[elnero]

Edwood,

You may want to get people to refrain from posting until everyone has had a chance to listen. That way people aren't influenced by others observations. If people wanted to get their impressions down on paper so to speak maybe they could PM their thought to you and you could post them at a later date.

Oh and if it does make it into Canada you can count me in.

 

[zachary80]

Sounds like a great dea, but the main problem will be deciding who is and isn't allowed to get the cables (and keeping track).

I agree that people should not post there findings before everyone has had their turn. Maybe a good way would be to send all the information to an email address for safekeeping

 

[Edwood]

Quote:

Originally Posted by elnero Edwood, You may want to get people to refrain from posting until everyone has had a chance to listen. That way people aren't influenced by others observations. If people wanted to get their impressions down on paper so to speak maybe they could PM their thought to you and you could post them at a later date. Oh and if it does make it into Canada you can count me in.

That could be part of the fun. Bandwagon jumping? I think I may label them by symbols rather than numbers or letters. Maybe one is "circle" and the other "square" or maybe "triangle".

I'm thinking maybe a wild card could be thrown in here. Maybe there could be three IC's.

1. Cotton insulated 20awg 99.99% Solid silver
2. Canare Starquad (should I terminate it half balanced? Ground the shield braid?)
3. The guts of the crappiest cable I can find

I bought some PVC tubing to play with today. The silver "twisted pair" just barely fits (the cotton is pretty thick). I think it may be a good idea to put the "guts" of the Canare Starquad inside the same PVC tubing, and like wise with the "crappiest cable I can find". That way, they would all "feel" the same. and would be closer in flexibility to each other.

So one would be labeled "circle", "square", and "triangle".

-Ed

 

[Edwood]

Quote:

Originally Posted by zachary80 Sounds like a great dea, but the main problem will be deciding who is and isn't allowed to get the cables (and keeping track). I agree that people should not post there findings before everyone has had their turn. Maybe a good way would be to send all the information to an email address for safekeeping

Hmmmm. What does everyone think?

A. New thread once Test is out and about, where everyone posts whatever they want. Which symbol they like better, and why. Guessing as to which cable is what. I reveal which symbols are which cable when the Test is over and all the cables are back to me.

B. New thread once Test is out and about , where everyone posts ONLY posts which order they prefer each cable. e.g. 1.Circle, 2.Square, 3.Triangle, or some standard format like that. I reveal which symbols are which cable when the Test is over and all the cables are back to me.

C. This thread is closed, people email/PM me their results, and I post them when the Test is over and all cables are back to me, and reveal which symbols are which cable.

D. This thread remains open, people email/PM me their results, and I post them when the Test is over and all cables are back to me. Everyone tries to refrain from mentioning specifics, like which symbol they like better and why. I reveal which symbols are which cable.

-Ed

 

[Earwax]

I don't think people should post any results until everyone has had a chance to listen. I think there's too large a chance that seeing posts will influence the later testers.

I marked on poll that I'm interested, but I'd best warn you that I don't expect to hear any difference in your 2 cable types, though I've heard (or at least think I've heard) really bad thin plastic cables grunge up a system.

 

[eric343]

Quote:

Originally Posted by Dane What is described here isn't a real blind test - the subject knows when the switch is done. A single blind test (which this is not) means that the subject doesn't know when the switch is done, but the assistent does. A real DBT means that both the subject and the assistent don't know what's going on. A third person is needed in this case.

So to do a true double blind test you'd need three people and at least three cables?

The first person would build two identical and one different cables. The second would label them in series (colors or letters or something). The third would then listen and try to determine which cables are different and which are the same, and what the sonic qualities of each are.

That way, the tester wouldn't know when he's switching (since he doesn't know which are the same) or what he's "supposed" to be hearing.

 

[acs236]

I'm in.

 

[ooheadsoo]

Definitely option C. We don't want any lemmingism at all showing up.

 

[fewtch]

Quote:

Originally Posted by Canman It is important to come up with a solid hypothesis so that the results are less arguable. The way I see it, a decent hypothesis is if a statistically significant majority is able to identify the correct interconnect, then there are audible differences between different conductor types (or just different cable).

One test would not be enough to prove this... after all, suppose you tossed a coin and it came up heads 8/10 times... does it prove something?

I'd say do this test on both coasts (as proposed) but use completely different colored cable jackets for each coast (e.g. blue and red on the West, green and orange on the East), gather the results, then do the test again -- swapping the cables between coasts! Then when it's all done, collate the results and make them available. IMO this would be a bit better than just a single test, although there's still an element of chance involved.

 

[WmAx]

Quote:

I want to make two visually identical InterConnect's. Label one a generic "A" and the other "B".

I believe that modification of your methodology would yeild more valuable results then your initially proposed test format.

I suggest use of an ABX-type of format. Make up at minimum 10 test sets of cables. Each set should be consist of 3 cable pairs. Pair A, Pair B and Pair X(control). The test subject will try to identify A or B to X. 30 pairs of cables may seem excessive at first mention, but this is needed in order to gather data of value, statistically, from a subject.

Before you begin to ship out cables for testing, they should be shipped to a naysayer(if that's the nickname you feel is appropriate). This person will sort, seal and mark the cables. Record the cable ID data and keep data sealed until the test cycle is complete. Note: For maximally valuable data in this test, ABX cable order should be re-arranged for every test subject. This would at least minimize possible cheating(for example a test subject that figured out which cables were which by whichever manner and then shared this data with other test subjects, thus propogating more errors). Cables must be tamper proofed in some manner. But nothing is iron-clad. I'm sure their is still some way to find out materials, perhaps measuring the respective effects within a magnetic field, like a metal detector? Perhaps the cables can be encased within a copper mesh tubing, in order to make such a cheating method difficult? After test cycle is complete, results need to be observed, but not yet released. If a person scores a significantly high result, then they should be re-tested with teh cables re-arranged. This is due to statisical law. In a large number of subjects tested, a small percentage MUST score signficantly high scores. Just like if you flip a quarter enough times, eventually you will get 10 heads in a row. By submitting that person to another trial, you will confirm their results with much higher confidence.

It MUST be noted that this test can not be scientifcally valid oustide of a controlled setting. Cheating/tampering with the test is still possible in this 'mail it around' method. IN addition, a switching system that allows the test subject to instantly change samples is needed in order to maximize sensitivity to small differences. A lab controlled test with postivie results, would be intriguing, though.

-Chris

 

[meat01]

The problem with doing the tests on both coasts and then swapping the cables is that you wouldn't get the results for a year. As it is, this test will take a while, and by the time the results are in, people may have forgotten about it.

I think option D is the best, as I don't see any reason to close this thread.

I think this is a good test the way you have already described it, even though there may be chance involved. If I couldn't tell the difference, I would say so rather than claiming that I can tell which one is which, only to be wrong. If there truly is such a huge difference as some claim, then it should be easy for people to figure out which cable is silver.

You could A/B the cables and guess/claim which one is better, or you could have someone give you a blind test and see if you can tell which one is which (by color or symbol) 10 times out of 10.

 

[Edwood]

Uh, keep in mind that this little project is coming out of my pocket. Remember: Simple and Fun to start with.
I think six IC's (pairs) is enough for now. Three for each half of the USA. Depending on how it goes, we can expand worldwide from there. This test will take long enough for the USA as it is.

The A,B,X may be the best route to go. (I think symbols may work the best, I want to cables otherwise to look the same) THe "X" factor in this test will be the guts of the crappiest cable I can find. Or just really crappy wire, no fancy twisting/braiding or shielding. Any recommendations for the crappiest and cheapest I can use for X?

-Ed

 

[Ruahrc]

Here's a suggestion for the "X" cable? Maybe go out and buy a common or popular cable (like a plain cheap RadioShack or Monster Cable or Acoustic Research) that the average guy would most likely buy if he were to enter a technology store like BB and get. That way we could see if it is possible to pick out the average cable from the good and/or silver one?

It would really shed a lot of light on to whether or not "Generic brand X" cable you can buy at BB is really that much worse than "audiophile" cables.

Just a thought

Ruahrc

 

[Edwood]

Quote:

Originally Posted by WmAx Before you begin to ship out cables for testing, they should be shipped to a naysayer(if that's the nickname you feel is appropriate). This person will sort, seal and mark the cables. Record the cable ID data and keep data sealed until the test cycle is complete. Note: For maximally valuable data in this test, ABX cable order should be re-arranged for every test subject. Cables must be tamper proofed in some manner. But nothing is iron-clad. I'm sure their is still some way to find out materials, perhaps measuring the respective effects within a magnetic field, like a metal detector? Perhaps the cables can be encased within a copper mesh tubing, in order to make such a cheating method difficult? After test cycle is complete, results need to be observed, but not yet released. If a person scores a significantly high result, then they should be re-tested with teh cables re-arranged. This is due to statisical law. -Chris

I guess that would make me the Nay Sayer as I won't be participating in the test, as I am the one making the cables and therefore I know which one is which.

If you guys want, someone else can be the verification. (Naturally they can't be participating in the test either.) Where I would email/PM them the "secret" of which cable is which, right before I send them out. That way there is an additional witness, and you know I'm not lying about the results. When you guys PM your rankings/results, PM them to both me and the third party "Nay Sayer." (you can send the same PM's to up to five people now, by separating multiple user names with ';' )

C'mon, guys. Some level of honesty and trust will have to take place here for two reasons.

1. Someone could always find out what's inside if they really wanted to. I don't think it should be necessary to encase the cables with lead for those with access to XRay machines. (so Radiologists and Airport security are out. heheh, just kidding) If you are someone that just can't resist spoiling it for everyone, don't participate.

2. Someone decides to spoil it for everyone and just keep the cables and not pass them on. (for this, maybe it would be best to limit the test to those that have a positive feedback thread here, apologies to the newbies, this time) or it get's lost in the transit from the shipping, or they get run over by a tank.

Try not to take this "Taste Test" too seriously, guys. I just want to provide a fun and simple test for those that want a semi controlled answer to the question: "Do cables really make a difference?" Without getting into a DBT flame fest.

Who knows, maybe the results will be predictable or surprising. Either way, this is not a strict scientific test.

-Ed

 

[Edwood]

Quote:

Originally Posted by Ruahrc Here's a suggestion for the "X" cable? Maybe go out and buy a common or popular cable (like a plain cheap RadioShack or Monster Cable or Acoustic Research) that the average guy would most likely buy if he were to enter a technology store like BB and get. That way we could see if it is possible to pick out the average cable from the good and/or silver one? It would really shed a lot of light on to whether or not "Generic brand X" cable you can buy at BB is really that much worse than "audiophile" cables. Just a thought Ruahrc

Good idea.

How about the cheapest Rat Shack cables?

Like the crappy stock ones you get included with cheap "hifi" equipment.

-Ed