Jump to content
IGNORED

Windows/Apple better than Linux for sound quality?


ggking7

Recommended Posts

There is actually a reason. Windoze and Linux both have the task of writing operating systems for pretty much all available hardware. Apple only writes operating system code for its own hardware, so there is much more opportunity to optimize things and anticipate what is needed.

 

Note that this is not the same thing as saying Apple OS X sounds better. I simply have no idea.

 

On the other hand, there is a large and active open source community that maintains linux, so if a problem does arise, they can deal with it immediately and openly. This is not the case for Windoze and OS X.

 

Link to comment

Hi wgscott,

 

You'd be surprised, but Apple actually use Intel's reference designs for all their Mac computers, and whatever optimizations they put in are typically geared towards optimizing the user experience (GUI, disk performance, etc). This doesn't equate to better sound quality, and usually actually compromises it because all the other computer subsystems get higher priority than the sound drivers and hardware.

 

To their credit (and I’m hardly the guy to praise Microsoft), Microsoft have gone to great lengths to improve the sound quality in Vista and even more so in Windows 7. Unlike Apple, they’re actually trying. Admittedly Apple's use of a Unix-based OS bore a significant advantage in SQ over early Windows, however parity was achieved somewhere around Vista SP1.

 

Both Windows and OSX use the microkernel system architecture where all the hardware drivers sit outside the core of the operating system. This certainly has many benefits as far as system stability and compatibility, but it isn't ideal for sound quality because the computer circuitry has to perform many context switches between things inside and outside the kernel.

 

linux, on the other hand, uses a monolithic kernel architecture. Carefully tuned and executed, this architecture allows for better realtime processing and fewer interruptions to the flow of the bits – and because it's linux, you get to choose what (and how) you optimize. It also means the entire kernel can blow up in your face, but that's what testing is for and music servers should operate like toasters – with minimum interference as they go about their processing.

 

Apologies for the geek-speek :-).

 

Enjoy the music,

 

Matan

 

 

Link to comment

Context switches between core kernel and drivers are irrelevant, as long as the data reaches the device in time to prevent audio dropouts. OS X is just as capable of doing this as Linux.

 

nigel[br]ALAC stored on Drobo -> Mac Mini -> iTunes -> Airport Express (1st gen) -> Monoprice toslink -> NAD M2 Direct Digital Amplifier -> Wilson Benesch Curve

Link to comment

 

matan: "..Both Windows and OSX use the microkernel system architecture where all the hardware drivers sit outside the core of the operating system..

 

 

 

No, neither Mac OS X or Windows have a microkernel architecture. At the time when Windows NT was first released the idea of 'microkernels' was fashionable, and so Microsoft described Windows NT as having a 'microkernel architure' for marketing reasons.

 

 

 

Mac OS X may be more plausibly described as having a microkernel architecture in that it has 'microkernel bits' in it. There is a part of Mac OS X that uses the Mach microkernel, but mainly it is based on BSD Unix-like operating system. But most of what Mac OS X does is in the BSD parts, and so it isn't really a microkernel.

 

 

 

All this discussion is totally irrelevant to the question that was originally asked, ie 'Is there any reason to think Windows or Apple might be better for sound quality than Linux?'. In my opinion my answer of 'No' is more accurate than all the other long winded descriptions of complicated operating system things. The real issue is whether or not is Linux more suitable for delivering appliances that 'just work' for non-technical people to use, such as the Sonore box recently reviewed on this site. I believe it is, and that is the real issue at hand, not all this 'microkernel' stuff.

 

 

System (i): Stack Audio Link > 2Qute+MCRU psu; Gyrodec/SME V/Ortofon 2M Black/EAT E-Glo Petit/Magnum Dynalab FT101A) > Glow Amp One > Klipsch RP-600M/REL T5x subs

System (ii): Allo USB Signature > Bel Canto uLink+AQVOX psu > Chord Hugo > Tandy LX5/REL Tzero v3 subs

System (iii) KEF LS50W/KEF R400b subs

 

Link to comment

I liked your answer Richard. Occasionally two (or three) letters is all you need.

 

There are issues like driver availability with Linux but fundamentally I agree with "No" as the correct answer.

 

Eloise

 

Eloise

---

...in my opinion / experience...

While I agree "Everything may matter" working out what actually affects the sound is a trickier thing.

And I agree "Trust your ears" but equally don't allow them to fool you - trust them with a bit of skepticism.

keep your mind open... But mind your brain doesn't fall out.

Link to comment

There is a part of Mac OS X that uses the Mach microkernel, but mainly it is based on BSD Unix-like operating system. But most of what Mac OS X does is in the BSD parts, and so it isn't really a microkernel.

 

The whole idea of a microkernel is that most of the (higher-level) operating stuff isn't within the kernel code itself. So you are pretty much saying it isn't a microkernel architecture because it is.

 

 

 

Link to comment

 

wgscott: "..The whole idea of a microkernel is that most of the (higher-level) operating stuff isn't within the kernel code itself. So you are pretty much saying it isn't a microkernel architecture because it is.."

 

 

 

No the code that Mac OS X uses from FreeBSD is kernel code, it isn't userland stuff. Maybe the best way to think of how Mac OS X works is to think that it has two kernels, the Mach one, and the BSD one. The two kernels communicate via Mach ipc.

 

 

 

You could argue that this is less efficient that how the Linux kernel works, but in my opinion for the purposes of using one or the other for driving a DAC this discussion is irrelevant, and it isn't helping the guy who posted the question that started this thread in the first place. He asked a simple question about whether Linux works well for music, and probably didn't expect to have to understand what a microkernel is.

 

 

System (i): Stack Audio Link > 2Qute+MCRU psu; Gyrodec/SME V/Ortofon 2M Black/EAT E-Glo Petit/Magnum Dynalab FT101A) > Glow Amp One > Klipsch RP-600M/REL T5x subs

System (ii): Allo USB Signature > Bel Canto uLink+AQVOX psu > Chord Hugo > Tandy LX5/REL Tzero v3 subs

System (iii) KEF LS50W/KEF R400b subs

 

Link to comment

He asked a simple question about whether Linux works well for music, and probably didn't expect to have to understand what a microkernel is.

 

You gave a simple answer: "No."

 

I gave a literal answer to the "is there any reason to think" part of the question, and therefore "yes", followed by a reason that is certainly valid for video drivers (for example) and then I clearly stated that I had no idea whether this would actually translate into an audible improvement.

 

I did not bring up any operating system details. Others, including you, did.

 

However, I do know a wee little bit about the operating system, and the suggestion that "the best way to think of how Mac OS X works is to think that it has two kernels" is, to put this as neutrally as possible, really odd.

 

If however you were to write this as "the best way to think of how xnu works is to think that it is a microkernel with a closely associated set of drivers or BSD-like kernel extensions" I think is would be less confusing.

 

 

 

Link to comment

 

wgscott: "..However, I do know a wee little bit about the operating system, and the suggestion that "the best way to think of how Mac OS X works is to think that it has two kernels" is, to put this as neutrally as possible, really odd.."

 

 

 

It may be odd, but that is how I personally think about how NeXTStep or Mac OS X work. I worked for many years as a NeXTStep/Objective-C programmer, and now I am am now a Linux programmer. So I'm pretty sure I have pretty good understanding of the basics of how both OSs work, and their relative merits.

 

 

 

When you make a system call on Mac OS X it goes via Mach ipc from BSD to the Mach kernel. If you make a system call on Linux, there isn't that level of indirection. However, for the purposes of using the OS for music, I don't think that this matters.

 

 

 

 

System (i): Stack Audio Link > 2Qute+MCRU psu; Gyrodec/SME V/Ortofon 2M Black/EAT E-Glo Petit/Magnum Dynalab FT101A) > Glow Amp One > Klipsch RP-600M/REL T5x subs

System (ii): Allo USB Signature > Bel Canto uLink+AQVOX psu > Chord Hugo > Tandy LX5/REL Tzero v3 subs

System (iii) KEF LS50W/KEF R400b subs

 

Link to comment

OK, you definitely know way more than I do, but that way of phrasing it was just confusing to me.

 

Since you've worked on both, can you tell me how best to implement the following test:

 

I would like to compare sound from OS X vs. linux on my Mac mini. Would installing, for example, the stock Ubuntu on the mini permit me to test this in a reasonably simple way? Would I have to recompile the kernel, or would it be able to deal with the apple hardware by default? Would there be an advantage to using the 64-bit version (I boot into the 64-bit kernel for OS X on this machine)?

 

Link to comment

I wonder what the mac mini would sound like with vortexbox linux installed on it:) I have not heard of anyone who did it, but I'm willing to help out if you want to try it. The iso for vortexbox is a free download from vortexbox.org

 

Jesus R

www.sonore.us

 

Link to comment

 

"..OK, you definitely know way more than I do.."

 

 

 

My apologies for 'pulling rank' on the topic of microkernels. I might have been a programmer using Unix, NeXTStep and Linux for a long time, but that doesn't make me any sort of kernel programmer or music OS expert. However, I certainly do know enough to know what I don't know. As this site is an audiophile site, it is very difficult to know what sort of level of systems programming understanding people have when we are talking about how suitable different operating systems are for implementing solutions for music applications. There are many reasons why Linux or Mac OS X might be better for music applications, and I think I can say pretty categorically that whether or not the OS is a microkernel is not one of the important considerations.

 

 

 

I am a relative 'newbie' to computer audio with only six months experience since getting my first USB DAC. I am only just discovering that issues like power supply quality (linear vs switching) are important, and so on. I have a Macbook and a HP Netbook running Linux. The Macbook sounds much better than the Netbook, but I don't think that is to do with Mac OS X sounding better that Linux. Maybe it is to do with the fact that my Macbook has a firewire port for my hard disk, whereas the Netbook only has USB ports and the USB bus must be shared with the DAC. Maybe the Macbook has a better power supply - I don't know yet.

 

 

 

For work I have an iMac and run Kubuntu under vmware, and it works very well. You could do the same on you Mac Mini. Vmware costs about 80 dollars, but you can get virtual box from Sun/Oracle for free that would allow you to run Linux under Mac OS X and experiment. Or you could use Boot Camp to install Ubuntu on your Mac Mini and dual boot. I think the client/server design of the MPD (music player daemon), is superior to how iTunes and the other Mac music players work on Mac OS X.

 

 

 

I would be surprised if a 64-bit kernel made a difference for reproducing music as an entire CD is only 600 Mb which fits easily into the address space of a 32 bit computer. But as a computer audio newbie, I expect surprises..

 

 

System (i): Stack Audio Link > 2Qute+MCRU psu; Gyrodec/SME V/Ortofon 2M Black/EAT E-Glo Petit/Magnum Dynalab FT101A) > Glow Amp One > Klipsch RP-600M/REL T5x subs

System (ii): Allo USB Signature > Bel Canto uLink+AQVOX psu > Chord Hugo > Tandy LX5/REL Tzero v3 subs

System (iii) KEF LS50W/KEF R400b subs

 

Link to comment

No problem; pull away.

 

These things always seem to happen with internet exchanges. I blame ascii-text autism. I think we are in complete agreement that the kernel architecture could matter. Maybe it does, but it is certainly beyond my comprehension. I find CoreAudio a lot weirder and harder to understand than most things OS X. BTW, you have one month up on me wrt computer audio.

 

I have VMware (academic discount made it worth it for children's games that only run in windoze and OS 9 and/or PPC), and have Xubuntu installed on that at the moment (the innards are all the same). It sounds terrible running in a virtual machine, but I think I would have to put Ubuntu on a separate partition and boot into it for a fair comparison.

 

Do you happen to know anything about afplay? Unlike every other player (iTunes, Play, etc) for OS X, it is 64-bit, so I reckon that should be a test of whether the 64-bit kernel is in any way superior. (Many people here believe that it sounds better, even with 32-bit iTunes, which seems odd to me.) I just left my kernel in 64-bit when I was trying to investigate a kernel panic I was having with the 2010 mini. It stopped panicking, so I have been loathe to change it back.

 

 

 

Link to comment

 

wgscott: "..I have VMware (academic discount made it worth it for children's games that only run in windoze and OS 9 and/or PPC), and have Xubuntu installed on that at the moment (the innards are all the same). It sounds terrible running in a virtual machine, but I think I would have to put Ubuntu on a separate partition and boot into it for a fair comparison.."

 

 

 

That's another interesting consideration - do operating systems running under virtual machines sound worse than they do natively?

 

 

 

wgscott: "..Do you happen to know anything about afplay?.."

 

 

 

No, I've never heard of it.

 

 

System (i): Stack Audio Link > 2Qute+MCRU psu; Gyrodec/SME V/Ortofon 2M Black/EAT E-Glo Petit/Magnum Dynalab FT101A) > Glow Amp One > Klipsch RP-600M/REL T5x subs

System (ii): Allo USB Signature > Bel Canto uLink+AQVOX psu > Chord Hugo > Tandy LX5/REL Tzero v3 subs

System (iii) KEF LS50W/KEF R400b subs

 

Link to comment

" ... do operating systems running under virtual machines sound worse than they do natively?"

 

I think the answer to that is definitely yes, but I think it is because a more limited set of resources (memory, CPU) is allocated to the processes running in a virtual machine, and also there is another abstraction layer. Video is similarly degraded, FWIW.

 

afplay lives in /usr/bin/afplay on OS X. The command

 

afplay -q 1 musicfile.m4a

 

plays the music file given as the argument, and the -q flag dictates quality (1 being full quality). It is 64-bit on 10.6. I can't find any other information apart from the essentially worthless man page. But to my ears, it sounds as good as anything else, and presumably is a simple component of CoreAudio (to which it links).

 

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now


×
×
  • Create New...