Windows resampling not actually that bad?

[T.J. Turner]

So was there any conclusion here?

 

[A800]

Any resampling is bad.

 

[whazzup]

Okay here is MPC-HC without using exclusive mode, playing a 76543Hz sweep file at 16 and 24-bit, recorded at 24-bit 48kHz using digital loopback. Left hand side is 100% volume and right hand side is 95% volume.

In terms of frequency...
View attachment 37526

and amplitude.
View attachment 37525

As you can see the test file's right channel is at 50% volume (-6dB) but it still distorts when using 100% playback volume. Just slightly reduce the built-in volume control in MPC-HC eliminates these artifacts and you can see how clean the Windows SRC is. When playing the 16-bit test file the noise floor is waaaaay higher than everything else. That means if Windows SRC is really that bad then everyone should delete all of their 16-bit files and sell all of their CDs.

This is a very interesting result. Why does reducing to 95% volume magically cleaned up 'audio artifacts' (not sure if that's the right term) in Windows? Windows 10 has similar issues?

 

[bennetng]

This is a very interesting result. Why does reducing to 95% volume magically cleaned up 'audio artifacts' (not sure if that's the right term) in Windows? Windows 10 has similar issues?

Explained in these posts:
https://www.audiosciencereview.com/...mpling-not-actually-that-bad.9092/post-253946
https://www.audiosciencereview.com/...ng-of-digital-volume-levels.12458/post-369292

 

[whazzup]

Explained in these posts:
https://www.audiosciencereview.com/...mpling-not-actually-that-bad.9092/post-253946
https://www.audiosciencereview.com/...ng-of-digital-volume-levels.12458/post-369292

Fantastic, thanks! Will go read up.

 

[anphex]

Digging this thread out again.

So is the consensus that Windows Resampling isn't bad and the culprit is the CAudioLimiter which you can disable by putting all software outputs to 99 % volume?
What about recent Windows 11 test?

I have some gear lying around, if someone could point me to the software and methodology to test this I'll do this today.

 

[Vincent Kars]

Are you familiar with this one: https://www.audiosciencereview.com/...nding-the-windows-audio-quality-debate.19438/ ?
To test is, you need a good ADC: http://archimago.blogspot.com/2015/11/measurements-windows-10-audio-stack.html

 

[anphex]

Are you familiar with this one: https://www.audiosciencereview.com/...nding-the-windows-audio-quality-debate.19438/ ?
To test is, you need a good ADC: http://archimago.blogspot.com/2015/11/measurements-windows-10-audio-stack.html

Yes! I am just not completely sure if Windows Resampling is really that bad or it's the CAudioLimiter. I vouch for the latter since even with the most archaic byte dropping or repeating the worst that could happen are minor rounding issues that produce the expected really low level distortions.

 

[Tangband]

Digging this thread out again.

So is the consensus that Windows Resampling isn't bad and the culprit is the CAudioLimiter which you can disable by putting all software outputs to 99 % volume?
What about recent Windows 11 test?

I have some gear lying around, if someone could point me to the software and methodology to test this I'll do this today.

Just remember that every upgrade in windows update can potentially change your setting , both for recording and playback , so this is why proffesional soundtechnicians often use Mac when doing recordings in concert halls.
With every update for Windows, you have to go trough all settings so they are correct. This is very uncreative .

The reason for Mac is also that when recording a whole orchestra , you cant afford any technical faults and the musicians gonna hate you if there is such problems.

 

[anphex]

Windows 11 lately only lets me switch my AVR output to surround when I use the old sound control panel instead of the new pretty one.
Also, the settings keep resetting to default sometime for no effing reason. So yeah, I totally agree with you. If Windows wasn't the top player for gaming I'd be a Ubuntu user all the way.

 

[valerianf]

As source sampling can be 44.1 khz, 48 khz or 96 khz, I chose to force the W10 digital output to 96 khz before going to the amp (spdif).
I did not noticed any side effect.
The measurement from @Degru seems to confirm it.