View Single Post
Old Aug 20, 2021, 02:56 AM   #14
drumphil
Radeon Volcanic Islands
 
Join Date: Mar 2003
Location: Australia Australia
Posts: 3,700
drumphil can beat 'Minesweeper' on any difficultydrumphil can beat 'Minesweeper' on any difficulty


Default

In that case you don't even have the kind of problem that fidelizer could fix, if it could do what it said it can, which it can't.

What can playing around with process priorities possibly do that can possibly make a difference to the output of the DAC on the sound card? The answer is that it can't do anything to make any difference to that.

If you really want to be sure download a copy of Right Mark Audio Analyzer.

connect the output of your sound card back to the input on your sound card, and use RMAA to do a loopback test.

It plays a set of test signals, records them back in though the input, and measures exactly how degraded the result is compared to the original test signal.

Do it with and without fidelizer. If you get the exact same results, then fidelizer is doing nothing.

I noticed the creator of fidelizer takes issue with this, and uses diffmaker to try and prove his point, but he uses VB-Audio Virtual Cable as his digital loop back, and it isn't a bit perfect loopback! He says he's using it to make sure no hardware errors are involved, but it doesn't claim to be bit perfect anyway. I can't follow exactly what his methodology is anyway, and usually if I can't make sense of it, it's because some of it doesn't make sense.

If I send audio to the digital outputs on an audio device I own, and I don't get a bit perfect bitstream, I declare the device broken, and use something else that isn't broken. Even if I'm running a massive project with multiple video files and a hundred audio tracks, putting large load on the CPU and graphics card, a non broken sound device should still get the exact same sequence of 1's and 0's as what the DAW software sent it.


I think it helps if you understand exactly how audio is passed from a bit of software to the sound card. The playback software sends chunks of audio data, which is ultimately a sequence of 1's and 0's. Assuming you're not running through a resampling process in shared mode, and the drivers for the audio device aren't broken, and that the volume is set to 0db, that exact sequence of 0's and 1's is what the sound card gets. Computers don't mix up the occasional 1 or 0 when they are under high load. If they do, then they are broken somehow. This is actually how CPU stress test programs work. They perform a series of calculations over and over, and check the results from the CPU against the correct result for that calculation that is already known. If the CPU ever comes back with an answer that doesn't match the known correct answer, then it declares an error has been made. This is usually used to check for faulty hardware, or to verify that an overclocked system hasn't been pushed too far. If the computer is broken enough to mix up the 1's and 0's sometimes under high load, then the answer isn't to play with process priorities, but to fix whatever is broken in the hardware.
__________________
Jasef: "I am a lawful man. By serving pee to cops, you support terrorism. I will lock this thread."

12Bass: re: 9/11 nano-thermite conspiracy theory "Nanotermites... turns out it was a spelling error. Also explains the lack of explosions before the collapse."

Last edited by drumphil : Aug 20, 2021 at 10:44 AM.
drumphil is offline   Reply With Quote