Atlas OS, low latency OS for sound quality

[Brian8591]

Hi, I just wondered if anyone had tried Atlas OS for running their USB DACs. It is a debloated version of W10 that is tuned to give lower latency for gamers, so I was wondering if it would be much better than the usual W10 that had been debloated manually.

[Archimago]

I'm not sure since when did low-latency alone make a difference to sound quality assuming all else bit-perfect.😉 Yes if one is gaming it would make a difference.

 

Would be interesting to hear of opinions, gear used, and what one is comparing with. But by default one should clearly not hold strong expectations of significant audible differences. 

[Brian8591]

3 hours ago, Archimago said:

I'm not sure since when did low-latency alone make a difference to sound quality assuming all else bit-perfect.😉 Yes if one is gaming it would make a difference.

 

Would be interesting to hear of opinions, gear used, and what one is comparing with. But by default one should clearly not hold strong expectations of significant audible differences. 

Well I've read quite a few posts saying that W10 sounded better than W7, and that the reasons were due to the software improvements and lower latency of W10. There is actually a whole operating system designed just for audiophiles called Audiophile Linux and it is designed to have the minimal number of processes and latency time to improve sound quality. I noticed myself that W10 sounded better than W7 and some people have advocated using Windows Server and Win PE to reduce the number of processes and improve sound quality. I'd rather use Windows 10 or Atlas OS because they are more convenient and user friendly.

[One and a half]

On 4/28/2023 at 6:35 PM, Brian8591 said:

Hi, I just wondered if anyone had tried Atlas OS for running their USB DACs. It is a debloated version of W10 that is tuned to give lower latency for gamers, so I was wondering if it would be much better than the usual W10 that had been debloated manually.

Having a look at what Atlas has to say makes a lot of sense. Why should the machine process activity increasing common mode noise, when it's not necessary. If this improves latency for gaming, then by all means a de-bloated windows works.  For audio, modern processors for playback aren't reliant on latency, but recording-playback certainly is in pro circles.

I would have thought that reduced activity/processes in the computer would lower heat, subsequently power usage, but some programs may rely on a regular type of Windows install to actually work, IIRC jriver made a song and dance about JPLAY's hibernation mode style of playback that would not work with their player. 

I would like to give Atlas a shot, and see if Audirvana can run on a de-bloated install. From the Atlas web site, the processes that are deleted, an audio server would not need anyway.

 

This harks to a concept of a  debloated OS that Fidelizer has undertaken over many years specifically for audio. There's the Swiss fella (sorry forgotten the name) using Windows Server and similar concept.  Maybe Atlas will work for audio, good question.

[Windows X]

Tried Atlas OS before due to client's suggestion. It's designed for gamers. It's fast and snappier that's for sure. But some tweaks also applied for gaming and audiophile wise, I don't quite like the changes that boost gaming performance and affect sonic performance.

 

If you want to try debloating with Playbook script, maybe you can consider ReviOS which focus on lightweight and non-intrusive approach. I used to try both AtlasOS and ReviOS and I found more positive results on ReviOS. Not as good as my on NimitraOS in my Nimitra server but I guess free debloating tool on active OS won't go wrong with ReviOS.

 

Regards,

Keetakawee

[yamamoto2002]

See this document https://learn.microsoft.com/en-us/windows-hardware/drivers/audio/low-latency-audio

 

Quote

Low latency means higher power consumption. If the system uses 10-ms buffers, it means that the CPU will wake up every 10 ms, fill the data buffer and go to sleep. However, if the system uses 1-ms buffers, it means that the CPU will wake up every 1 ms. In the second scenario, this means that the CPU will wake up more often and the power consumption will increase. This will decrease battery life.

 

(I'm not sure if Windows can wake up user thread with 1ms granularity reliably.)

 

In most cases there is no sonic difference when latency value (PCM buffer size) is changed. Sometimes more power consumption means more noise (electronically and acoustically). Sometimes noise pitch is changed when power consumption is increased, it is YMMV.

 

My opinion is, larger buffer is better with music listening scenario, because of less occurrence of sound stuttering due to buffer underruns. As Archimago-san said, low-latency audio is absolute necessary for some usage scenario such as gaming application or realtime monitoring of recorded sound, therefore low-latency audio is introduced on Windows 10, not to improve music listening experience

 

[Zauurx]

I've always taken great care to balance load and cpu latency, and it has an impact.