Triple blind test setup for comparing headphone amps

[magicscreen]

Player->USB cable->DAC-
-RCA cable1 RCA1->Amp1->headphone cable1 HC1-
-RCA cable2 RCA2->Amp2->headphone cable2 HC2-
-headphone cable3 HC3->Headphone

RCA1 and RCA2 are the same type
HC1 and HC2 are the same type extension cables

1. randomly plug one of the RCA1/2 into the DAC.
2. randomly plug the HC3 into one of the HC1/2
3. if there is no sound then plug the HC3 into the other HC1/2
4. enjoy music and try to find out which amp listening to

Single - I do not know which amp is connected
Double - no other person in the test, cannot influence me
Triple - amps cannot see me, my personality cannot influence them (joking, for experiment only )

Using pink noise and android decibel meter for volume level-match.

Is this any good?

 

[Trouble Maker]

Triple - amps cannot see me, my personality cannot influence them

Is this any good?

You need quadruple blind.
Wearing a cover so god can't see your sins while you are doing the test.

 

[Blumlein 88]

Player->USB cable->DAC-
-RCA cable1 RCA1->Amp1->headphone cable1 HC1-
-RCA cable2 RCA2->Amp2->headphone cable2 HC2-
-headphone cable3 HC3->Headphone

RCA1 and RCA2 are the same type
HC1 and HC2 are the same type extension cables

1. randomly plug one of the RCA1/2 into the DAC.
2. randomly plug the HC3 into one of the HC1/2
3. if there is no sound then plug the HC3 into the other HC1/2
4. enjoy music and try to find out which amp listening to

Single - I do not know which amp is connected
Double - no other person in the test, cannot influence me
Triple - amps cannot see me, my personality cannot influence them (joking, for experiment only )

Using pink noise and android decibel meter for volume level-match.

Is this any good?

Don't match levels by measuring the sound. Usually easier to match with a sine wave tone than noise. Best is to use a voltmeter across the output to the headphones. Set it for a comfortable level for music on one amp. Then measure a -20 db 1khz tone. Swap to the other phone amp and set it until you get the same reading on the 1 khz tone.

 

[Fluffy]

If you are serious at attempting to hearing differences between electronics, you should use an A/B switch that instantly switches one amp from another (2 inputs, 1 output). Unplugging and re-plugging RCA connectors takes way too long for the brain to remember what it heard.

 

[odyo]

In my opinion, personally i think best way to compare is use them for a long time then switch to other and use it long time too.
Use amp a for 1 month, use amp b for 1 month. Get used to them.

 

[maverickronin]

Triple blind is actually a thing, at least in large studies. It's when all the specifics of the test are hidden from whoever is compiling the statistics so that they don't know what's actually being tested.

Automated tests (i.e. foobar's ABX plugin) are already triple bind.

 

[Bob-23]

you should use an A/B switch that instantly switches one amp from another

Indeed it's necessary switching instantaneously - otherwise the comparison is of not much use - plugíng-in and -out takes too long. Optimal is a switch box with which you don't know when it is just being switched, and the switching time itself is minimal. Rod Elliott presents such an AB-Box on his ESR-site - great site, btw, it switches by relays, and even if you don''t built it yourself, it is described there how to use such a box. As blumlein pointed out it's crucial getting the exactest possible concordance ot the voltages of both amps. Even very slight differences of the output voltages - differences that are too small for being perceived as loudness differences - may nevertheless be perceived as 'more open' and 'better sounding'.

In the ESR-article Phil Allison introduces the project:
"This is probably going to be a very controversial device. Its purpose is to prove people wrong and that is very confronting. If you don't wish to have your cherished beliefs about amplifiers and audio generally challenged then do not build or use this unit."
https://sound-au.com/index.html

Projects>Miscellaneous Projects>A-B-Switch-box

 

[LTig]

In my opinion, personally i think best way to compare is use them for a long time then switch to other and use it long time too.
Use amp a for 1 month, use amp b for 1 month. Get used to them.

Nope, scientific research says otherwise.

 

[solderdude]

Not only scientic research does but I do as well

 

[Fluffy]

Optimal is a switch box with which you don't kow when it is just being switched

That set up is useful in very narrow cases. If you want to definitely prove there is no difference, you put the tester alone in the room and secretly switch between a/b to see if he can spot when was there a switch. But I think it's only valid if you play a constant signal, like pink noise, or a very short loop of sound. Music is way too variable to conclude anything about the ability of someone to blindly perceive that switch, because you can get a lot of false positives. And this test isn't useful in assessing what are the differences if there are any.

I think an ABX test where the tester can control when the switch occur is more useful to know definitively if he can distinguish between a and b. And if he can, a simple non-blind a/b switch is the method to sort out the minute differences. Any method that takes more than a second to switch between a and b is useless in all of these cases.

 

[Leporello]

I suggest a quintuple blind test: the test is run in quadruple manner, but - in addition - the results will not be revealed to anyone, ever.

 

[Bob-23]

If you want to definitely prove there is no difference, you put the tester alone in the room and secretly switch between a/b to see if he can spot when was there a switch.

Agree: That's what I meant by 'you don't know' - that is optimal. Second best method is - which I often practise when there's nobody around I could engage: I myself switch, and, of course, then know that I switch, but as the switching time is so short - neatless -, I usually then get quite clearly if there's a difference, or not. Obfuscating (before yourself) which position of the toggle switch (in case you have a such one) switches which amp, optimizes that method.

 

[magicscreen]

I have tried the (cheap) HC1 extension cable in the chain and it was inducing some sound quality loss.
The length of the cables all together was 4 meter. Is it too much?
First of all I have to blind-test the cables. But I have no idea how can I do that? Without spending a lot of money.

 

[Music1969]

The problem with an A/B switch though is you need to use an objectively transparent one.

Otherwise it may become the weakest link in the chain and this MAY be the reason both things sound similar/same...

 

[Fluffy]

The problem with an A/B switch though is you need to use an objectively transparent one.

Otherwise it may become the weakest link in the chain and this MAY be the reason both things sound similar/same...

It's not really a problem. Electrically, a passive switch is no different than a simple cable that you can plug and unplug instantaneously. It's just wire and contacts, nothing there to ruin the sound. Here is a completely transparent one, for example:

https://www.audiosciencereview.com/...und-3-in-1-out-xlr-audio-switch-review.11062/

https://www.audiosciencereview.com/...eardown-of-douk-audio-mc103-pro-switch.11020/

 

[A800]

Contact resistance.

 

[Music1969]

It's not really a problem. Electrically, a passive switch is no different than a simple cable that you can plug and unplug instantaneously. It's just wire and contacts, nothing there to ruin the sound. Here is a completely transparent one, for example:

https://www.audiosciencereview.com/...und-3-in-1-out-xlr-audio-switch-review.11062/

https://www.audiosciencereview.com/...eardown-of-douk-audio-mc103-pro-switch.11020/

That nobsound one is definitely not a problem.

 

[solderdude]

The problem with an A/B switch though is you need to use an objectively transparent one.

Otherwise it may become the weakest link in the chain and this MAY be the reason both things sound similar/same...

So... for instance if someone built 2 quite different amplifiers but used the same switch or relay at the output this could make them measure (and sound) the same ?

That would be a neat trick.

 

[Music1969]

So... for instance if someone built 2 quite different amplifiers but used the same switch or relay at the output this could make them measure (and sound) the same ?

That would be a neat trick.

Certainly not a problem with the nobsound linked above (obviously).

But for an extreme example:

If amp A had SINAD 70dB,

Amp B had SINAD 120dB

And both went through a switch with SINAD 60dB, will this give amp B a fair shot ?

 

[A800]

Certainly not a problem with the nobsound linked above (obviously).

But for an extreme example:

If amp A had SINAD 70dB,

Amp B had SINAD 120dB

And both went through a switch with SINAD 60dB, will this give amp B a fair shot ?

60dB for both.
Looks fair to me.