Whats your favourite Linux Distro?

[InVasMani]

I think this could be of interest to the topic

Ubuntu 24.04 Flavours: Kubuntu, Budgie, MATE, Cinnamon & More​

I'm interested in how the Cinnamon tastes...is it good?

Spoiler: yummy...

 

[Ferrum Master]

Believe me I know

I use pulse on top of jackd which is reasonably stable when you do it yourself.

Actually if I listen something in background I use a pretty wild approach and bypass the whole audio stack. I use Foobar still in Linux via wine... as all other music players simply suck also... but there I use UPnP MediaRender output and launch BubbleUPnPServer in background, config it via web and foobar sees the network speakers, stream audio data on your network. No pops, no crackles... just works.

 

[Courier 6]

I'm interested in how the Cinnamon tastes...is it good?

Spoiler: yummy...

I don´t use it, so it´s better you ask one of the other guys for a better evaluation

 

[xtreemchaos]

well ive only 1 mint because the last time i used linux was when i had hair and teeth . its so easy nowa days.

 

[Mindweaver]

I mainly use Ubuntu. My work route is pretty simple. I power on Ubuntu after login I press Ctrl+Alt+T. Aww who am I kidding.. I mainly use Ubuntu Server without the GUI. lol

 

[Solaris17]

I mainly use Ubuntu. My work route is pretty simple. I power on Ubuntu after login I press Ctrl+Alt+T. Aww who am I kidding.. I mainly use Ubuntu Server without the GUI. lol

Same lol ssh life. The wars of the DEs are for the privileged. I don’t even see a desktop for my job

 

[Ferrum Master]

I mainly use Ubuntu. My work route is pretty simple. I power on Ubuntu after login I press Ctrl+Alt+T. Aww who am I kidding.. I mainly use Ubuntu Server without the GUI. lol

So this is how you do TPU? w3m?

 

[GoldenX]

Fedora.

 

[fec32a4de]

A stupid question, because someone needs to start asking stupid questions around here and its gonna be me:

Whenever I play Hitman (2016) which has a native Linux client through Steam or Hitman 3/World of Assassination via Wine (but not proton for some reason).

The audio despite being the same speakers, same connection (3.5mm) becomes dynamic and I can hear a wider range of sounds, footsteps, etc. Than whenever I play any of these games on Windows.

In fact most sound video game sounds sound a lot better under Linux (music is unchanged). I tested on Linux Mint and Ubuntu (forgot the version numbers) about a year ago.

Whats Linux doing under the hood that makes it better than whatever MS is doing? (Realtek ALC1200)

 

[Ferrum Master]

Whenever I play Hitman (2016) which has a native Linux client through Steam or Hitman 3/World of Assassination via Wine (but not proton for some reason).

Because WINE imho does resample I am not sure about the default config for hitman, if it bypasses Winepusle and uses ALSA. But in general it looks like more that something does not work than work as it should. Hitman uses Wwise, there should not be some differences.

Most probably you are on pipewire now and use pw-top in terminal and look for sample rate differences to figure out what's going on versus proton, sinks and sources.

 

[SchumannFrequency]

The audio despite being the same speakers, same connection (3.5mm) becomes dynamic and I can hear a wider range of sounds, footsteps, etc. Than whenever I play any of these games on Windows.

The best sound Linux has for games is probably OpenAL. OpenAL is particularly interesting because it is a 3D sound system.
OpenAL Soft is a fork licensed under LGPL that's still up-to-date and includes HRTF on headphones and up to 3rd-order Ambisonics.
OpenAL Soft can do a form of audio raytracing for convincing occlusion effects.
Games can also use sdl and maybe sdl works well on your hardware/operating system combination.

Linux' sound quality for audio players such as Qmmp is a hit-or-miss situation.
It depends heavily on the motherboard and the audio chip you are using.

On most hardware I have, Linux sounds horrible.
But I once installed Linux on an acquaintance's old (high-end) Dell Latitude.
I immediately noticed that Qmmp produced abnormally detailed and clean sound on this hardware on Linux.
I don't know if this is because the hardware is much better, or because Dell is popular with Linux developers. Perhaps a combination of both.

For most hardware, OpenBSD and FreeBSD do better than Linux.
If you combine OpenBSD/FreeBSD with music player daemon you can produce high-quality sound on the very cheapest motherboards.
With Linux this is not the case.

 

[SchumannFrequency]

The audio despite being the same speakers, same connection (3.5mm) becomes dynamic and I can hear a wider range of sounds, footsteps, etc. Than whenever I play any of these games on Windows.
In fact most sound video game sounds sound a lot better under Linux (music is unchanged). I tested on Linux Mint and Ubuntu (forgot the version numbers) about a year ago.
Whats Linux doing under the hood that makes it better than whatever MS is doing? (Realtek ALC1200)

Here you have two recordings that give a pretty good idea of how ALSA and OSS sound over an analogue connection.

OpenBSD + music player daemon + OSS

OpenBSD sound demo with Infinity Alpha 5 + Sony STR-DB790.wav

drive.google.com

Calculate Linux + Qmmp + ALSA (direct ALSA connection without PulseAudio)

Calculate Linux sound demo with Infinity Alpha 5 + Sony STR-DB790.wav

drive.google.com

Both recordings were made 99% the same way. Same micro, same recording operating system (Void Linux), same recording notebook, same recording app.
The only difference is that one recording is in 48 KHz and the other in 96 KHz.
According to most experts, people usually cannot hear any difference between 48 KHz and 96 KHz.

The second track on both recordings is the same song. This makes comparison of both sound architectures easy.
To me, there are noticeable differences in how the electric guitar sounds on both recordings.

The way the electric guitar comes to the foreground is different in both recordings.
The volume balance of the instruments is maybe also different.

On the sound system used (Infinity speakers + Sony receiver), Calculate Linux sounds good.
However, my preference is OpenBSD for most songs.

There is hardware on which the audio sounds better in Linux than in my above example. (certain DELL hardware for example)
My general idea is that Linux and Windows are similar in terms of audio and it's going to depend on the specific hardware which of the two sounds better.
NetBSD, FreeBSD and OpenBSD have independent OSS implementations that tend to give better analogue sound than Linux and Windows. (on most hardware..)

 

[Ferrum Master]

Here you have two recordings that give a pretty good idea of how ALSA and OSS sound over an analogue connection.

Get a XMOS(~316) backed DAC/AMP and thank me later. It also can bug, so you know, it is PC, if it has some weird profile... and bad USB, even my PC sneezes USB and dmesg is error spammed if I put the DAC in higher USB numbers during boot, but I have shitsus mobo, so no wonders there.

Those work out of the box, drivers are in kernel, well almost, we will never see proper Linux DSD in our life time frames most probably, but not that it is needed. Let's not indulge ourselves into comparing audio stacks and how they operate in reality, you will have to take into account what the DAC does itself on hardware level and afterwards filtering it... none of us has R/2R or begin to dvelve into quantization artifacts for other topologies and their hybrids even. Basically that's what are you actually influencing here. For guitars adding pre ringing or actually not taking away in filtering phase, you get pleasant distortion pedal effect, thus win win situation, more HF distortion sounds more bright to our ears, so... there's your catch.

Compare different solutions, do proper SINAD... we have enough propehic reviewers across sites claiming healing properties when using some devices... If you cannot measure it, it ain't there. And for limited spectrum human hearing audio you always can... we aren't chasing Higgs boson particles here.

Each hardware solution needs to be analyzed their own to judge what's happening, it is not that simple. Windows WASAPI(Bypass windows volume service resampler) + Foobar is the best for bitperfect playback and nothing holds a candle to it. ASIO did before, but due to new hardware and OS nature it ain't usable anymore as it was in the past.

But either way... using XMOS based devices is a way to use Linux and while being with audio that doesn't suck, bark, hiccup, crackle, snore etc...

I did too much... in the age most people use Bluetooth shit earbuds, causing them permanent hearloss...

 

[SchumannFrequency]

Windows WASAPI(Bypass windows volume service resampler) + Foobar is the best for bitperfect playback and nothing holds a candle to it.

Foobar2000 has been a largely dead project for a long time.
There are many Linux/BSD music players that get much more frequent updates than Foobar2000.

We could do a FreeBSD users poll whether they think their audio in bitperfect mode + real-time sound + music player daemon sounds less good than Foobar + ASIO bitperfect mode.

My guess is that most people will prefer FreeBSD because among the FreeBSD forums, it seems to me that most FreeBSD users never say that audio is ‘less good’ than on windows.
Which of course means it's most likely the other way around.

Open Sound System (OSS4) superior to ALSA https://www.audiocircle.com/index.php?topic=95824.0
OSS https://linuxreviews.org/OSS
ALSA vs OSS4 https://www.overclock.net/threads/alsa-vs-oss4.826933/

There is also a ‘Music’ topic on the FreeBSD forums that is frequently posted and has become a bigger thread than 99.9% of the ‘Music’ threads started by windows10/11 users.

 

[Ferrum Master]

Foobar2000 has been a largely dead project for a long time.

Not sure what you mean by that. Peter does update it all the time, including plugins and keeps up with Windows API releases, fixing and speeding up code parts. You trust that piece of software.

What Linux/BSD player has more updates? In what? Show me one capable of DSD, same amount of working plugins and UI not made by 3year old kid. That's not even a competition, Foobar is superior and years ahead, so is Windows audio stack. I use Linux and root for it... but I am not blind, to be more precise - deaf this time and I am not turned into a rooting cheerleader.

I did all this talk for nothing... it not only in the software side. But strict conjunction with the hardware. Audio Drivers suck hard on Linux/BSD no OSS/ALSA can fix that, don't turn it into distro challenge or Linux vs BSD, save us from that nonsense. Only after drivers are fixed you can go splitting hairs about sound backends. Bloody nobody will fix latency issues or other limitations if the device manufacturer simply does not care and blame each other for most problems. Don't indulge ourselves into esoteric talks about what's better sounding. Measure it, SINAD, hard numbers, distortion figures, harmonics. Do charts and figures then do talking. Basically human beings barely keep their pants clean, even more trust our subjective hearing capabilities. There is no point discussing it with people who rely only on their ears... considering age, service in army, hearing defects and the last one is taste preferences, people like distortion and doesn't understand that they really prefer the worst performing product, but good for them, it is cheaper that way.

 

[SchumannFrequency]

Not sure what you mean by that. Peter does update it all the time, including plugins and keeps up with Windows API releases, fixing and speeding up code parts. You trust that piece of software.

Does one single person develop perhaps the most popular audio player in Windows?
rhythmbox has 300 developers.
music player daemon has more than 240 developers.
These are easily understandable numerical differences.

When I look at the software updates of Foobar2000, I see that there are hardly any (new) features added in new versions.
Many of the popular old skins still do not have a 64 bit version in 2024.
I also frequently see that bugs that have existed for 15 years before they get fixed.
These are the signs of a lack of manpower to deliver something of high quality.

Audio Drivers suck hard on Linux/BSD no OSS/ALSA can fix that, don't turn it into distro challenge or Linux vs BSD, save us from that nonsense. Only after drivers are fixed you can go splitting hairs about sound backends.

I really like music player daemon on Linux. Is foobar actually superior at repro... | Hacker News

news.ycombinator.com

#1 foobar can do bitperfect output. MPD too, given configured properly. A lot of players is missing this option (specifically bypassing system mixer), and it making sound noticably worse.
On MacOS situation is dire (there are some players, but their UX is terrible, imo)
#2 No, the quality of the sound itself isn't better. If anything a higher quality pipeline is easier in Linux IMO.

OSS vs ALSA https://forums.freebsd.org/threads/oss-vs-alsa.43785/
There is very little interest in ALSA among the FreeBSD developers as far as I know and remember reading from the mailing lists.
It's down to the fact that on hardware that is supported by the FreeBSD base system the FreeBSD sound system (that isn't quite OSS but uses the OSS API) is vastly superior to ALSA.

I understand the logic why many people think windows11 + foobar2000 + bitperfect ASIO should sound perfect. You think it will be the same bits as other (bitperfect) apps/operating systems that will be sent through the cable to the receiver.

With Flatpak and RPM you have a similar phenomenon. There are Flatpak developers who say this is just another way to package an app that (purely theoretically) has no impact on performance whatsoever. However, in practice I constantly measure that there are Flatpak apps that perform more than 10% lower than the RPM versions of the same app versions. With Steam you see that in (completely) CPU-bound situations the Flatpak Steam often performs lower than native Steam.

From a purely theoretical perspective, there is no reason why macOS and BSD systems should have better audio than Windows. But go to Microsoft and compare the skills of their developers with some Linux/BSD developers.

The skills, manpower and/or work commitment of some Windows developers are often not on par with the skills of open-source developers.

Q: Can I get a job at Microsoft with low grades?

Can I get a job at Microsoft with low grades?

Answer: Ofcourse you can ! There is a condition however - the amount of hard work you should have put in academics will have to be put in now. Sharpen your technical skills. Get relevant experience in a profile that Microsoft hires for. It might take some time to build your resume, but it is poss...

www.quora.com

A: Ofcourse you can!

I think that in the text above you will find several correct explanations why the audio that comes from Windows systems is usually of lower quality than the audio that comes from a free operating system.

 

[Ferrum Master]

Calm down dude, I did ask completely different questions, it seems you still do not understand the hardware side of things, not even a clue, that's the problem with coding monkeys especially in audio areas, they don't have the gear or lab to test what they exactly are doing, words mean nothing here. There is no point to argue. Chill.

Windows API is superior, nobody even argues about it, quit trolling.

 

[Athlonite]

I'm interested in how the Cinnamon tastes...is it good?

Spoiler: yummy...

I've a mate who uses Mint debian edition and from what I've seen of it and the little bit of playing around on it it's bleeding great he used to use the Ubuntu edition but they're being dicks about flat packs and few other things he dislikes with a near blood vessel popping fury
I can say if I were to go Linux at all it would be Mint Debian edition with cinnamon

 

[SchumannFrequency]

Windows API is superior, nobody even argues about it, quit trolling.

Why Vista and its successors sound worse https://www.theguardian.com/technology/2008/jan/31/microsoft.technology

For most pro audio recording applications, ASIO is the preferred driver format to use, particularly when using an audio interface.

Many professional DAW developers seem to be of the opinion that WASAPI is not the highest audio standard, otherwise the advice to use ASIO would not be given.

According to the REAPER user guide, in terms of performance, “ASIO drivers generally offer better latency than others.

Do most windows users use ASIO?

FreeBSD has the option kern.timecounter.alloweddeviation=0
= noticable sound latency <5ms

Does BSD or windows usually have the best performance/latency?

https://optim.tildacdn.net/tild6335-3537-4364-b731-353132636631/-/resize/652x/-/format/webp/image.png

Contributors to FreeBSD

A list of organizations and individuals who have contributed to FreeBSD

docs.freebsd.org

Many developers on the BSD team come from countries with better programming skills than Germany.
I suspect that ASIO was mainly created by German programmers.

It would make statistical sense if BSD systems have a better audio API than windows if the programmers have more expertise.

 

[gurusmi]

...Windows API is superior, nobody even argues about it,...

Only people not keen in system programming think so. But hey. Nowadays the lessest computer developers even know what a system developer is.

 

[SchumannFrequency]

Only people not keen in system programming think so.

I have the Infinity Alpha 5 + Sony STR-DB790 combination.
I have this setup fully custom tuned. I use the most natural settings for the receiver.
The subwoofer I have set slightly louder than the average setting and some other settings of the subwoofer are tuned as well.

When I play music on windows or Linux on this setup, +- 65% of the songs I play sound 'very good’.

However, when I play music on OpenBSD, 75% of the music sounds 'phenomenal'. Let's just say I can use my imagination to imagine how music might sound best.
OpenBSD + music player daemon takes the best I can think of and makes it ten times better.

I have heard very expensive systems in Hi-Fi shops but they were not on the same level.
I bought a GXT 259 RUDOX microphone to record this sound. This is the result.

OpenBSD sound demo with Infinity Alpha 5 + Sony STR-DB790.wav

drive.google.com

The GXT 259 RUDOX mic is a lot better than my previous mic, but is not close to the level of detail and quality needed to accurately reproduce this Infinity system.
What is the most optimal microphone to make this type of recording?

 

[unwind-protect]

The GXT 259 RUDOX mic is a lot better than my previous mic, but is not close to the level of detail and quality needed to accurately reproduce this Infinity system.
What is the most optimal microphone to make this type of recording?

Schoeps with omni capsules.

Schoeps CMC 62H Omnidirectional Condenser Mic Stereo Set

Complete modular small diaphragm condenser microphone stereo set including two amplifiers, two omnidirectional capsules, and accessories - features a flat frequency response and very low harmonic distortion and noise

vintageking.com

 

[Ferrum Master]

When I play music on windows or Linux on this setup, +- 65% of the songs I play sound 'very good’.

However, when I play music on OpenBSD, 75% of the music sounds 'phenomenal'. Let's just say I can use my imagination to imagine how music might sound best.
OpenBSD + music player daemon takes the best I can think of and makes it ten times better.

Holy... I haven't seen so much contradiction in while regarding digital to analog data conversion. While you previously even argued that audio cannot differ in between audio backends, now you fully claim it does.

There is no point discussing about devs, not sure why even you started it, I asked not to bring up this BSD vs Linux nonsense. After mentioning Rythmbox it made clear where the wind blows... 300 devs barely managing to do commits to 4-6 years old problems... 300 devs... but that's GNOME for you, ain't it?

Read up about signal reconstruction principles and what happens in DACs at hardware level, I mentioned it two times already... don't trust your ears, measure, hard facts. It is all snake oil you are saying here.

 

[gurusmi]

I have the Infinity Alpha 5 + Sony STR-DB790 combination....

But you realized that API is the official abbreviation for Application Programming Interface? There is no special system function or program behind. In general with the API you "Steer the underlying OS and Hardware. No? As i thought. To explain: An API is used in general between a Applicaation program and the underlying OS, and Hardware. So what this well defined fact has to do with your mass of senseless and useless words?

 

[SchumannFrequency]

While you previously even argued that audio cannot differ in between audio backends, now you fully claim it does.

I do not (literally) see that I write ‘audio cannot differ in between audio backends’. Can you quote me such that I know which statement you are talking about?

What I did say is that theoretically, with certain specific settings, different audio backends should give the same sound. I then also explained why theory and practice differ in reality. Programmers' skills are limited and things are always implemented in a specific (different) way. There are easily audible differences between audio architectures and audio-related software.

But you realized that API is the official abbreviation for Application Programming Interface? There is no special system function or program behind. In general with the API you "Steer the underlying OS and Hardware. No? As i thought. To explain: An API is used in general between a Applicaation program and the underlying OS, and Hardware. So what this well defined fact has to do with your mass of senseless and useless words?

I have answered far more difficult questions (in tests) than what the word API means.
You hopefully realise that macOS, Linux, BSD and windows use vastly different software protocols to manage sound cards and MIDI ports?
Software can have a stronger impact on sound quality than hardware in many situations.

Schoeps with omni capsules.

The price of a really good micro is higher than the budget I have for it.
I made a (new) recording with the GXT 259 RUDOX and this time I did not place the mic centrally between the L/R boxes but just in front of the left box furthest from the subwoofer.
This is the recording. https://drive.google.com/file/d/1pkv-QqQjMVRs4Pya32fwjSmSRb3DdtiY/view

My intention is to have a bit more detail in the recording by placing the mic closer to the speaker.
The subwoofer is producing problematic tones for this mic and perhaps distorting all the other frequencies as well.
In the last recording, the mic is slightly less in the direct output radius of the subwoofer. There should be slightly less subwoofer distortion in this new recording.

The first track still has enough subwoofer to significantly distort the song. But the second and third tracks in this recording fared better.

If you listen very carefully to the second and third tracks, you can notice that my Infinity speakers are capable of producing purer sound than the following ONKYO + Polk setup:

For my previous work, I had to go for a medical check-up once and they also did a detailed hearing test then.
According to the doctor, I had achieved an exceptionally good result in this hearing test. Better than 98% of people.

I am not going to claim that everyone is going to hear the same as me.
But I can hear (clearly) that the ONKYO TX-NR 6100 + Polk Signature ELITE Series produces less pure sound than my OpenBSD + Infinity Alpha 5 + Sony STR-DB790 setup.