Skip to main content
Topic: Software for ABX testing of output devices (Read 6761 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Software for ABX testing of output devices

Hi all,

I've been in contact with the developer of AudioSwitcher (http://audioswit.ch/er), Sean Chapman, about possibly adding a random switching capability to facilitate ABX testing of output devices (onboard, external DAC &c). He responded positively and is thinking of creating a separate application specifically for this purpose. Is there are any interest in such a program? If so, do you have any concerns? I've alerted him to the existence of this thread, and he should be popping by at some pt. Looking forward to your replies!

Software for ABX testing of output devices

Reply #1
Sounds interesting! Would be really cool to know if I actually hear any difference between headset directly to the laptop, or headset through my DAC setup
How would hardware setup be? A Y jack splitter, one female jack, two males both connected to a output source. Then the software would change between the output devices?

Software for ABX testing of output devices

Reply #2
[quote author=o-l-a-v link=msg=874319 date=1410342347]How would hardware setup be? A Y jack splitter, one female jack, two males both connected to a output source. Then the software would change between the output devices?[/quote]
That's more or less it in a nutshell.

Considerations:
- The source audio file should be of high (pref lossless) quality, i.e Wave or Flac.
- Source audio should be controlled directly through the application where appropriate
- Switching between devices should be consistent as to not give hints to the listener
- Device A and B are able to be selected as required
- Multiple A/B switches should be allowed, before testing X
- After testing X, require the user to choose either A or B to produce a test result

Optional:
- Repeatable tests
- Stats/Graphs?

Software for ABX testing of output devices

Reply #3
[quote author=o-l-a-v link=msg=874319 date=1410342347]Sounds interesting! Would be really cool to know if I actually hear any difference between headset directly to the laptop, or headset through my DAC setup
How would hardware setup be? A Y jack splitter, one female jack, two males both connected to a output source. Then the software would change between the output devices?[/quote]

In practice, a Y-splitter wouldn't be enough - there will be significant (EM?) interference from the mere presence of the other DAC. I have found that this is true even with a simple A/B switch box. With the Phono-Preamps TC-716 switcher, you can have more than one selected, but it attenuates all of them significantly when you do that.  So the only real solution is a cheapish mixer with several un-balanced inputs.

Software for ABX testing of output devices

Reply #4
Hi all,

I've been in contact with the developer of AudioSwitcher (http://audioswit.ch/er), Sean Chapman, about possibly adding a random switching capability to facilitate ABX testing of output devices (onboard, external DAC &c). He responded positively and is thinking of creating a separate application specifically for this purpose. Is there are any interest in such a program? If so, do you have any concerns? I've alerted him to the existence of this thread, and he should be popping by at some pt. Looking forward to your replies!



I've always seen a device like this:

http://www.ebay.com/itm/SainSmart-8-channe...=item3f1f4d42cc



slaved to one of the software ABX comparators (which already have all the logic in place) as a better solution.

Software for ABX testing of output devices

Reply #5
I've always seen a device like this:

http://www.ebay.com/itm/SainSmart-8-channe...=item3f1f4d42cc

slaved to one of the software ABX comparators (which already have all the logic in place) as a better solution.


I know it's been quite a while, but I never turned on notifications for this thread and just stumbled upon it again. I'm not quite sure what that does, can you enlighten me? Thanks.

Software for ABX testing of output devices

Reply #6
I've always seen a device like this:

http://www.ebay.com/itm/SainSmart-8-channe...=item3f1f4d42cc

slaved to one of the software ABX comparators (which already have all the logic in place) as a better solution.


I know it's been quite a while, but I never turned on notifications for this thread and just stumbled upon it again. I'm not quite sure what that does, can you enlighten me? Thanks.


Uses electromagnets to move which piece of wire is connected to which input.

Software for ABX testing of output devices

Reply #7
I've always seen a device like this:

http://www.ebay.com/itm/SainSmart-8-channe...=item3f1f4d42cc

slaved to one of the software ABX comparators (which already have all the logic in place) as a better solution.


But Arny... won't those cheap Japanese copper relay contacts cause intermodulation and veiling of the audio signal?

      Sorry, I've been dealing with science deniers all day today and my sarcasm is showing.

Software for ABX testing of output devices

Reply #8
I've always seen a device like this:

http://www.ebay.com/itm/SainSmart-8-channe...=item3f1f4d42cc

slaved to one of the software ABX comparators (which already have all the logic in place) as a better solution.


But Arny... won't those cheap Japanese copper relay contacts cause intermodulation and veiling of the audio signal?

      Sorry, I've been dealing with science deniers all day today and my sarcasm is showing.


Agreed that relay contacts are very common in the audio world.  Most audio gear with power amps and any pretension to quality uses relays to disconnect and thus protect the speakers when one of several common kinds of faults are detected. It's that click that one hears shortly after power is turned on.  Analog audio consoles, especially the elaborate high quality ones, also use relays to route audio signals.

Audiophiles often have no idea about how pervasive relays are in the audio world, and therefore they can be mislead into thinking that they have some kind of inherent audible problems.

I'm sure that you are aware that due the corrosion problem, relay contacts are rarely if ever copper.

Software for ABX testing of output devices

Reply #9
Yessir, I am.  I used to design radar circuitry and had to work with high power relays occasionally.  It just amuses me that the subjectivists overthink things like contacts but voluntarily introduce huge amounts of distortion via single ended tube stages and such. 

Back on topic, I'm actually excited about getting one of these; it would be very handy.  Looks like you use it with an Arduino board?

Software for ABX testing of output devices

Reply #10
Yessir, I am.  I used to design radar circuitry and had to work with high power relays occasionally.  It just amuses me that the subjectivists overthink things like contacts but voluntarily introduce huge amounts of distortion via single ended tube stages and such. 

Back on topic, I'm actually excited about getting one of these; it would be very handy.  Looks like you use it with an Arduino board?



Sadly, there is no latter day hardware ABX box.  I identified the 8-relay USB board as a pregnant looking hardware switching core, bought one, and the project languishes.  It seems like mating the relay board with an Arduino board could work out.



Software for ABX testing of output devices

Reply #13
Would that tool be good enough to conduct professional tests?


Software for ABX testing of output devices

Reply #14
Would that tool be good enough to conduct professional tests?

How would you define "professional tests"?

Software for ABX testing of output devices

Reply #15
Would that tool be good enough to conduct professional tests?

How would you define "professional tests"?



Good point. The other question is exactly what is the tool itself?  The spec sheet is hyper terse and the published reviews are not very clear to me.

I still think that the best way to test hardware these days is to record its output under test with some really good converters and go the software ABX route. That way there can be  many more test participants. 

Of course this is not very satisfying to people who believe that converters come in the sonic flavor of the month. Many people know that is a long-settled issue, but may others are poorly informed.



Software for ABX testing of output devices

Reply #17
Would that tool be good enough to conduct professional tests?

How would you define "professional tests"?


For use in a professional organisation. Where credibility was paramount.



IME professional organizations were always the easiest to please. It is the audiophile and audiophile publicaitons that make the biggest demands on credibility.

Notice that the AES leadership recently gave high praise to a paper describing DBTs that were based on a Matlab simulation, not even a real world piece of equipment. Yet the test results were claimed to be relevant to typical audio gear.

What's the chances of that flying in audiophile land?

Software for ABX testing of output devices

Reply #18
Ahhh! Replies.

Is this still of any use? I'm not sure how useful a software solution is without the hardware counterpart.

It may make the hardware requirements for ABX testing slightly easier to come by; just a mixer.

Software for ABX testing of output devices

Reply #19
Is this still of any use? I'm not sure how useful a software solution is without the hardware counterpart.


The hardware solution is strictly for those who don't believe that ADCs and DACs can be sonically transparent.

Recording the output of the units under test (UUT) with a good ADC, and using a good DAC for monitoring during the ABX (or other software-driven DBT) works exceedingly well and avoids a lot of complexity.


Quote
It may make the hardware requirements for ABX testing slightly easier to come by; just a mixer.


IME professional audio mixers are neither needed nor as reliably as sonically transparent as good  Digital <-> Analog converters.  The only mixing consoles I really trust are digital.

 
SimplePortal 1.0.0 RC1 © 2008-2019