GSIC AUDIO

[Moose]

That's only because Mr. Anda insisted he (or his rep) had to be the ones to treat the disks. If neither the applicant nor his/her observer, nor the applicant's equipment have physical contact with the disks during the treatment, and proper double-blinding security is maintained to prevent information leakage, then it simply doesn't matter if the applicant handles the disks or plays with his/her own equipment during the test.

Restrictions in terms of the applicant's setup only gives the applicant wiggle room for excuses, as we've seen.

That's all just thoroughly unnecessary complication and expense which can be avoided entirely if JREF handles the treatments on neutral equipment, with the applicant's designated observer present.

Even if GSIC optimizes for the equipment on which it was applied, it doesn't matter. You're not looking for "detectably optimized". You're simply looking for "detectably different".

 

[LostAngeles]

Re: Please....

Uh, OK, we really don't need all this data about your day-to-day schedule. Reminds me of Anda, frankly, and I need no reminders.

The JREF itslef doesn't need ANY of this information, as the test date is determined by whomever will conduct the test - hopefully CFI-West in LA. We have nothing to do with it.

Regarding the protocol, once again we have a dummy chip.

It's gotta go. We rejected the notion of a dummy chip in the Anda negotiations, so I'm surprised it was even IN your protocol, especially considering how closely you followed that fiasco.

So let's kill that dummy chippy thingy and proceed asap.

I also strongly feel that this test will be immediately voided by any audiophile unless we can secure a truly excellent listening system. No walkmen or ipods. As the test will occur in LA, there are hundreds of audiophile houses who could be contacted about performing the test in an excellent, soundproof room with the best equipment. That's how it ought to be done.

I haven't read thru all the other posts on this thread but it seems there are many, so I'm sure I'll have more comments once I've seen them all.

Ah, I just wanted to thumb my nose at Anda's whining is all.

I noted there was a bunch of debate about the dummy chip but I wasn't exactly sure what was going on. Mainly because my eyes were rolling on every Wellfed post.

Reading on...

 

[LostAngeles]

Re: Re: Re: Sorry, but...

I don't see any problem with this, either.

I wonder why Mr. Anda did?

I don't wonder about it too much, though.

I'm not seeing a problem with this either, but let me finish reading, just in case, and I'll redraft the protocol.

 

[LostAngeles]

Really, this doesn't need to be very complicated at all.

You need eleven bit-by-bit identical disks. One of these disks is designated (and identified) as the control and is guaranteed to not have been treated.

The other ten are either treated or not treated by a coin toss, and recorded by paper in a sealed envelope and continuous video tape.

It's okay if a rep for the applicant is present, so long as this rep does not touch the equipment nor the coin, nor contacts the applicant in any way between that point and the time of the test. If any of these things happen, the test is invalidated.

The treatment machine is borrowed/provided by JREF, set once prior to the initial treatment, and not adjusted until the treatment is done.

This is done entirely out of sight of the applicant.

Once this is done, all "treaters" leave the area. One neutral, who was not present for the treatment, enters and transports the CDs to the room where the applicant awaits with their preferred equipment setup. (This can include packaging and mailing, in terms of a remote test.)

The applicant then receives all eleven CDs, and determines by any means they desire which CDs have been treated and which have not.

The applicant can handle the disks, or not handle the disks as he/she desires. The applicant may tweak the equipment at any point during the test. They may do bitwise comparisons, or as many listens as they like. They may switch CDs at will. The applicant can eat the disks if that is what it takes for them to tell the difference.

The applicant provides whatever equipment and environment they feel they require to make the determination, subject to inspection for communication vectors (smuggled cell phones, etc.)

KRAMER, I like this one. Other than removing the handling of the discs by me, should we go with this one?

 

[KRAMER]

Seems fine to me.

Comments? Forum members? Time to present it to Randi?

Oh and by the way, if you REALLY mean this to be an OFFICIAL protocol, you'll need to have it embossed in gold, notarized by Justice Renquist, and hand-courriered to Randi personally by a Nubian prince who will then wait for Randi's reply and run the 3,000 miles back to you. If he doesn't make it there within 48 hours, it's not an official protocol.

Sorry, but those are the rules, after all.

 

[Beleth]

I just want to add that at least one test CD must be treated. If, by some fluke, all ten coin flips indicate that no CD is to be treated, the sequence must be flipped again, from scratch, until at least one of the ten test CDs is treated.

I was going to point out that this changes the odds from 1024:1 up to 512:1, but on further reflection I think that it only changes them up to 1023:1.

Maybe there should be two controls: a known treated disc as well as the known untreated one.

 

[jmercer]

Maybe there should be two controls: a known treated disc as well as the known untreated one.

That's an interesting idea... how would you use them?

 

[LostAngeles]

Seems fine to me.

Comments? Forum members? Time to present it to Randi?

Oh and by the way, if you REALLY mean this to be an OFFICIAL protocol, you'll need to have it embossed in gold, notarized by Justice Renquist, and hand-courriered to Randi personally by a Nubian prince who will then wait for Randi's reply and run the 3,000 miles back to you. If he doesn't make it there within 48 hours, it's not an official protocol.

Sorry, but those are the rules, after all.

*snerk* You know, I was going to try and come up with a "reason" to not do it every week until the test, but since a) we should probably not kid around so much because of credibility concerns and b) you don't want to be reminded of Mr. Anda.

I believe my first one was going to involve gold-plated space hamsters.

 

[Moose]

I was going to point out that this changes the odds from 1024:1 up to 512:1, but on further reflection I think that it only changes them up to 1023:1.

Yes, it only removes one possibility "off the top". 1023:1 odds instead of 1024:1.

Maybe there should be two controls: a known treated disc as well as the known untreated one.

This would indeed restore that one possibility, and it isn't a bad idea at all. It would remove considerable (perceived) wiggle-room on the treatment side of things.

Kramer, that was funny.

 

[Moose]

That's an interesting idea... how would you use them?

You search for treated CDs against the untreated control. However, using a coin to decide which CDs are treated and which aren't opens the possibility that all test CDs are untreated. This renders the test indistinguishable from "couldn't do it".

So you add the treated control CD for direct comparison, so that the applicant can be definite about declaring that no test CD was treated.

The applicant may also wish to confirm "no treatment" against the treated control CD in any case.

It actually gives the applicant more play options without compromising the validity of the test any.

The more I think about this, the more I like it.

 

[BPSCG]

Seems fine to me.

Comments? Forum members?

What is the actual paranormal claim? Your OP on this thread says:

As Mr. Anda has led us to a place wherein we had no choice but to reject his claim, we have accepted an identical claim from an applicant who is also a forum member,

Emphasis mine. Wellfed's claim, IIRC, was that he could identify GSIC-treated CD's. LA expresses doubt (well-founded, IMO) that she herself can do it. That being the case, how is her claim identical to Wellfed's?

If it is not identical, then what exactly is her paranormal claim? If her claim is simply that she thinks there's a small chance that she might be able to do it, isn't that a serious lowering of the bar for a paranormal claim?

Apologies all around if the consensus is that I'm picking nits, especially as I appear to have already irritated Kramer enough for one day.

 

[Timothy]

Maybe there should be two controls: a known treated disc as well as the known untreated one.

In the interest of scientific integrity and honesty, this suggestion is a good one.

Rather than the four possible outcomes being:

- I could hear a difference, I selected the treated disks correctly, it works.
- I could hear a difference, I failed to select the treated disks correctly, it doesn't work.
- I couldn't really hear a difference, I selected the treated disks correctly (either statistical fluke, or device works subconsciously ... in which case you do the test over.)
- I couldn't really hear a difference, I failed to select the treated disks correctly, it doesn't work.

You now can add a fifth:

- I could not hear any difference between the known treated and untreated controls, therefore the results of the test are inconclusive.

The first protocol implies (not explicitly, but from the discussion raging here certainly implicitly among the more combative participants) that the results of the test will provide one of two answers --- CHIP WORKS, CHIP DOESN'T WORK.

The second, more intellectually honest protocol changes the outcomes to -- CHIP WORKS, CHIP DOESN'T WORK, TEST WAS INCONCLUSIVE.

If the fidelity of the system (including the listener) is not good enough to distinguish between known treated and untreated disk, you can't run a test on unknown disks with the same system and claim that the chip failed. That's simply intellectually dishonest, and I'm disenchanted to see JREF sanctioning a test with these limitations whose sole purpose seems to be an ego bashing exercise to show up a previous applicant.

- Timothy

 

[LostAngeles]

Assorted thoughts:

I am really not impressed by Mr. Anda at all. I'm certain he feels the same about Kramer and I, but I for one and I think I can reliably say Kramer for two, don't care. We really went from something simple to something overly and needlessly complex. It's like counting jelly beans using matrices of complex numbers. Why?

That's why I jumped into this. It's simple. It's straight forward. We got us a protocol in what? A day?

I want to make jokes on this a lot more than I am and I'm going to. It's only because of "credibility." I don't see why we shouldn't enjoy ourselves as we do this. I know it's serious, but that doesn't mean we can't have fun and be amicable and just enjoy it.

I don't see anyone as being "hostile" to the test, just hostile to waffling and picky demands. Everyone's pretty much saying, "Let's do this."

Lastly, I'd like to thank everyone... mostly everyone who's been involved in this because there's been a lot of good thinking and good suggestions on this and it's all just been very helpful and conducive to taking the test. Y'all are cool.

And none of that was mocking Anda, for once.

Forgot to add: Should this pass, I absolutely think we should start doing test with larger groups of people. Passing the Randi Challenge opens the door, IMO, to being able to say, "Hey, looks like we have something here. Let's really put this through the works and check it out."

Hell, if it passes, I'll freaking fund the larger tests.

 

[Timothy]

And as long as you're tossing out *no* treated disks, you should also toss out *all* treated disks for the same reason.

1022:1

- Timothy

 

[LostAngeles]

What is the actual paranormal claim? Your OP on this thread says: Emphasis mine. Wellfed's claim, IIRC, was that he could identify GSIC-treated CD's. LA expresses doubt (well-founded, IMO) that she herself can do it. That being the case, how is her claim identical to Wellfed's?

If it is not identical, then what exactly is her paranormal claim? If her claim is simply that she thinks there's a small chance that she might be able to do it, isn't that a serious lowering of the bar for a paranormal claim?

Apologies all around if the consensus is that I'm picking nits, especially as I appear to have already irritated Kramer enough for one day.

*hands BPSG a nit comb*

Someone has to do it.

I thought Wellfed's claim was that the GSIC made a audible difference that he could detect. Mine is that it makes an audible difference that I can detect. The difference is that he's an audiophile and I'm an average Jane. They're identical in that we're testing a product and not a person, that can do something that doesn't have an apparent scientific explanation.

 

[Timothy]

That's an interesting idea... how would you use them [treated and untreated control disks]?

Applicant gets the two labelled controls and the ten test disks. Applicant gets to swap back and forth as much as desired to try to distinguish the test disks as being treated or untreated.

As stated in previous post, if (skeptical) applicant can't distinguish between the two controls, any results showing inability to distinguish test disks must not be labelled "failure" but "inconclusive".

- Timothy

 

[drkitten]

The first protocol implies (not explicitly, but from the discussion raging here certainly implicitly among the more combative participants) that the results of the test will provide one of two answers --- CHIP WORKS, CHIP DOESN'T WORK.

The second, more intellectually honest protocol changes the outcomes to -- CHIP WORKS, CHIP DOESN'T WORK, TEST WAS INCONCLUSIVE.

If the fidelity of the system (including the listener) is not good enough to distinguish between known treated and untreated disk, you can't run a test on unknown disks with the same system and claim that the chip failed. That's simply intellectually dishonest, and I'm disenchanted to see JREF sanctioning a test with these limitations whose sole purpose seems to be an ego bashing exercise to show up a previous applicant.

I think I disagree. No, in fact, I'm fairly confident that I disagree. From a practical standpoint, this device is being marketed and sold with the claim that it will improve the sound of a CD, and not with the claim that it will improve the sound of particular kinds of CD's when used on specific kinds of high-end systems. Legally speaking, whenever you offer something for sale, there are two crucial implicit waranties -- "merchandisability" and "fitness for purpose." Basically, if the gadget has to do what the seller says it does, or the seller is at fault.

I therefore disapprove in principle of any test design where a fraudulent device that does nothing whatsoever cannot ever, even in principle, be proven not to work, because all tests return "inconclusive."

From an epistemological standpoint, the claim is that the chip works. If no one can detect any differences whatsoever, the chip does not work (under those circumstances). In which case another claimant may be able to produce another set of circumstsances under which the chip works -- but that's another claim.

So I disapprove in principle of any test where a complete failure of the claimant to perform as claimed can be regarded as "inconclusive."

 

[alfaniner]

You search for treated CDs against the untreated control. However, using a coin to decide which CDs are treated and which aren't opens the possibility that all test CDs are untreated. This renders the test indistinguishable from "couldn't do it".

...

The nature of the test is biased in favor of hearing nothing. (Meaning that by assumption, the treating of the discs will have no effect on them.)

Wouldn't this render the test very probably passable if for some reason they all wound up untreated?? (Granted, the probability of a coin flip coming up the same 10x in a row is the same probability needed for passing this test, right?)

1) Can't hear nuthin
2) Nope
3) Nuh-uh
4) Nada
5) Zip
6) Zilch
7) Negatory
8-10) (Same because I ran out of synonyms...)

So, if the applicant heard nothing at all, they'd get the million.

I'd suggest consulting a statistician for this, or at least do five and five which would eliminate the need for the extra control discs.

 

[Timothy]

Mine is that it makes an audible difference that I can detect.

The first part of your protocol, then, should be to confirm your ability to do it on the treated/untreated control pair. If you can't, the rest of the test is meaningless.

- Timothy

 

[drkitten]

As stated in previous post, if (skeptical) applicant can't distinguish between the two controls, any results showing inability to distinguish test disks must not be labelled "failure" but "inconclusive".

As stated in my previous post, if the applicant can't distinguish between the two controls, then the chip does not work.