DL. Similarly, half of L1 and half of R1 then goes to a detector
> DR. The beams emerging from both sides of the mirror are thus fully mixed,
> and the which-path information for all photons is lost. In this case Bob
> must see an interference pattern. If Alice then diverts her beams
> directly to detectors, the which-way information is then restored to 100
> percent available, and Bob must see a bimodal distribution.

I recommend you to compute the probabilities instead of using
verbal descriptions like "must see an interference pattern".
It doesn't work. The point is that in constructions of your type
one does not see an interference pattern without additional
information. One sees the sum of two interference patterns
which does not look like an interference pattern.
Later, with the information of what has been done at the other end,
one can decompose the visible result into two parts which look
like interference patterns, or two parts which don't look this way.

But you have to compute it yourself to see how this works.

Ilja

Jesse Mazer

Nov14-04, 07:08 AM

<jabberwocky><div class="vbmenu_control"><a href="jabberwocky:;" onClick="newWindow=window.open('','usenetCode','toolbar=no, location=no,scrollbars=yes,resizable=yes,status=no ,width=650,height=400'); newWindow.document.write('<HTML><HEAD><TITLE>Usenet ASCII</TITLE></HEAD><BODY topmargin=0 leftmargin=0 BGCOLOR=#F1F1F1><table border=0 width=625><td bgcolor=midnightblue><font color=#F1F1F1>This Usenet message\'s original ASCII form: </font></td></tr><tr><td width=449><br><br><font face=courier><UL><PRE>Bruce Zweig wrote:\n\n>Your FTL will not work because you cannot get an interference pattern\n>from the signal beams R2 and L2. I did a post called "Why quantum\n>erasers cannot communicate backwards in time" which addresses this\n>issue.\n>\n>Greene oversimplifies things a little too much in his noble efforts to\n>popularize quantum physics, so he left out the part where the signal\n>beams and idler beams are coincidence counted. So far, no one ever\n>has gotten signal beams to interfere: you need to do a coincidence\n>count with the idler photons if you want to see the interference\n>pattern emerge, and this requires relativistic communication between\n>the detectors.\n>\n>A good paper on this subject is at\n>http://arxiv.org/abs/quant-ph/0201036 by Chioa and Kwiat.\n>\n>A good forum to ask these kinds of questions is:\n>http://www.physicsforums.com\n>\n>Bruce\n>\n\nI\'m not sure exactly what you mean by \'a coincidence count\', but Greene\ndoes address the fact that you can only retroactively detect\ninterference in the subset of signal photons whose corresponding idler\nphotons have had their which-path information erased, if that\'s what\nyou\'re talking about:\n\n> Does this erasure of some of the which-path information--even though\n> we\'ve done nothing directly to the signal photon--mean interference\n> effects are recovered? Indeed it does--but only for those signal\n> photons whose idlers wind up in either detector 2 or detector 3.\n> Namely, the totality of impact positions of the signal photons on the\n> screen will look like the data in Figure 7.5a, showing not the\n> slightest hint of an interference pattern, as is characteristic of\n> photons that have travelled one path or the other. But if we focus on\n> a subset of the data points--for example, those signal photons whose\n> idlers entered detector 2--then that subset of points will fill out an\n> interference pattern! These signal photons--whose idlers happened, by\n> chance, not to provide any which-path information--act as though\n> they\'ve travelled both paths! If we were to hook up the equipment so\n> that the screen displays a red dot for the position of each signal\n> photon whose idler was detected by detector 2, and a green dot for all\n> others, someone who is color-blind would see no interference pattern,\n> but everyone else would see that the red dots were arranged with\n> bright and dark bands--an interference pattern. The same holds true\n> with detector 3 in place of detector 2. But there would be no such\n> interference pattern if we single out signal photons whose idlers wind\n> up in detector 1 or detector 4, since these are the idlers that yield\n> which-path information about their partners.\n\n\n(from \'The Fabric of the Cosmos\', p. 197)\n\nMy question would be, what if instead of using the setup described by\nGreene where the idler photons having a 50% chance of having their\nwhich-path information erased, you instead use a setup where all the\nidler photons will have their which-path information erased? Will that\nmean 100% of the signal photons form an interference pattern, which can\nbe seen before the idler photons have actually been detected? I assume\nthat can\'t be what happens, since that would leave open the possibility\nof then detecting all the idlers before they had their which-path\ninformation erased, but what\'s the explanation for *why* the signal\nphotons don\'t show interference even when 100% of the idler photons have\ntheir which-path information erased?\n\nJesse\n\n</UL></PRE></font></td></tr></table></BODY><HTML>');"> <IMG SRC=/images/buttons/ip.gif BORDER=0 ALIGN=CENTER ALT="View this Usenet post in original ASCII form">  View this Usenet post in original ASCII form </a></div><P></jabberwocky>Bruce Zweig wrote:

>Your FTL will not work because you cannot get an interference pattern
>from the signal beams R2 and L2. I did a post called "Why quantum
>erasers cannot communicate backwards in time" which addresses this
>issue.
>
>Greene oversimplifies things a little too much in his noble efforts to
>popularize quantum physics, so he left out the part where the signal
>beams and idler beams are coincidence counted. So far, no one ever
>has gotten signal beams to interfere: you need to do a coincidence
>count with the idler photons if you want to see the interference
>pattern emerge, and this requires relativistic communication between
>the detectors.
>
>A good paper on this subject is at
>http://arxiv.org/abs/http://www.arxiv.org/abs/quant-ph/0201036 by Chioa and Kwiat.
>
>A good forum to ask these kinds of questions is:
>http://www.physicsforums.com
>
>Bruce
>

I'm not sure exactly what you mean by 'a coincidence count', but Greene
does address the fact that you can only retroactively detect
interference in the subset of signal photons whose corresponding idler
photons have had their which-path information erased, if that's what
you're talking about:

> Does this erasure of some of the which-path information--even though
> we've done nothing directly to the signal photon--mean interference
> effects are recovered? Indeed it does--but only for those signal
> photons whose idlers wind up in either detector 2 or detector 3.
> Namely, the totality of impact positions of the signal photons on the
> screen will look like the data in Figure 7.5a, showing not the
> slightest hint of an interference pattern, as is characteristic of
> photons that have travelled one path or the other. But if we focus on
> a subset of the data points--for example, those signal photons whose
> idlers entered detector 2--then that subset of points will fill out an
> interference pattern! These signal photons--whose idlers happened, by
> chance, not to provide any which-path information--act as though
> they've travelled both paths! If we were to hook up the equipment so
> that the screen displays a red dot for the position of each signal
> photon whose idler was detected by detector 2, and a green dot for all
> others, someone who is color-blind would see no interference pattern,
> but everyone else would see that the red dots were arranged with
> bright and dark bands--an interference pattern. The same holds true
> with detector 3 in place of detector 2. But there would be no such
> interference pattern if we single out signal photons whose idlers wind
> up in detector 1 or detector 4, since these are the idlers that yield
> which-path information about their partners.

(from 'The Fabric of the Cosmos', p. 197)

My question would be, what if instead of using the setup described by
Greene where the idler photons having a 50% chance of having their
which-path information erased, you instead use a setup where all the
idler photons will have their which-path information erased? Will that
mean 100% of the signal photons form an interference pattern, which can
be seen before the idler photons have actually been detected? I assume
that can't be what happens, since that would leave open the possibility
of then detecting all the idlers before they had their which-path
information erased, but what's the explanation for *why* the signal
photons don't show interference even when 100% of the idler photons have
their which-path information erased?

Jesse

Ralph Hartley

Nov16-04, 02:59 AM

<jabberwocky><div class="vbmenu_control"><a href="jabberwocky:;" onClick="newWindow=window.open('','usenetCode','toolbar=no, location=no,scrollbars=yes,resizable=yes,status=no ,width=650,height=400'); newWindow.document.write('<HTML><HEAD><TITLE>Usenet ASCII</TITLE></HEAD><BODY topmargin=0 leftmargin=0 BGCOLOR=#F1F1F1><table border=0 width=625><td bgcolor=midnightblue><font color=#F1F1F1>This Usenet message\'s original ASCII form: </font></td></tr><tr><td width=449><br><br><font face=courier><UL><PRE>\n\nJesse Mazer wrote:\n\n> My question would be, what if instead of using the setup described by\n> Greene where the idler photons having a 50% chance of having their\n> which-path information erased, you instead use a setup where all the\n> idler photons will have their which-path information erased?\n\nThe problem is that there *isn\'t* any such setup!\n\nRalph Hartley\n</UL></PRE></font></td></tr></table></BODY><HTML>');"> <IMG SRC=/images/buttons/ip.gif BORDER=0 ALIGN=CENTER ALT="View this Usenet post in original ASCII form">  View this Usenet post in original ASCII form </a></div><P></jabberwocky>Jesse Mazer wrote:

> My question would be, what if instead of using the setup described by
> Greene where the idler photons having a 50% chance of having their
> which-path information erased, you instead use a setup where all the
> idler photons will have their which-path information erased?

The problem is that there *isn't* any such setup!

Ralph Hartley

Jesse Mazer

Nov17-04, 10:51 AM

<jabberwocky><div class="vbmenu_control"><a href="jabberwocky:;" onClick="newWindow=window.open('','usenetCode','toolbar=no, location=no,scrollbars=yes,resizable=yes,status=no ,width=650,height=400'); newWindow.document.write('<HTML><HEAD><TITLE>Usenet ASCII</TITLE></HEAD><BODY topmargin=0 leftmargin=0 BGCOLOR=#F1F1F1><table border=0 width=625><td bgcolor=midnightblue><font color=#F1F1F1>This Usenet message\'s original ASCII form: </font></td></tr><tr><td width=449><br><br><font face=courier><UL><PRE>Ralph Hartley wrote:\n\n>Jesse Mazer wrote:\n>\n>\n>\n>>My question would be, what if instead of using the setup described by\n>>Greene where the idler photons having a 50% chance of having their\n>>which-path information erased, you instead use a setup where all the\n>>idler photons will have their which-path information erased?\n>>\n>>\n>\n>The problem is that there *isn\'t* any such setup!\n>\n>Ralph Hartley\n>\n>\n\nSure there is--as I said in another post which I guess hasn\'t gotten\nthrough the moderators yet, if you look at the setup in Fig. 1 of the\npaper at http://xxx.lanl.gov/PS_cache/quant-ph/pdf/9903/9903047.pdf ,\nyou could replace the beam splitters BSA and BSB with mirrors, so that\nall the idler photons end up in detectors D1 or D2, in which case their\nwhich-path information has been erased. The joint detection rates of the\nsignal photons with the idlers at D1 is given by Fig. 3, and the joint\ndetection rates of the signal photons with the idlers at D2 is given by\nFig. 4--as I said in the other post, you can see that although both\nthese graphs show interference individually, the minima of Fig. 3 line\nup with the maxima of Fig. 4 and vice versa, so I think the solution is\nthat the total pattern of signal photons, which is the sum of these two\ngraphs, would not show any interference (it would look like Fig. 5) even\nin the case where all the idlers end up at D1 or D2 and have their\nwhich-path information erased.\n\nJesse\n\n</UL></PRE></font></td></tr></table></BODY><HTML>');"> <IMG SRC=/images/buttons/ip.gif BORDER=0 ALIGN=CENTER ALT="View this Usenet post in original ASCII form">  View this Usenet post in original ASCII form </a></div><P></jabberwocky>Ralph Hartley wrote:

>Jesse Mazer wrote:
>
>
>
>>My question would be, what if instead of using the setup described by
>>Greene where the idler photons having a 50% chance of having their
>>which-path information erased, you instead use a setup where all the
>>idler photons will have their which-path information erased?
>>
>>
>
>The problem is that there *isn't* any such setup!
>
>Ralph Hartley
>
>

Sure there is--as I said in another post which I guess hasn't gotten
through the moderators yet, if you look at the setup in Fig. 1 of the
paper at http://xxx.lanl.gov/PS_cache/quant-ph/pdf/9903/9903047.pdf ,
you could replace the beam splitters BSA and BSB with mirrors, so that
all the idler photons end up in detectors D1 or D2, in which case their
which-path information has been erased. The joint detection rates of the
signal photons with the idlers at D1 is given by Fig. 3, and the joint
detection rates of the signal photons with the idlers at D2 is given by
Fig. 4--as I said in the other post, you can see that although both
these graphs show interference individually, the minima of Fig. 3 line
up with the maxima of Fig. 4 and vice versa, so I think the solution is
that the total pattern of signal photons, which is the sum of these two
graphs, would not show any interference (it would look like Fig. 5) even
in the case where all the idlers end up at D1 or D2 and have their
which-path information erased.

Jesse

Jesse Mazer

Nov19-04, 12:07 PM

<jabberwocky><div class="vbmenu_control"><a href="jabberwocky:;" onClick="newWindow=window.open('','usenetCode','toolbar=no, location=no,scrollbars=yes,resizable=yes,status=no ,width=650,height=400'); newWindow.document.write('<HTML><HEAD><TITLE>Usenet ASCII</TITLE></HEAD><BODY topmargin=0 leftmargin=0 BGCOLOR=#F1F1F1><table border=0 width=625><td bgcolor=midnightblue><font color=#F1F1F1>This Usenet message\'s original ASCII form: </font></td></tr><tr><td width=449><br><br><font face=courier><UL><PRE>\n>My question would be, what if instead of using the setup described by\n>Greene where the idler photons having a 50% chance of having their\n>which-path information erased, you instead use a setup where all the\n>idler photons will have their which-path information erased? Will that\n>mean 100% of the signal photons form an interference pattern, which can\n>be seen before the idler photons have actually been detected? I assume\n>that can\'t be what happens, since that would leave open the possibility\n>of then detecting all the idlers before they had their which-path\n>information erased, but what\'s the explanation for *why* the signal\n>photons don\'t show interference even when 100% of the idler photons have\n>their which-path information erased?\n>\n\nOK, after discussing this on another forum (\nhttp://www.iidb.org/vbb/showthread.php?p=1972215 ) I think I see the\nanswer now. It\'s a lot clearer if you look at the diagrams at the bottom\nof the paper at\nhttp://xxx.lanl.gov/PS_cache/quant-ph/pdf/9903/9903047.pdf -- in Fig. 1,\nthe "signal photons" would be the ones that travel to the right after\nbeing emitted by either atom A or atom B, ending up in detector D0, and\nthe "idler photons" would be the ones that travel to the left after\nbeing emitted at the same moment by A or B, ending up in either\ndetectors D3 or D4 (which preserve the which-path information about\nwhich atom emitted them) or in detectors D1 or D2 (which erase this\nwhich-path information). The charts in Figures 3, 4 and 5 show how the\njoint detection rate between D0 and other detectors varies as you vary\nthe position of D0 along the x-axis, so you could just as easily imagine\nreplacing D0 with a continuous screen parallel to the x-axis, like in\nthe double-slit experiment. Then Fig. 3 would show the positions on the\nscreen of the subset of signal photons whose corresponding idlers ended\nup in detector D1, Fig. 4 would show the positions of the subset of\nsignal photons whose idlers ended up in D2, and Fig. 5 would show the\npositions of the subset of signal photons whose idlers ended up in D3 or D4.\n\nSo, if you modified the experiment by replacing the beam splitters BSA\nand BSB with mirrors, insuring that *all* the idlers ended up in D1 or\nD2 and thus had their which-path information erased, the total pattern\nof signal photons hitting the screen should be the sum of Fig. 3 and\nFig. 4. Even though individually Fig. 3 and Fig. 4 show an interference\npattern, the peaks of one match up with the valleys of the other, so\ntheir sum would look just like Fig. 5, showing no interference. Thus\neven in this experiment you can\'t detect any interference until *after*\nthe idlers have been detected and you look at the subset of signal\nphotons whose idlers ended up in one of the two detectors. So there\'s no\nviolation of complementarity here, and also no potential for FTL or\nbackwards-in-time communication.\n\nI don\'t understand Ralph Hartley\'s earlier comment that if you modify\nthe experiment so 100% of the idler photons have their which-path\ninformation erased, as in the setup I described above, this would just\nbe "a different form of the Aspect experiment." Isn\'t the Aspect\nexperiment about comparing the spins of entangled photons and seeing a\nviolation of the Bell inequality?\n\nJesse\n</UL></PRE></font></td></tr></table></BODY><HTML>');"> <IMG SRC=/images/buttons/ip.gif BORDER=0 ALIGN=CENTER ALT="View this Usenet post in original ASCII form">  View this Usenet post in original ASCII form </a></div><P></jabberwocky>>My question would be, what if instead of using the setup described by
>Greene where the idler photons having a 50% chance of having their
>which-path information erased, you instead use a setup where all the
>idler photons will have their which-path information erased? Will that
>mean 100% of the signal photons form an interference pattern, which can
>be seen before the idler photons have actually been detected? I assume
>that can't be what happens, since that would leave open the possibility
>of then detecting all the idlers before they had their which-path
>information erased, but what's the explanation for *why* the signal
>photons don't show interference even when 100% of the idler photons have
>their which-path information erased?
>

OK, after discussing this on another forum (
http://www.iidb.org/vbb/showthread.php?p=1972215 ) I think I see the
answer now. It's a lot clearer if you look at the diagrams at the bottom
of the paper at
http://xxx.lanl.gov/PS_cache/quant-ph/pdf/9903/9903047.pdf -- in Fig. 1,
the "signal photons" would be the ones that travel to the right after
being emitted by either atom A or atom B, ending up in detector D0, and
the "idler photons" would be the ones that travel to the left after
being emitted at the same moment by A or B, ending up in either
detectors D3 or D4 (which preserve the which-path information about
which atom emitted them) or in detectors D1 or D2 (which erase this
which-path information). The charts in Figures 3, 4 and 5 show how the
joint detection rate between D0 and other detectors varies as you vary
the position of D0 along the x-axis, so you could just as easily imagine
replacing D0 with a continuous screen parallel to the x-axis, like in
the double-slit experiment. Then Fig. 3 would show the positions on the
screen of the subset of signal photons whose corresponding idlers ended
up in detector D1, Fig. 4 would show the positions of the subset of
signal photons whose idlers ended up in D2, and Fig. 5 would show the
positions of the subset of signal photons whose idlers ended up in D3 or D4.

So, if you modified the experiment by replacing the beam splitters BSA
and BSB with mirrors, insuring that *all* the idlers ended up in D1 or
D2 and thus had their which-path information erased, the total pattern
of signal photons hitting the screen should be the sum of Fig. 3 and
Fig. 4. Even though individually Fig. 3 and Fig. 4 show an interference
pattern, the peaks of one match up with the valleys of the other, so
their sum would look just like Fig. 5, showing no interference. Thus
even in this experiment you can't detect any interference until *after*
the idlers have been detected and you look at the subset of signal
photons whose idlers ended up in one of the two detectors. So there's no
violation of complementarity here, and also no potential for FTL or
backwards-in-time communication.

I don't understand Ralph Hartley's earlier comment that if you modify
the experiment so 100% of the idler photons have their which-path
information erased, as in the setup I described above, this would just
be "a different form of the Aspect experiment." Isn't the Aspect
experiment about comparing the spins of entangled photons and seeing a
violation of the Bell inequality?

Jesse

Ralph Hartley

Nov19-04, 01:32 PM

<jabberwocky><div class="vbmenu_control"><a href="jabberwocky:;" onClick="newWindow=window.open('','usenetCode','toolbar=no, location=no,scrollbars=yes,resizable=yes,status=no ,width=650,height=400'); newWindow.document.write('<HTML><HEAD><TITLE>Usenet ASCII</TITLE></HEAD><BODY topmargin=0 leftmargin=0 BGCOLOR=#F1F1F1><table border=0 width=625><td bgcolor=midnightblue><font color=#F1F1F1>This Usenet message\'s original ASCII form: </font></td></tr><tr><td width=449><br><br><font face=courier><UL><PRE>Jesse Mazer wrote:\n> Ralph Hartley wrote:\n>\n>>Jesse Mazer wrote:\n>>\n>>>My question would be, what if instead of using the setup described by\n>>>Greene where the idler photons having a 50% chance of having their\n>>>which-path information erased, you instead use a setup where all the\n>>>idler photons will have their which-path information erased?\n>>\n>>The problem is that there *isn\'t* any such setup!\n\nPerhaps my statement was a bit too broad. You can\'t exactly erase the which\npath information, but you can convert it into a *different* bit of information.\n\n> Sure there is--as I said in another post which I guess hasn\'t gotten\n> through the moderators yet, if you look at the setup in Fig. 1 of the\n> paper at http://xxx.lanl.gov/PS_cache/quant-ph/pdf/9903/9903047.pdf ,\n> you could replace the beam splitters BSA and BSB with mirrors, so that\n> all the idler photons end up in detectors D1 or D2, in which case their\n> which-path information has been erased. The joint detection rates of the\n> signal photons with the idlers at D1 is given by Fig. 3, and the joint\n> detection rates of the signal photons with the idlers at D2 is given by\n> Fig. 4--as I said in the other post, you can see that although both\n> these graphs show interference individually, the minima of Fig. 3 line\n> up with the maxima of Fig. 4 and vice versa, so I think the solution is\n> that the total pattern of signal photons, which is the sum of these two\n> graphs, would not show any interference (it would look like Fig. 5) even\n> in the case where all the idlers end up at D1 or D2 and have their\n> which-path information erased.\n\nSo far as I can tell this is entirely correct.\n\nThe experiment described is actually just a different form of EPR experiment.\n\nInstead of having two electrons with the same spin, you have two photons\nthat take the same path.\n\nSuppose the two observers just measure which path. They both get the same\nresults just as if they had measured the spin relative to the same axis.\n\nCombining the two paths with a beam splitter corresponds to changing the\nthe axis along which the spin is measured. The new axis depends on the\nrelative length of the two paths, and on the percentage reflectance of the\nbeam splitter.\n\nNote that the mechanism for "erasing" which-path information is *identical*\nto a device for detecting interference, combine the paths with a beam\nsplitter and observe where the photon comes out.\n\nIf they both combine beams with a 50% beam splitter, and are careful that\nall the path lengths are the same (modulo the wavelength), then the same\ndetector will fire for them both every time, just as if they had measured\nthe spin about the same axis, at right angles to the original axis.\n\nIf one of them changes one of the path lengths by a quarter of a\nwavelength, or if one uses a 50% mirror and the other uses a 0% mirror,\nthey will see no correlation at all, just as if they had measured the spins\nabout orthogonal axes.\n\nBy doing something in between, they could observe the violation of Bells\ninequality.\n\nRalph Hartley\n\n</UL></PRE></font></td></tr></table></BODY><HTML>');"> <IMG SRC=/images/buttons/ip.gif BORDER=0 ALIGN=CENTER ALT="View this Usenet post in original ASCII form">  View this Usenet post in original ASCII form </a></div><P></jabberwocky>Jesse Mazer wrote:
> Ralph Hartley wrote:
>
>>Jesse Mazer wrote:
>>
>>>My question would be, what if instead of using the setup described by
>>>Greene where the idler photons having a 50% chance of having their
>>>which-path information erased, you instead use a setup where all the
>>>idler photons will have their which-path information erased?
>>
>>The problem is that there *isn't* any such setup!

Perhaps my statement was a bit too broad. You can't exactly erase the which
path information, but you can convert it into a *different* bit of information.

> Sure there is--as I said in another post which I guess hasn't gotten
> through the moderators yet, if you look at the setup in Fig. 1 of the
> paper at http://xxx.lanl.gov/PS_cache/quant-ph/pdf/9903/9903047.pdf ,
> you could replace the beam splitters BSA and BSB with mirrors, so that
> all the idler photons end up in detectors D1 or D2, in which case their
> which-path information has been erased. The joint detection rates of the
> signal photons with the idlers at D1 is given by Fig. 3, and the joint
> detection rates of the signal photons with the idlers at D2 is given by
> Fig. 4--as I said in the other post, you can see that although both
> these graphs show interference individually, the minima of Fig. 3 line
> up with the maxima of Fig. 4 and vice versa, so I think the solution is
> that the total pattern of signal photons, which is the sum of these two
> graphs, would not show any interference (it would look like Fig. 5) even
> in the case where all the idlers end up at D1 or D2 and have their
> which-path information erased.

So far as I can tell this is entirely correct.

The experiment described is actually just a different form of EPR experiment.

Instead of having two electrons with the same spin, you have two photons
that take the same path.

Suppose the two observers just measure which path. They both get the same
results just as if they had measured the spin relative to the same axis.

Combining the two paths with a beam splitter corresponds to changing the
the axis along which the spin is measured. The new axis depends on the
relative length of the two paths, and on the percentage reflectance of the
beam splitter.

Note that the mechanism for "erasing" which-path information is *identical*
to a device for detecting interference, combine the paths with a beam
splitter and observe where the photon comes out.

If they both combine beams with a 50% beam splitter, and are careful that
all the path lengths are the same (modulo the wavelength), then the same
detector will fire for them both every time, just as if they had measured
the spin about the same axis, at right angles to the original axis.

If one of them changes one of the path lengths by a quarter of a
wavelength, or if one uses a 50% mirror and the other uses a 0% mirror,
they will see no correlation at all, just as if they had measured the spins
about orthogonal axes.

By doing something in between, they could observe the violation of Bells
inequality.

Ralph Hartley

bruce2g

Nov25-04, 03:42 AM

<jabberwocky><div class="vbmenu_control"><a href="jabberwocky:;" onClick="newWindow=window.open('','usenetCode','toolbar=no, location=no,scrollbars=yes,resizable=yes,status=no ,width=650,height=400'); newWindow.document.write('<HTML><HEAD><TITLE>Usenet ASCII</TITLE></HEAD><BODY topmargin=0 leftmargin=0 BGCOLOR=#F1F1F1><table border=0 width=625><td bgcolor=midnightblue><font color=#F1F1F1>This Usenet message\'s original ASCII form: </font></td></tr><tr><td width=449><br><br><font face=courier><UL><PRE>I had the same thought, and I found out that so far no one can get the\ndown converted signal photons to interfere by themselves. There\'s a\npaper by Shimizu et. al.(www.arxiv.org/abs/quant-ph/0210142) in which\nthe authors get normal photons to interfere, and then they get SDLC\ncoincident pairs to interfere, but they just can\'t get the darn signal\nphotons by themselves to interfere. A paper by Altschul & Altschul\n(www.arxiv.org/abs/quant-ph/0106113) which describes the erasure\nexperiment makes the statement "This process produces double-slit\ninterference if and only if it is subjected to coincidence counting\nwith an A+B idler beam." Finally, there\'s a nice discussion at the end\nof www.arxiv.org/abs/quant-ph/0201036 by Chiao and Kwiat that talks\nabout some hypotheses concerning the role of the coincidence counts.\n\nBasically, Greene had to oversimplify the erasure experiment in his\nexplanation in order to make it even remotely comprehensible to general\nreaders; so he apparently left out the part where the interference\npattern is obtained through coincidence counting. It would be nice to\nknow if it is even theoretically possible to obtain an interference\npattern among the down converted signal photons in the absence of\ncoincident counting. Everything I\'ve read says it isn\'t, though I\nhaven\'t seen a proof yet.\n\nBruce Zweig\n\n------------------------------------------------------------------------\nThis post submitted through the LaTeX-enabled physicsforums.com\nTo view this post with LaTeX images:\nhttp://www.physicsforums.com/showthread.php?t=50120#post359402\n\n</UL></PRE></font></td></tr></table></BODY><HTML>');"> <IMG SRC=/images/buttons/ip.gif BORDER=0 ALIGN=CENTER ALT="View this Usenet post in original ASCII form">  View this Usenet post in original ASCII form </a></div><P></jabberwocky>I had the same thought, and I found out that so far no one can get the
down converted signal photons to interfere by themselves. There's a
paper by Shimizu et. al.(www.arxiv.org/abs/http://www.arxiv.org/abs/quant-ph/0210142) in which
the authors get normal photons to interfere, and then they get SDLC
coincident pairs to interfere, but they just can't get the darn signal
photons by themselves to interfere. A paper by Altschul & Altschul
(www.arxiv.org/abs/http://www.arxiv.org/abs/quant-ph/0106113) which describes the erasure
experiment makes the statement "This process produces double-slit
interference if and only if it is subjected to coincidence counting
with an A+B idler beam." Finally, there's a nice discussion at the end
of www.arxiv.org/abs/http://www.arxiv.org/abs/quant-ph/0201036 by Chiao and Kwiat that talks
about some hypotheses concerning the role of the coincidence counts.

Basically, Greene had to oversimplify the erasure experiment in his
explanation in order to make it even remotely comprehensible to general
readers; so he apparently left out the part where the interference
pattern is obtained through coincidence counting. It would be nice to
know if it is even theoretically possible to obtain an interference
pattern among the down converted signal photons in the absence of
coincident counting. Everything I've read says it isn't, though I
haven't seen a proof yet.

Bruce Zweig

------------------------------------------------------------------------
This post submitted through the LaTeX-enabled physicsforums.com
To view this post with LaTeX images:
http://www.physicsforums.com/showthread.php?t=50120#post359402