Audiophile PC vs Streamer

[Diet Kokaine]

I'm waaaay late to this thread, but I just wanted to say its not the late 1990s/early 2000s anymore. I knew someone that fried a PC in the 1990s because they had long hair and leaned over a computer while it was running and the side panel was off and their hair touched something. Poof. Dead motherboard. These days there are YouTube videos of people intentionally trying to fry motherboards with electricity and failing.

Back when Linux was still in its infancy and Windows 9X was mainstream, I think that best case scenario the Windows audio subsystem could do 15-bits. I remember I had a PS/2 mouse and when you would move the mouse you would hear a high pitched noise with no audio playing. There were things like ASIO/Kernel Streaming but no software in the world can overcome hardware limitations. Isolation and such are much better on modern PCs.

A modern PC with properly configured software outputting to a USB isolator to an external DAC(or optical if you are really paranoid) should always be well above the range of human hearing when it comes to distortion as long as the DAC is competently designed.

I wish it was as simple as you're making it out to be.
This is not the case in any of the tests that I've carried out.
PC's are an extremely noisy source. You absolutely need a Linear ATX power supply and an externally powered (Linear powered) USB card to get a clean signal out of it.
You can 100% hear this noise on a High end system.
I think about 50% of the sound coming out of a DAC is your "Source".

Headphones are more forgiving than Speakers. This might be why your experience differs.

 

[fliz]

I wish it was as simple as you're making it out to be.
This is not the case in any of the tests that I've carried out.
PC's are an extremely noisy source. You absolutely need a Linear ATX power supply and an externally powered (Linear powered) USB card to get a clean signal out of it.
You can 100% hear this noise on a High end system.
I think about 50% of the sound coming out of a DAC is your "Source".

Headphones are more forgiving than Speakers. This might be why your experience differs.

I agree with everything you just said *except* the externally powered (linear powered) USB card is only necessary if your DAC relies on the 5v rail of the USB connection in any way.

There's a simple test for whether this is the case.

Turn off and unplug your DAC and plug it into your computer. If the computer "sees" it in this state, the DAC is using the 5v rail. If the computer only recognizes there's a DAC connected after you turn it on, then you're fine.

 

[theveterans]

I wish it was as simple as you're making it out to be.
This is not the case in any of the tests that I've carried out.
PC's are an extremely noisy source. You absolutely need a Linear ATX power supply and an externally powered (Linear powered) USB card to get a clean signal out of it.
You can 100% hear this noise on a High end system.
I think about 50% of the sound coming out of a DAC is your "Source".

Headphones are more forgiving than Speakers. This might be why your experience differs.

A lot of headphones aren't revealing enough of the limitations of the source (in combination of DAC), but with the likes of Raal 1995 Immanis, yes, the limitations are easily heard. It's the culmination of "believability factor" that's easily exposed with the Immanis.

 

[Diet Kokaine]

I agree with everything you just said *except* the externally powered (linear powered) USB card is only necessary if your DAC relies on the 5v rail of the USB connection in any way.

There's a simple test for whether this is the case.

Turn off and unplug your DAC and plug it into your computer. If the computer "sees" it in this state, the DAC is using the 5v rail. If the computer only recognizes there's a DAC connected after you turn it on, then you're fine.

My DAC doesn't use 5V from the source.
My DIY USB cable doesn't have 5V or ground connected. Just data+ and data-.
The clock on the JCAT XE USB card uses external power as well.
The sound changes quite audibly with the power connected (in a positive way). Clarity, dynamics, Depth of stage.
It's a pretty big difference.

 

[theveterans]

There is absolutely no upside to using ASIO over ALSA. There are downsides though: ASIO drivers are closed source and written by 3rd parties whereas ALSA is free open source software.

There is one which is to use Diretta to stream from Windows PC -> LAN -> Rpi streamer

 

[fliz]

My DAC doesn't use 5V from the source.
My DIY USB cable doesn't have 5V or ground connected. Just data+ and data-.
The clock on the JCAT XE USB card uses external power as well.
The sound changes quite audibly with the power connected (in a positive way). Clarity, dynamics, Depth of stage.
It's a pretty big difference.

No ground and you still get lock? neat

What USB to I2S board is in your DAC?

USB is an asynchronous connection and the USB to I2S board has clocks on it.

here is one which is to use Diretta to stream from Windows PC -> LAN -> Rpi streamer

This only applies if you're getting your LAN involved. I just have a SSD plugged directly into my RPi5.

 

[Svperstar]

PC's are an extremely noisy source. You absolutely need a Linear ATX power supply and an externally powered (Linear powered) USB card to get a clean signal out of it.
You can 100% hear this noise on a High end system.

Note I said use a USB isolator. When I have nothing playing and I turn my estat amps to max I hear nothing. Under high GPU load, 300+ watts there is some noise you can hear if you play no music but turn the amp to the max, but the USB isolator gets rid of that.

Note I run a power conditioner and a UPS which may be overkill. Optical is always an option if you don't trust USB can be clean enough.

 

[theveterans]

This only applies if you're getting your LAN involved. I just have a SSD plugged directly into my RPi5.

I love streaming with Apple Music so LAN using Diretta is the cheapest way to get bang for banger sound in my case. Never liked to buy downloads or CD since it’s way too costly for me

 

[AlexMa]

There is absolutely no upside to using ASIO over ALSA. There are downsides though: ASIO drivers are closed source and written by 3rd parties whereas ALSA is free open source software.

ASIO and ALSA are basically incomparable levels of audio stream transmission quality. How can you compare a professional interface that is a generally accepted standard and the patchwork quilt of ALSA, to which bug patches are sewn every month and for which no one is responsible.

 

[Maxpain]

Do ASIO or ALSA change the sound in any way?

 

[fliz]

ASIO and ALSA are basically incomparable levels of audio stream transmission quality. How can you compare a professional interface that is a generally accepted standard and the patchwork quilt of ALSA, to which bug patches are sewn every month and for which no one is responsible.

I've listened to all sorts of music servers and streamers, on every OS, just like you claim to have as well. You made a decision to go with windows 7 years ago. I was testing things as recently as last year, despite being in this hobby since the early aughts. You can see I joined this forum *for enthusiasts* (not shills) in 2003.

A purpose built OS that explicitly exists ONLY for media playback is a no-brainer compared to something as bloated as windows. (and that's not even touching on the ways in which windows violates your privacy). That's what LibreELEC is. It's a Linux distribution that specifically is "Just enough OS" to run Kodi. No more, no less. Kodi itself has been in active development for over 20 years.

Trying to badmouth FOSS is the icing on the cake though. Commercial products, developed on a budget, by people who are only working on them because they're paid to be, can never compare to projects that are developed by GIANT collaborative efforts of people who are volunteering their time because they're passionate about what they're working on.

The linux foundation calculated (back in 2008) what it would have cost to commercially develop Debian and it came out to over 10 billion dollars (again, in 2008 dollars).

https://www.linuxfoundation.jp/events/2008/10/estimating-the-total-cost-of-a-linux-distribution/

Enterprise systems are most commonly built on linux and microsoft has been losing market share w/ it's inferior product for ages.

Do ASIO or ALSA change the sound in any way?

No. Both are "bit perfect" in that they are both subsystems that allow drivers to communicate directly with hardware, w/o being modified by the OS in any way. Both are extremely low latency.

ALSA is the Linux solution for this, and is native. ASIO is the Windows solution to this, and is made by third parties (not Microsoft).

 

[AlexMa]

You are so funny)) You provided a graph with the shares of different OS installed on servers, but don't you know that no system administrator would even think of installing ALSA on a server?
Linux's share on desktops is less than 5%, and many of them don't have ALSA installed either.

 

[Diet Kokaine]

Note I said use a USB isolator. When I have nothing playing and I turn my estat amps to max I hear nothing. Under high GPU load, 300+ watts there is some noise you can hear if you play no music but turn the amp to the max, but the USB isolator gets rid of that.

Note I run a power conditioner and a UPS which may be overkill. Optical is always an option if you don't trust USB can be clean enough.

USB isolator is like putting a band aid on a gaping wound. It makes very little difference (if any), at least in my application.
Once noise is added to the playback, even at the CPU or RAM level, it's almost impossible to clean the signal from that point forward. You can reclock it, which helps to a certain extent, but there is no replacement for providing clean power to these components from the get go.

Again, I'm not questioning what you're hearing in your setup. In mine, this difference is clear as day.

 

[Maxpain]

No. Both are "bit perfect" in that they are both subsystems that allow drivers to communicate directly with hardware, w/o being modified by the OS in any way. Both are extremely low latency.

ALSA is the Linux solution for this, and is native. ASIO is the Windows solution to this, and is made by third parties (not Microsoft).

That is my point...ASIO and ALSA both do the same thing...they do not ''sound'' better or worse unless the user (or the dev) configure them improperly.

ASIO drivers offered by the manufacturers yes...ALSA is native on linux and in any case manufacturers rely on the usb 2 generic driver to talk to linux devices....windows needs their ''special'' drivers...but this is because manufacturers do not want to invest in linux because of the small market share and the plethora of the distros as I understand it.


Also because it was mentioned above....the vast majority of data servers on the world is running linux....and every streamer out there...(it is a dedicated server too) uses linux...not one of them uses windows because even if they wanted they could not. It is also in their benefit to use an open source software and code on it....

FOSS is a different beast in general....to say the least

Windows offers an easy solution for desktop only platforms...with certain software but the real difference in sound quality do not come from software but from hardware implementation of the streamer/desktop etc

To be fair ALSA is not easy to configure even with all the documentation...but to be more fair....ASIO is not configurable at all of the manufacturer does not offer a solution like that..or your software (foobar, roon or whatever...)

 

[Maxpain]

I didn't know about Oppo using windows platform...it must been a pain in the ass to do so and make it work..

I have tried over the past on my previous windows desktops software that claims to give you better sound quality by killing processes and minimizing windows tasks...never did anything for me....not I ever could tell a difference from ASIO vs ALSA.

The only thing that made an audible difference was going from desktop/laptop to a dedicated streamer (zen stream) with a nice psu. The zen stream uses a custom volumio I think. I would have gone for re rpi soluton too but I do not have the time to research and built it.

As for windows loaded on a streamer...it has more complications other than making it work with that hardware....there are copyright stuff and security holes everywhere....

personally I use either Linux or macos devices...I will never go back to windows...not for the audio...but for the privacy invasion...though this has nothing to do with audio.