Topic by jay_chi

  • Avatar
  • jay_chi
  • Member
  • Offline
  • Registered: 15-03-2025
  • Posts: 1

Topic: Same Hardware Linux vs Windows version. min latency?

I am a prospective user.
I am not going to use Apple for anything. So "use a mac" is not helpful.
TL;DR: Is there any/enough latency difference in a linux install to make it worth the hassle over windows?

I am considering creating a dedicated pianoteq system with a few features:
1) minimal latency
2) quick GUI interface
3) likely on a recent low-end thinkpad - either current windows or recent linux on it.
4) works with Yamaha CK-88
5) can be updated over time both OS and Pianoteq (reasonably until end-of-life of hardware)
6) total cost (including thinkpad) < US$1000.
7) system would be fully dedicated to Pianoteq, no other DAW or anything significant installed on it.
8) this is primarily for high intermediate or low expert level practice. RCM 8+.

I am (extremely) competent with linux, but I do not enjoy fooling around with it. So, I prefer not to mess around much with settings or to optimize via the trial-error-research-banghead cycle that often occurs. I'd rather spend that time practicing. I am happy to follow a howto if there is one out there ( such as https://wiki.linuxaudio.org/wiki/system_configuration ). A prepackaged distro/bundle for pianoteq would be even better.

I have read that windows has hardware limitations on sound "cards" that increase latency and just in general, windows is less tunable for process management than Linux is. Additionally, I'd spend some time optimizing windows and uninstalling all the extras (this device is 100% dedicated to pianoteq). Windows prep would be less work than tuning linux, so in comparison, windows would be more ready out of the box, but may not be able to minimize latency like linux, and I guess that is my question to the group.

Is there any/enough latency improvement in a linux install to make it worth the hassle over windows?

FYI, I am considering this setup versus a (more expensive and older) fully managed hardware expansion such as Dexibell SX8, or possibly but unlikely a V3 Piano Grand XXL.

I'd love the actual experiences of people that have used linux and windows and if they see a difference and if they find linux to be reasonably robust?
TIA

-JC

Reply by robisme

  • Avatar
  • robisme
  • Member
  • Offline
  • Registered: 28-03-2022
  • Posts: 17

Re: Same Hardware Linux vs Windows version. min latency?

jay_chi wrote:

I am a prospective user.
I am not going to use Apple for anything. So "use a mac" is not helpful.
TL;DR: Is there any/enough latency difference in a linux install to make it worth the hassle over windows?

I am considering creating a dedicated pianoteq system with a few features:
1) minimal latency
2) quick GUI interface
3) likely on a recent low-end thinkpad - either current windows or recent linux on it.
4) works with Yamaha CK-88
5) can be updated over time both OS and Pianoteq (reasonably until end-of-life of hardware)
6) total cost (including thinkpad) < US$1000.
7) system would be fully dedicated to Pianoteq, no other DAW or anything significant installed on it.
8) this is primarily for high intermediate or low expert level practice. RCM 8+.

I am (extremely) competent with linux, but I do not enjoy fooling around with it. So, I prefer not to mess around much with settings or to optimize via the trial-error-research-banghead cycle that often occurs. I'd rather spend that time practicing. I am happy to follow a howto if there is one out there ( such as https://wiki.linuxaudio.org/wiki/system_configuration ). A prepackaged distro/bundle for pianoteq would be even better.

I have read that windows has hardware limitations on sound "cards" that increase latency and just in general, windows is less tunable for process management than Linux is. Additionally, I'd spend some time optimizing windows and uninstalling all the extras (this device is 100% dedicated to pianoteq). Windows prep would be less work than tuning linux, so in comparison, windows would be more ready out of the box, but may not be able to minimize latency like linux, and I guess that is my question to the group.

Is there any/enough latency improvement in a linux install to make it worth the hassle over windows?

FYI, I am considering this setup versus a (more expensive and older) fully managed hardware expansion such as Dexibell SX8, or possibly but unlikely a V3 Piano Grand XXL.

I'd love the actual experiences of people that have used linux and windows and if they see a difference and if they find linux to be reasonably robust?
TIA

-JC


I would say the hassle is clearly windows, which is a pita to install, avoid the traps of every 30 days trial of this bloatware, enter your credencies here and there.
You know linux ? Go Debian 12, almost no need to update. It's a set it once and enjoy. Just be sure to have your user in the audio group and set limits.conf as pianoteq says to.
Cheers.

Reply by brundlefly

  • Avatar
  • brundlefly
  • Member
  • Offline
  • Registered: 31-03-2022
  • Posts: 221

Re: Same Hardware Linux vs Windows version. min latency?

robisme wrote:

Is there any/enough latency difference in a linux install to make it worth the hassle over windows?

Regardless of the O/S, getting low latency is all about the performance of your MIDI and audio interface hardware, their drivers, and the bus by which they're connected to the host. I would choose the O/S that gives you the broadest choice of I/O hardware which is pretty clearly Windows, I think.

Reply by dv

  • Avatar
  • dv
  • Member
  • Offline
  • Registered: 01-11-2021
  • Posts: 344

Re: Same Hardware Linux vs Windows version. min latency?

brundlefly wrote:
robisme wrote:

Is there any/enough latency difference in a linux install to make it worth the hassle over windows?

Regardless of the O/S, getting low latency is all about the performance of your MIDI and audio interface hardware, their drivers, and the bus by which they're connected to the host.

Totally true.

brundlefly wrote:

I would choose the O/S that gives you the broadest choice of I/O hardware which is pretty clearly Windows, I think.

That was maybe true 10 years ago.
But most importantly you don't need the *broadest choice*!! What do you care about the support for the 723 audio interfaces you don't own??

You care that the two (or maybe even one!) piece of hardware you have decided to buy are supported. And you care that support (driver) are solid, stable and blazing fast.

That was the case for me when I used a Foucrite Scarlett Solo under Linux 4 years ago. At the moment I am not using anything so I can't comment on most recent improvements (if any) with either OS.

I prefer not to mess around much with settings or to optimize via the trial-error-research-banghead cycle that often occurs.

Frankly, that was the case 15 years ago, now it pretty much everythings works-fine-out-of-the-box and optimize-by-following-documentation-or-blog-posts. At the time (4 years ago as said above), it was actually *much* harder to do this for Windows than from Linux!

But most importantly, in my experience (not for audio, but for anything else my experience is current), windows "pull the chair you are seated from under you", i.e. constantly changes things and forces you to re-do the same thing over and over or learn whatever crap they decided to change. With Linux, once you have made the (limited) effort to configure it correctly once, things stay as you set them forever. If I weren't forced by work, I would never use Windows.

Where do I find a list of all posts I upvoted? :(

Reply by brundlefly

  • Avatar
  • brundlefly
  • Member
  • Offline
  • Registered: 31-03-2022
  • Posts: 221

Re: Same Hardware Linux vs Windows version. min latency?

dv wrote:

What do you care about the support for the 723 audio interfaces you don't own?

The point was to have 723 to choose from in the first place... or at least some reasonable number from an array of reputable manufacturers. Admittedly I don't know what the state of audio/MIDI hardware/driver support is for Linux now. If it's reasonable, just go with the O/S that's the most familiar, neither is going to make or break the latency performance. The devices on the MOBO that determine bus performance and (and DPC latency in the case of Windows) are also more important than the O/S itself.

Reply by Professor Leandro Duarte

  • Avatar
  • Professor Leandro Duarte
  • Member
  • Offline
  • From: Brasil
  • Registered: 03-03-2019
  • Posts: 217

Re: Same Hardware Linux vs Windows version. min latency?

I've been using Linux (Ubuntu Studio) for over five years and I don't think I'll ever go back to Windows.

For working with music, Linux distros are much more fluid, with low-latency or real-time kernels.

I don't know about other software, but Pianoteq with Ubuntu Studio is a perfect match!

Respeito, Esforço e Sabedoria

Reply by marcos daniel

  • Avatar
  • marcos daniel
  • Member
  • Offline
  • Registered: 02-09-2014
  • Posts: 280

Re: Same Hardware Linux vs Windows version. min latency?

In my experience, Linux had slightly better performance compared to Windows on the same PC. I used a laptop that was somewhat weak on the CPU side and occasionally experienced CPU overloads in Windows, but never in Linux on that setup.

With a reasonably powerful PC, there is no appreciable difference.

That said, latencies are similar. Some time ago, in a thread, other users and I posted some comparisons. The main difference at that time was that when changing the buffer size, latency increased more in Linux than in Windows.

old post: https://forum.modartt.com/viewtopic.php?id=8170

Reply by brundlefly

  • Avatar
  • brundlefly
  • Member
  • Offline
  • Registered: 31-03-2022
  • Posts: 221

Re: Same Hardware Linux vs Windows version. min latency?

marcos daniel wrote:

The main difference at that time was that when changing the buffer size, latency increased more in Linux than in Windows.

That's pretty weird. Latency should just change by the difference in samples between the two buffer sizes (times two for RTL) unless the interface driver is adding 'safety' buffers as some do. But in that case, the result should still be the same for a given interface.

Reply by hebele

  • Avatar
  • hebele
  • Member
  • Offline
  • Registered: 23-07-2024
  • Posts: 8

Re: Same Hardware Linux vs Windows version. min latency?

I did not measure latency difference. But I agree with others here. Linux is much easier to tame. It is possible to control what exactly running in the OS. Windows is a mess. Surprisingly it is also harder to set up a low latency environment. Midi drivers are not great. ASIO can be painful. In Linux default pipewire config works with low latency so that I didn't bother to customise it.

Linux can also have better hardware support. I have a microkorg bluetooth keyboard. It just works in Linux. I couldn't manage to connect it Windows via bluetooth. Although this is mostly on Korg having terrible windows drivers.

However, Windows has better commercial software support. Pianoteq is an exception having great Linux support. Most other VSTs require wine to run and that is hit or miss. And even it works, things can break after updates.

So pros and cons to each.

Currently I use an old surfacebook running Pianoteq and Mobilesheets. It sits on the music rest. I like the setup. I have to say Windows 11 touchscreen support is fine, probably better than any Linux DE. But I have to ignore daily reminders about Windows being Windows. Recently it started to show some ads via notification popups! I am pretty sure I can disable it if I spent some time to figure out. But why I have to fight with my OS constantly.

Reply by Kengrayfield

  • Avatar
  • Kengrayfield
  • Member
  • Offline
  • Registered: 22-03-2009
  • Posts: 31

Re: Same Hardware Linux vs Windows version. min latency?

I have to agree that it's quite a fight to get Windows 11 to finally work without audio pops, because it has about 10 different settings that would put the various hardware to sleep. If I don't do most of my work on Windows, I might really switch to Linux one day.

Reply by Chopin87

  • Avatar
  • Chopin87
  • Member
  • Offline
  • Registered: 18-10-2012
  • Posts: 322

Re: Same Hardware Linux vs Windows version. min latency?

Quick reminder that you don't necessarily need a modded kernel any more if you use a kernel from 6.12 onwards given RealTime patch was added to the mainline from that release.

"And live to be the show and gaze o' the time."  (William Shakespeare)

Reply by dubc

  • Avatar
  • dubc
  • Member
  • Offline
  • Registered: 07-08-2022
  • Posts: 32

Re: Same Hardware Linux vs Windows version. min latency?

Chopin87 wrote:

Quick reminder that you don't necessarily need a modded kernel any more if you use a kernel from 6.12 onwards given RealTime patch was added to the mainline from that release.


I never knew that. Im still on https://liquorix.net/

Im going to look to update to 6.12 RT as long as pre-compiled exists. Turns out Ubuntu Pro is free for 5 installs and has RT included!!. I would defeinately pay if required. Yay!!!

Thank-you!

Also - I record only on linux and its many times better with stability than windows. Its easier to use once set up also. It allows me to really focus on the music.

Last edited by dubc (02-06-2025 01:54)

Reply by dubc

  • Avatar
  • dubc
  • Member
  • Offline
  • Registered: 07-08-2022
  • Posts: 32

Re: Same Hardware Linux vs Windows version. min latency?

dubc wrote:
Chopin87 wrote:

Quick reminder that you don't necessarily need a modded kernel any more if you use a kernel from 6.12 onwards given RealTime patch was added to the mainline from that release.


I never knew that. Im still on https://liquorix.net/

Im going to look to update to 6.12 RT as long as pre-compiled exists. Turns out Ubuntu Pro is free for 5 installs and has RT included!!. I would defeinately pay if required. Yay!!!

Thank-you!

Also - I record only on linux and its many times better with stability than windows. Its easier to use once set up also. It allows me to really focus on the music.

I updated to the realtime kernel. That was a major pain. There is 5 free license for Ubuntu Pro Subscription https://ubuntu.com/pro/subscribe

Last edited by dubc (03-06-2025 07:59)

Reply by dikrek

  • Avatar
  • dikrek
  • Member
  • Offline
  • Registered: 06-02-2023
  • Posts: 147

Re: Same Hardware Linux vs Windows version. min latency?

dubc wrote:
dubc wrote:
Chopin87 wrote:

Quick reminder that you don't necessarily need a modded kernel any more if you use a kernel from 6.12 onwards given RealTime patch was added to the mainline from that release.


I never knew that. Im still on https://liquorix.net/

Im going to look to update to 6.12 RT as long as pre-compiled exists. Turns out Ubuntu Pro is free for 5 installs and has RT included!!. I would defeinately pay if required. Yay!!!

Thank-you!

Also - I record only on linux and its many times better with stability than windows. Its easier to use once set up also. It allows me to really focus on the music.

I updated to the realtime kernel. That was a major pain. There is 5 free license for Ubuntu Pro Subscription https://ubuntu.com/pro/subscribe

Why not the low latency kernel instead? RT has some drawbacks

Reply by dubc

  • Avatar
  • dubc
  • Member
  • Offline
  • Registered: 07-08-2022
  • Posts: 32

Re: Same Hardware Linux vs Windows version. min latency?

dikrek wrote:

Why not the low latency kernel instead? RT has some drawbacks

I just bought a new laptop, installed ubuntu 25.04 with 6.14 RT + lowlatency kernel.

Latency is:

Default Lunar Lake integrated soundcard
- 1.3ms 64 frames @48kHz <- not bad at all but I will never even test that again as I never use integrated audio.

SSL 2 32bit 192kHz USB interface with lowlatency

- 0.2 ms 32 bit 192kHz 32 frames, small buffer
- 0.1 ms 32 bit 192 kHz 16 frames, small buffer
- 0.03 ms 32 bit 192 kHz 8 frames, small buffer

This is on a brand new ASUS Vivobook 16 Flip OLED TP3607SA-RJ019X Core Ultra 7

CPU: Intel Core Ultra 7-258V
CPU Cores: 8 Core || 8 Threads
RAM Size: 32GB DDR5
Storage Size: 1TB SSD Upgradable
GPU: Intel Arc iGPU
Screen Size: 16" QHD+ 120Hz OLED

I am more than blown away with this.

Battery life is a reasonable 8h (windows 11 proclaims to get 22h max but I installed ubuntu as a fresh install and never booted into it to test).

Last edited by dubc (Yesterday 10:21)

Reply by dikrek

  • Avatar
  • dikrek
  • Member
  • Offline
  • Registered: 06-02-2023
  • Posts: 147

Re: Same Hardware Linux vs Windows version. min latency?

dubc wrote:
dikrek wrote:

Why not the low latency kernel instead? RT has some drawbacks

I just bought a new laptop, installed ubuntu 25.04 with 6.14 RT + lowlatency kernel.

Latency is:

Default Lunar Lake integrated soundcard
- 1.3ms 64 frames @48kHz <- not bad at all but I will never even test that again as I never use integrated audio.

SSL 2 32bit 192kHz USB interface with lowlatency

- 0.2 ms 32 bit 192kHz 32 frames, small buffer
- 0.1 ms 32 bit 192 kHz 16 frames, small buffer
- 0.03 ms 32 bit 192 kHz 8 frames, small buffer

This is on a brand new ASUS Vivobook 16 Flip OLED TP3607SA-RJ019X Core Ultra 7

CPU: Intel Core Ultra 7-258V
CPU Cores: 8 Core || 8 Threads
RAM Size: 32GB DDR5
Storage Size: 1TB SSD Upgradable
GPU: Intel Arc iGPU
Screen Size: 16" QHD+ 120Hz OLED

I am more than blown away with this.

Battery life is a reasonable 8h (windows 11 proclaims to get 22h max but I installed ubuntu as a fresh install and never booted into it to test).

What if you just use lowlatency and not the RT kernel?

Reply by dv

  • Avatar
  • dv
  • Member
  • Offline
  • Registered: 01-11-2021
  • Posts: 344

Re: Same Hardware Linux vs Windows version. min latency?

dubc wrote:

- 0.2 ms 32 bit 192kHz 32 frames, small buffer
- 0.1 ms 32 bit 192 kHz 16 frames, small buffer
- 0.03 ms 32 bit 192 kHz 8 frames, small buffer

I am more than blown away with this.

That is indeed impressive, but how did you measure that? More than the "declared" latency (which includes only what the software declaring it knows about), I think it's important to know actual latency. The only way I know to "scientifically" measure the latter is with a highly accurate high-speed audio-video recording device -- checking the time delay between your pushing down the key(s) and the sound being recorded.
It's very important to keep in mind that the recording device needs to have very tight sync between audio and video, and that the speed of sound in air on typical conditions is about 33cm for each ms, so if recorded with a microphone (rather than with a line-in) the distance from the speakers is utterly important.

I explored using my super-slow-motion settings in my phone to do such measurements, but in the end I was not able to achieve anything accurate enough and decided to go with "gut feeling" wrt what feels "right" when playing.

Where do I find a list of all posts I upvoted? :(

Reply by dubc

  • Avatar
  • dubc
  • Member
  • Offline
  • Registered: 07-08-2022
  • Posts: 32

Re: Same Hardware Linux vs Windows version. min latency?

dikrek wrote:

What if you just use lowlatency and not the RT kernel?

I'm using the lowlatency rt kernel. Are you disappointed in these results?

Reply by dubc

  • Avatar
  • dubc
  • Member
  • Offline
  • Registered: 07-08-2022
  • Posts: 32

Re: Same Hardware Linux vs Windows version. min latency?

dv wrote:

That is indeed impressive, but how did you measure that? More than the "declared" latency (which includes only what the software declaring it knows about), I think it's important to know actual latency. The only way I know to "scientifically" measure the latter is with a highly accurate high-speed audio-video recording device -- checking the time delay between your pushing down the key(s) and the sound being recorded.
It's very important to keep in mind that the recording device needs to have very tight sync between audio and video, and that the speed of sound in air on typical conditions is about 33cm for each ms, so if recorded with a microphone (rather than with a line-in) the distance from the speakers is utterly important.

I explored using my super-slow-motion settings in my phone to do such measurements, but in the end I was not able to achieve anything accurate enough and decided to go with "gut feeling" wrt what feels "right" when playing.


No test can truly be perfect as the testing overhead comes into play if running on the same system.

I've been recording 35 years and as many know, some latency is just too much. I tested this with  jackd.

jackd -R -P75 -t2000 -dalsa -dhw:1,0 -p16 -n2 -r192000 -s

Reply by dikrek

  • Avatar
  • dikrek
  • Member
  • Offline
  • Registered: 06-02-2023
  • Posts: 147

Re: Same Hardware Linux vs Windows version. min latency?

dubc wrote:
dikrek wrote:

What if you just use lowlatency and not the RT kernel?

I'm using the lowlatency rt kernel. Are you disappointed in these results?

I thought there was another kernel option with just lowlatency, not using the RT mode.

Reply by dikrek

  • Avatar
  • dikrek
  • Member
  • Offline
  • Registered: 06-02-2023
  • Posts: 147

Re: Same Hardware Linux vs Windows version. min latency?

dubc wrote:
dv wrote:

That is indeed impressive, but how did you measure that? More than the "declared" latency (which includes only what the software declaring it knows about), I think it's important to know actual latency. The only way I know to "scientifically" measure the latter is with a highly accurate high-speed audio-video recording device -- checking the time delay between your pushing down the key(s) and the sound being recorded.
It's very important to keep in mind that the recording device needs to have very tight sync between audio and video, and that the speed of sound in air on typical conditions is about 33cm for each ms, so if recorded with a microphone (rather than with a line-in) the distance from the speakers is utterly important.

I explored using my super-slow-motion settings in my phone to do such measurements, but in the end I was not able to achieve anything accurate enough and decided to go with "gut feeling" wrt what feels "right" when playing.


No test can truly be perfect as the testing overhead comes into play if running on the same system.

I've been recording 35 years and as many know, some latency is just too much. I tested this with  jackd.

jackd -R -P75 -t2000 -dalsa -dhw:1,0 -p16 -n2 -r192000 -s

Try sudo apt install linux-lowlatency
sudo update-grub

That way you’ll get the lowlatency version (which has all the tunings minus the hard realtime scheduling).

It might be a better balance.

RT can significantly degrade overall throughput, battery life, and preempt things you may not want preempted.

So if I am running a DAW with lots of plugins, I would test to see which kernel doesn’t impact overall throughput too much yet also allows low latency.

For a general use machine I’d probably just go lowlatency.

Last edited by dikrek (Today 12:55)