Jump to content
IGNORED

Blue or red pill?


Recommended Posts

12 minutes ago, manisandher said:

No, I'm not looking to satisfy random requests. But if someone thinks that a specific set of tests would help them understand what's going on, I'm happy to do this

 

Indeed there is nothing wrong with this and only very nice of you.

 

13 minutes ago, manisandher said:

What's the downside, other than the time it takes me to conduct them, and the time it takes the requester to analyse the captures?

 

That without notice you fall into more and more pitfalls, want to climb out but fall deeper again. So now your analysis software ain't right. Next attempt ? *IS* it wrong ?

So it goes deeper and deeper while the actual work has been done. OK, that's what I think (and you too) but this does not mean (all) others are satisfied with it. So let's say it is justified to require more. But what would be useful in this regard ?

 

Increase the 10 tests to 30 ? Nah

 

Have better test signals ? maybe. But not a random (sorry) general test set for testing DACs etc. but more to the point perhaps. Could be difficult. In the Lush thread a "field" of IMD tones was suggested. So a set of 20 or so throughout the audible spectrum. Observe those with Mans's expertise (he could try to smoothen those signals themselves to dig out differences). Btw, it might not be easy to make them (read : it is relatively easy to let such test signals fail as it requires software which makes them right in a proven fashion (and it is not quite common, I think)). I once tried to create a test quickly (!) myself but failed.

 

Peter

 

 

Lush^3-e      Lush^2      Blaxius^2.5      Ethernet^3     HDMI^2     XLR^2

XXHighEnd (developer)

Phasure NOS1 24/768 Async USB DAC (manufacturer)

Phasure Mach III Audio PC with Linear PSU (manufacturer)

Orelino & Orelo MKII Speakers (designer/supplier)

Link to comment
50 minutes ago, manisandher said:

I'm confused... perhaps someone could help me?

 

We put the following 10kHz test tone into the Altmann DAC:

5ad462435c6d2_1.10kHz_24_44.1_testtone_non-averaged.thumb.JPG.0170c5be1725a818e9df21e78a2a022d.JPG

 

The low-level spikes in the signal lie at around -165dB.

 

And here's the output from the DAC:

5ad46266708dd_1.Altmann-10k_24_44.1-Customfilter.thumb.JPG.fdfd783e882bda909b65285ea6506fdb.JPG

 

Yes, the above is averaged, but the spikes are visible even in the non-averaged output:

5ad4630d113a7_1.Altmann-10k_24_44.1-Customfilter_non-averaged.thumb.JPG.914bcf65778bd32b98f425c5954b1d03.JPG

 

 

Now, how can a DAC using a 25-year-old 16-bit chip possibly resolve these spikes?

 

I'd love to know...

Those spikes in the test file are caused by rounding errors in the LSB of the 24-bit signal. If the DAC simply truncates the input to 16 bits, this will create bigger rounding errors resulting in the measured spikes.

Link to comment
12 minutes ago, adamdea said:

The other possibility which had occurred to me would be to have two live streams which could be switched between -either two computers each running a version of the software or possibly two versions of the software running at the same time on the same machine.

 

Adam, Neither would work because it requires all double, up to the DAC. So technically two streams (programs) can not output to the same DAC at the same time. Of course this is solved by pausing playback and switch, but this actually already happens (happened).

 

I need to say it again (and not addressing anyone in particular) : instead of putting Mani (and Mans) now to the burden of re-doing things in any more satisfactory way than currently has been done, why don't you try it yourself. I really don't see why this is not happening unless you in the end trust Mani's ears better than your own. Now that would be strange, right ?

 

19 minutes ago, adamdea said:

but one possibility would be to see whether the outputs can be tested as files in the foobar abx comparator. This could be done from recordings

1) of the spdif output of the computer

2) of the analogue output of the dac.

 

This was not what I meant in my text above, but that too should work. But keep in mind : never in the same fashion because of the "double effect" at play. So mind you, any player you use already implies the effect as such but in its one-setting way (think over why Foobar sounds different from JRiver etc.). Still, underlaying would be the recorded effect between the two different settings. So it is two layers now, the layer of your player of choice being the common denominator. And please, if that is too (literally) bad, it will mask what happened at Mani's. Keep in mind the diff files I showed myself a couple of weeks back (which were also from Foobar against XXHighEnd). So if the implied difference = in the end a literal noise is too large, the minimalistic differences will be masked. So it doesn't work by guarantee.

Downloading XXHighEnd and do nothing else but changing the SFS from 0.1 to 200 should always work, already because it has been proven for so many years that it works (out). The demo version can be used for it and is free. But my advice could really be to try Mani's recordings. It would be more in your own environment. Right ?

 

 

Lush^3-e      Lush^2      Blaxius^2.5      Ethernet^3     HDMI^2     XLR^2

XXHighEnd (developer)

Phasure NOS1 24/768 Async USB DAC (manufacturer)

Phasure Mach III Audio PC with Linear PSU (manufacturer)

Orelino & Orelo MKII Speakers (designer/supplier)

Link to comment
2 minutes ago, PeterSt said:

 

Adam, Neither would work because it requires all double, up to the DAC. So technically two streams (programs) can not output to the same DAC at the same time.

1) I have a dac with two inputs and I bet most people do too.

2) I believe s/pdif switching devices exist.

 

 

You are not a sound quality measurement device

Link to comment
14 minutes ago, PeterSt said:

This was not what I meant in my text above, but that too should work.

Do I understand you correctly that you consider that a recording of the analogue out of the dac should be sufficiently sensitive to reveal whatever difference the software settings make (although it could conceivably be masked by playback software and dac)

You are not a sound quality measurement device

Link to comment
42 minutes ago, mansr said:

Those spikes in the test file are caused by rounding errors in the LSB of the 24-bit signal. If the DAC simply truncates the input to 16 bits, this will create bigger rounding errors resulting in the measured spikes.

 

What if Mani were to ABX with 16-bit tracks?

Would this remove the spikes?

And do the spikes differ in the two software settings?

"Science draws the wave, poetry fills it with water" Teixeira de Pascoaes

 

HQPlayer Desktop / Mac mini → Intona 7054 → RME ADI-2 DAC FS (DSD256)

Link to comment

I've just done a quick analysis of the two ABXXXXXXXXXX tests:

 

5ad482aee2c16_1stABXXXXXXXXXXResults.thumb.JPG.b89ccc0dbdbff174d83797e6acfe710d.JPG

 

5ad487c46e5eb_2ndABXXXXXXXXXXResults.thumb.JPG.576b15b11f1896f54e04ae1be704a001.JPG

 

So my scores for the four different transitions in the two ABXXXXXXXXXX tests were:

 

AA - 0/2

AB - 2/5

BA - 4/7

BB - 2/6

 

My scores for the two different transitions in the A/B/X were:

 

BA - 4/4

BB - 5/6

 

And although this has been said on numerous occasions, but some just don't care to acknowledge, the 54 digital captures taken during all 3 tests were all bit-identical.

 

I hope this finally puts to rest that I scored 9/10 in the A/B/X because there might have been some audible differences in the transitions between BA and BB... that weren't present in the ABXXXXXXXXXX but magically appeared in the A/B/X!

 

If people still don't believe the results of the A/B/X, that's their problem. I won't be pursuing this any further.

 

Mani.

Main: SOtM sMS-200 -> Okto dac8PRO -> 6x Neurochrome 286 mono amps -> Tune Audio Anima horns + 2x Rotel RB-1590 amps -> 4 subs

Home Office: SOtM sMS-200 -> MOTU UltraLite-mk5 -> 6x Neurochrome 286 mono amps -> Impulse H2 speakers

Vinyl: Technics SP10 / London (Decca) Reference -> Trafomatic Luna -> RME ADI-2 Pro

Link to comment
33 minutes ago, adamdea said:

Do I understand you correctly that you consider that a recording of the analogue out of the dac should be sufficiently sensitive to reveal whatever difference the software settings make (although it could conceivably be masked by playback software and dac)

 

Adam, maybe it is not clear to me what you're asking. The first part of your sentence has been done by Mani and Mans. The part between braces is probably my lacking English.

 

Anyway what I suggest is listen for yourself with or the two recordings of analogue out which are available (can be made available by Mani/Mans) - and which was your own idea I think, or use XXHighEnd yourself with your own DAC to receive the result in direct fashion (and if you like to ABX it, you may get your wife involved - but I'd say, start with being placeboed to begin with, why not)

 

It is not more difficult than this, really.

Lush^3-e      Lush^2      Blaxius^2.5      Ethernet^3     HDMI^2     XLR^2

XXHighEnd (developer)

Phasure NOS1 24/768 Async USB DAC (manufacturer)

Phasure Mach III Audio PC with Linear PSU (manufacturer)

Orelino & Orelo MKII Speakers (designer/supplier)

Link to comment
6 minutes ago, PeterSt said:

Anyway what I suggest is listen for yourself with or the two recordings of analogue out which are available...

 

I think the question is whether these would have captured the audible differences I heard.

 

Edit: I'll post the files in the next hour...

 

Mani.

Main: SOtM sMS-200 -> Okto dac8PRO -> 6x Neurochrome 286 mono amps -> Tune Audio Anima horns + 2x Rotel RB-1590 amps -> 4 subs

Home Office: SOtM sMS-200 -> MOTU UltraLite-mk5 -> 6x Neurochrome 286 mono amps -> Impulse H2 speakers

Vinyl: Technics SP10 / London (Decca) Reference -> Trafomatic Luna -> RME ADI-2 Pro

Link to comment
3 minutes ago, manisandher said:

 

I hope this finally puts to rest that I scored 9/10 in the A/B/X because there might have been some audible differences in the transitions between BA and BB... that weren't present in the ABXXXXXXXXXX but magically appeared in the A/B/X!

 

If people still don't believe the results of the A/B/X, that's their problem. I won't be pursuing this any further.

 

Mani.

I'm afraid I just can't follow your reasoning here at all, either on what the analysis of the ABXXXX shows or on what "believing the results" means. The ABXXXXXX ..s seem to me to have been an impossible task and I would be cautious about drawing any conclusions from them.

I not disbelieving any results, only the analysis of what they mean. People do that with experiments. 

You are not a sound quality measurement device

Link to comment
2 minutes ago, manisandher said:

 

I think the question is whether these would have captured the audible differences I heard.

 

Edit: I'll post the files in the next hour...

 

Mani.

Thanks this would be helpful

You are not a sound quality measurement device

Link to comment
14 minutes ago, PeterSt said:

 

Adam, maybe it is not clear to me what you're asking. The first part of your sentence has been done by Mani and Mans. The part between braces is probably my lacking English.

 

Anyway what I suggest is listen for yourself with or the two recordings of analogue out which are available (can be made available by Mani/Mans) - and which was your own idea I think, or use XXHighEnd yourself with your own DAC to receive the result in direct fashion (and if you like to ABX it, you may get your wife involved - but I'd say, start with being placeboed to begin with, why not)

 

It is not more difficult than this, really.

Foobar ABX comparator seems like the best bet. 

You are not a sound quality measurement device

Link to comment
16 minutes ago, adamdea said:

The ABXXXXXX ..s seem to me to have been an impossible task...

 

Agreed.

 

16 minutes ago, adamdea said:

... and I would be cautious about drawing any conclusions from them.

 

The conclusion that there were no 'tells' in the A/B/X can be drawn from the analysis of BA and BB transitions in the ABXXXXXXXXXX.

 

16 minutes ago, adamdea said:

I'm afraid I just can't follow your reasoning here at all...

 

No problem.

 

Mani.

Main: SOtM sMS-200 -> Okto dac8PRO -> 6x Neurochrome 286 mono amps -> Tune Audio Anima horns + 2x Rotel RB-1590 amps -> 4 subs

Home Office: SOtM sMS-200 -> MOTU UltraLite-mk5 -> 6x Neurochrome 286 mono amps -> Impulse H2 speakers

Vinyl: Technics SP10 / London (Decca) Reference -> Trafomatic Luna -> RME ADI-2 Pro

Link to comment
1 hour ago, mansr said:

Those spikes in the test file are caused by rounding errors in the LSB of the 24-bit signal. If the DAC simply truncates the input to 16 bits, this will create bigger rounding errors resulting in the measured spikes.

 

That makes sense. Did the Altmann DAC have dither and was that enabled for the captures?

 

Link to comment
35 minutes ago, manisandher said:

If people still don't believe the results of the A/B/X, that's their problem. I won't be pursuing this any further.

Mani

 No matter what you do, some people will NEVER accept results that don't agree with currently accepted theory.

They will always try to find fault with the methodology used .

 Given another couple of months to theorise over it, for many it will just be like this thread and it's results never happened !:o

Alex

 

How a Digital Audio file sounds, or a Digital Video file looks, is governed to a large extent by the Power Supply area. All that Identical Checksums gives is the possibility of REGENERATING the file to close to that of the original file.

PROFILE UPDATED 13-11-2020

Link to comment
7 minutes ago, sandyk said:

... some people will NEVER accept results...

 

I'm hopeful that everyone now agrees that I heard clear differences between two bit-identical streams. No 'tells' involved, just a difference in sound between them.

 

7 minutes ago, sandyk said:

... that don't agree with currently accepted theory.

 

That's a different matter :)

 

Mani.

Main: SOtM sMS-200 -> Okto dac8PRO -> 6x Neurochrome 286 mono amps -> Tune Audio Anima horns + 2x Rotel RB-1590 amps -> 4 subs

Home Office: SOtM sMS-200 -> MOTU UltraLite-mk5 -> 6x Neurochrome 286 mono amps -> Impulse H2 speakers

Vinyl: Technics SP10 / London (Decca) Reference -> Trafomatic Luna -> RME ADI-2 Pro

Link to comment
1 hour ago, manisandher said:

I've just done a quick analysis of the two ABXXXXXXXXXX tests:

 

45 minutes ago, adamdea said:

I'm afraid I just can't follow your reasoning here at all, either on what the analysis of the ABXXXX shows or on what "believing the results" means.

 

Mani, maybe I am not talking about the same Adam talks about, but I (too) did not understand much of what your tables tried to tell us.

I also don't understand why the ABXXXXX keeps on showing up. It is a nice subject in itself but unrelated to what it in the end is about : two bit identical situations showed a difference in the analogue (after D/A) result which you heard and which can be measured.

And the "which you heard" is the most weak thing because without exception everybody can hear it, if they only use software to show it. Also see upcoming post ...

Lush^3-e      Lush^2      Blaxius^2.5      Ethernet^3     HDMI^2     XLR^2

XXHighEnd (developer)

Phasure NOS1 24/768 Async USB DAC (manufacturer)

Phasure Mach III Audio PC with Linear PSU (manufacturer)

Orelino & Orelo MKII Speakers (designer/supplier)

Link to comment
3 minutes ago, PeterSt said:

I also don't understand why the ABXXXXX keeps on showing up.

 

It makes sense to me :)

 

Some people may think that there was a 'tell' during the A/B/X. i.e. there was something (not related to the sound per se) that was helping me (consciously or unconsciously) identify whether the X was A or B. In the A/B/X, there were only two transitions after the initial AB, i.e. BA or BB. Let's imagine that BA had a slight, but audible 'tell' such, as a slight glitch (from the transition from B back to A). OK, first, this should have shown up on the digital captures. After pointing this out many times, it seemed that not everyone was convinced (amazingly!). So I went back to the ABXXXXXXXXXX to see if I did better in BA transitions vs BB transitions there. And the answer is an emphatic "no".

 

Ergo, there were no tells in the A/B/X, because there were none in the ABXXXXXXXXXX.

 

If it still doesn't make sense, no problem.

 

Mani.

Main: SOtM sMS-200 -> Okto dac8PRO -> 6x Neurochrome 286 mono amps -> Tune Audio Anima horns + 2x Rotel RB-1590 amps -> 4 subs

Home Office: SOtM sMS-200 -> MOTU UltraLite-mk5 -> 6x Neurochrome 286 mono amps -> Impulse H2 speakers

Vinyl: Technics SP10 / London (Decca) Reference -> Trafomatic Luna -> RME ADI-2 Pro

Link to comment
4 minutes ago, manisandher said:

 

I'm hopeful that everyone now agrees that I heard clear differences between two bit-identical streams. No 'tells' involved, just a difference in sound between them.

 

Mani.

Fraid not. Sorry. I'm afraid I don;t think your reasoning from the ABXXXX transitions bears the conclusion you wish to draw given the change in activity (and ignores learning). After all why do you think the test in the first two tasks was impossible. I think it's becasue the X is being compared with a long distant audio memory whereas in the ABX is it with the last thing played. And that focusses the attention on what just happened. I accept that a switching tell could in principle still operate with the ABXXXX, but I can't see any reason to infer from it not applying in the first case that it could not work in the second.   

 

Even then, I just can't imagine why you think that no one would wonder about what might be being detected. This stuff happens. Surprising experimental results happen. 

You are not a sound quality measurement device

Link to comment
Just now, adamdea said:

Fraid not.

 

Believe what you want...

 

Mani.

Main: SOtM sMS-200 -> Okto dac8PRO -> 6x Neurochrome 286 mono amps -> Tune Audio Anima horns + 2x Rotel RB-1590 amps -> 4 subs

Home Office: SOtM sMS-200 -> MOTU UltraLite-mk5 -> 6x Neurochrome 286 mono amps -> Impulse H2 speakers

Vinyl: Technics SP10 / London (Decca) Reference -> Trafomatic Luna -> RME ADI-2 Pro

Link to comment

For anyone who's interested in doing some analysis of the digital (taken in real-time during the A/B/X) and analogue (taken immediately after A/B/X) captures, in the following link you'll find 4 files:

 

1. digital capture _ A

2. digital capture _ B

3. analogue capture _ A

4. analogue capture _ B

 

https://drive.google.com/open?id=10SD3SSdXV-P0ZFEte31NSFXeO0UhBYBi

 

Enjoy!

 

Mani.

Main: SOtM sMS-200 -> Okto dac8PRO -> 6x Neurochrome 286 mono amps -> Tune Audio Anima horns + 2x Rotel RB-1590 amps -> 4 subs

Home Office: SOtM sMS-200 -> MOTU UltraLite-mk5 -> 6x Neurochrome 286 mono amps -> Impulse H2 speakers

Vinyl: Technics SP10 / London (Decca) Reference -> Trafomatic Luna -> RME ADI-2 Pro

Link to comment
4 minutes ago, adamdea said:

Surprising experimental results happen. 

 

The only surprise to any XXHighEnd user is why I didn't get 10/10 in the A/B/X!

 

Take Peter's advice. Download a free trial of XXHighEnd, and take a listen for yourself. Or, if you can't be bothered with that, take a listen to the analogue captures I've just ULed. (Personally, I'd recommend the former, because I'm not sure how clear the audible differences will be in the analogue captures.)

 

(And of course, this has nothing to do with XXHighEnd - it's just the tool we happened to have used. By all means delete it from your HD once you've tried listening for differences in bit-identical playback with it.)

 

Mani.

Main: SOtM sMS-200 -> Okto dac8PRO -> 6x Neurochrome 286 mono amps -> Tune Audio Anima horns + 2x Rotel RB-1590 amps -> 4 subs

Home Office: SOtM sMS-200 -> MOTU UltraLite-mk5 -> 6x Neurochrome 286 mono amps -> Impulse H2 speakers

Vinyl: Technics SP10 / London (Decca) Reference -> Trafomatic Luna -> RME ADI-2 Pro

Link to comment
3 minutes ago, manisandher said:

 

The only surprise to any XXHighEnd user is why I didn't get 10/10 in the A/B/X!

 

Take Peter's advice. Download a free trial of XXHighEnd, and take a listen for yourself. Or, if you can't be bothered with that, take a listen to the analogue captures I've just ULed. (Personally, I'd recommend the former, because I'm not sure how clear the audible differences will be in the analogue captures.)

 

Mani.

Already downloaded the analogue captures and loaded into foobar ABx  (am listening on Benchmark dac2 HGC and AKG 701s, not tried formal test yet.).

I would recommend anyone else to try. I will be interested to know whether anyone can ABx them. I'm more interested in other people's results than my own.

You are not a sound quality measurement device

Link to comment

Yup I can get 10/10 (or for that matter 1000/1000) by setting the test to start at 19.7s

But of course I'm playing. I'll try doing it properly with it the test finishing at 19 seconds.

 

foo_abx 2.0.2 report
foobar2000 v1.3.3
2018-04-16 14:10:42

File A: 3. analogue capture _ A.wav
SHA1: 9c7dbfe1ac5f00c6f3d3acd73c3d6a3fd1eaa59d
File B: 4. analogue capture _ B.wav
SHA1: 6c1ec457e27079d3e864c0396851e91185fcc982

Output:
DS : Primary Sound Driver
Crossfading: NO

14:10:42 : Test started.
14:10:56 : 01/01
14:11:03 : 02/02
14:11:12 : 03/03
14:11:22 : 04/04
14:11:29 : 05/05
14:11:34 : 06/06
14:11:38 : 07/07
14:11:42 : 08/08
14:11:48 : 09/09
14:11:52 : 10/10
14:11:52 : Test finished.

 ---------- 
Total: 10/10
Probability that you were guessing: 0.1%

 -- signature -- 
b69cab17981073e545c7f532dfd3e2d3349a61dd

You are not a sound quality measurement device

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...