View Single Post
Old Aug 19, 2021, 12:30 AM   #8
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

Let's have a look at the claims made about what it does, and what that achieves.

Quote:
Improve audio performance by giving audio task more priority, improve low latency stability, and increase accuracy of clock resolution.

Take control of system and network utilization completely by optimizing Windows multimedia scheduling platform.
This makes no sense. The sound card has internal clocking. Audio is moved in fixed buffer size units. The sound card plays this data at a rate fixed by it's own internal clock. Either the audio chunks make it through in time, for perfect playback, or they don't and you get clicks and pops.

There is no range of quality. The card gets the chunks of audio in time to avoid dropping a buffer, or it doesn't and you get clicks and pops.


Quote:
Isolate non-audio processes and reduce its priority to minimize any possible interference to audio task and related processes.

Launch music player application at high priority without interference from user permissions for solid performance.
Again, if you aren't getting clicks and pops, the sound card is getting the data fast enough, and audio quality isn't improved by giving these processes higher priority.


Quote:
Stop/Disable most non-audio system services greatly reducing system footprints for pure audio performance.

Smartly adjust system services based on audio profiles to maintain required features like network and remote control on demand.
OK, now we get to a concept that could in theory (sorta kind of) improve audio fidelity, but would only matter if your sound device has a broken design.

Reducing system activity can in theory reduce the amount of noise the sound card is exposed to, through the PCIE data and power lanes, and from RF radiation inside the computer case.

But, playing around with process priorities doesn't stop the rest of the system from making this noise. Even if the audio processes are given a higher priority, doing stuff that puts load on the system will still cause the hardware to create this noise. Doing something that causes high CPU load will still make the same amount of noise. Process priorities may give the audio services higher priority, but they only have so much work to do, and the rest of the time will be spent running whatever else needs doing as fast as it can be done.

Any sound card that achieves good fidelity successfully filters out this noise in the first place. Designs that fail at this have measurable noise problems. The solution is to get a card that doesn't have a faulty design. If the Nu Audio didn't do this successfully, you wouldn't see the low noise levels shown in the measurements from the review I linked to. Sure, there are cards that do better, but the results are still pretty good, and if you want better than that, get a better sound card.

Now, if you have a crappy enough sound card, and a crappy enough PSU, and a crappy enough motherboard, there may be enough noise to be audible, but in general, a well designed sound card shouldn't care, and the claims made by fidelizer about reducing this noise are faulty anyway.

So why is it that in some combinations of computer, amp, speakers and sound card, even with a good sound card, that noise can be heard from high load, especially high GPU load?

The answer is ground loops. This happens when your computer has a PSU power cable with a ground pin, and your amp, or speaker with amp in it, has a PSU cable with a ground pin.

Because amateur gear like the Nu Audio has unbalanced outputs, with left signal, right signal, and an audio ground, you now have two ground paths between the computer and the speakers. The audio ground between the sound card and the amp, and the PSU ground from the computer and the amp.

For electrical engineering reasons that would take a few hours to explain properly, this is a fundamentally incompatible connection system. Ground loops mean that using sound cards that are well designed and do properly filter the internal noise from the PC, can still end up with noise being induced when connected to an amp that has a safety pin on it's PSU cable (three pin power plug). No amount of extra filtering on the sound card can avoid this, and even if you reduce system load, the noise will still happen.

Even professional gear that uses balanced connections can have this problem. Balanced connections don't use a signal and ground wire. They use a positive polarity signal, and a negative polarity signal connection, with a shield that isn't used as an audio signal ground.

Even then to avoid ground loop noise the shield often has to be disconnected between devices.

Some enterprising minds have though to themselves that if two connected grounds cause problems, how about snipping the ground plug off the PSU cable to break the ground loop while leaving the audio ground connected so that the audio circuit is still complete.... Bad idea, because now your PSU has no safety ground!

So, the options are, use an amp that is double insulated and requires no safety ground, use balanced IO like professional gear where you can break the ground path by disconnecting the shield without breaking the audio circuit because ground isn't needed for signal flow, or use a DI box that can convert the unbalanced signal to a balanced signal and can disconnect the ground path.


Gawd, I was going to try and give a "short version" answer, but as is usually the case it's turned into an essay.

I'm the technical officer for a recording and video production studio. If you have audio fidelity problems with your system I'm happy to help you sort them out, but the answers aren't found in snake oil placebo stuff like fidelizer.

What system do you have? What amps and speakers?
__________________
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 19, 2021 at 12:52 AM.
drumphil is offline   Reply With Quote