Head-Fi.org › Forums › Equipment Forums › Computer Audio › Windows vs Mac vs Linux
New Posts  All Forums:Forum Nav:

Windows vs Mac vs Linux - Page 2

post #16 of 98
Not to prolong the agony, but I'm on all three platforms and I agree with the posters who say a Windows install - properly tended - should not produce the kind of errors '123456' is seeing.

I don't run a resident anti-virus program either; I use clam once a week overnight, along with Spybot and another anti-malware program.

I also use the jptools registry cleaner.

Most importantly, I'm very conservative about where I go, what I look at and what attachments I open. And in six years, I have never been virused on an xp install.

Scott Atkinson
Watertown NY
post #17 of 98
I too have never had these types of problems with Windows. Yeah, it slows down after a while and can become clunky (for lack of a better word), but properly maintained, it's prefectly adequate for nearly all intents and purposes.

I won't try and drive you away from Linux though Linux is very nice if you like to tinker with things, and like to read. More than half the time, installing something isn't as easy as downloading a .exe file and double clicking, answering yes to some questions, agreeing to a contract, yes to install location, and waiting. Newer distros like ones from Debian and Red Hat (not to mention their branches Ubuntu and Fedora, respectively), have made great strides in ease of use, but there's still times where you'll have no clue what you're doing.

Installing Linux isn't hard. It's pretty intuitive if you know basic computer jargon. Once you get it installed, you'll have to install audio codecs. This is usually an ALSA (Advanced Linux Sound Architecture) plug-in called Gstreamer. For a Debian/Ubuntu system, you'd open up a terminal and as root, type

apt-get install gstremer

After it's done instlalling, you'll have to configure ALSA to use your sound card. Still as root,

alsaconf

After following the prompts and choosing your card, the volume will be at it's lowest by default. Either as root or as a normal user, type

alsamixer

Adjust your volumes here and then start using your audio player. Ubuntu and Debian come default with Rhythmbox. It's a very easy to use, easy to configure audio player, but it's very basic. It doesn't even have an EQ. Amarok is prettier, still damn easy to use and has some nicer, more polished features. If you're usually a FooBar2K user, than you'll appreciate XMMS.

Learning Linux takes work, so if you're not afraid to learn, then Linux can be very rewarding and the limits to what you can do are endless. If you're impatient, and have the money, than a Mac is for you.
post #18 of 98
i'd recommend osx though i will say it's not clearcut better. there are some people i know who just don't really like the interface. it's different and they're not used to it and it doesn't really offer obvious advantages. i've been using both my whole life (well not osx my whole life... but apple os) and i've found that osx is a lot more dependent on keyboard shortcuts for some reason. if i unplug my keyboard it really messes me up where as it's no so much of a problem with windows.

anyway you shouldn't be having as many problems as you are with windows it's really not as bad as people say. you still get freezes and stuff with osx so yeah something to think about.

and to the person wondering about the java thing and how you'll have to turn it off on osx that's not true. viruses are machine specific because they have to deal with different holes in security so viruses that work on macs work on macs only and vice versa. so you don't need to worry about those.

as for linux... i've been using that in some of my classes and (depending on the distro of course) it's pretty nice too. kinda reminded me more of windows than osx though it's based on unix similarly to osx.

anyway most of that was probably useless to you, so here's my recommendation: osx
post #19 of 98
It really doesn't matter, any of the three can do equally well. It's the same data in any case; the OS only needs to not muddy the sound or video along its route to output.

In Windows you do need to bypass the kmixer for good audio though, whereas this isn't a concern in OSX / Linux. It's a pretty simple and one-time thing though, not something to base OS choice on.

As for video, this isn't an area I'm well versed in, but I'd suggest googling the terms postprocessing, upscaling and ffdshow. As I understand it, many videophiles prefer windows due to ffdshow, a freeware program with many postprocessing options. Some of the before/after screens I've seen are eye-opening.

If you're having issues with malicious software in Windows, I'd suggest reading up on basic computer security. It's likely you have many online habits which are at odds with good security. I have a/v software as well as spyware scanners, and I occasionally do online scans to see if the resident software I have missed anything. I've never found anything though.
post #20 of 98
Quote:
Originally Posted by Gurck View Post
As for video, this isn't an area I'm well versed in, but I'd suggest googling the terms postprocessing, upscaling and ffdshow. As I understand it, many videophiles prefer windows due to ffdshow, a freeware program with many postprocessing options. Some of the before/after screens I've seen are eye-opening
ffdshow is ffmpeg for DirectShow. ffmpeg is available on Linux and the Mac, and works behind the scenes in VLC and mplayer, both of which can enable postprocessing. On Linux, I use mplayer with the option '-vf pp=ac', which enables full-quality deblocking and deringing. In VLC, if I remember correctly, it's a checkbox somewhere. To get cleaner video scaling without melting my CPU, I use '-vo gl:yuv=3:lscale=1:cscale=1', but that may not work on all video cards.

Quicktime also has its own postprocessing filters, though you do have to find and enable the option.
post #21 of 98
Quote:
Originally Posted by LnxPrgr3 View Post
Straight ALSA will mix. Is PulseAudio any better at it? It would be the first audio server I've seen that doesn't completely suck at resampling.
Yes, sort of. libsamplerate, using Secret Rabbit Code, can do very good sync resampling. Technically, you can use libsamplerate with just ALSA, but ALSA configuration is a PITA. ALSA is good at talking to hardware, but not use by applications.

With Pulse, you set it to use SRC's sinc of medium or best, a default sample rate, and there you go. One problem is that it uses a good bit of CPU (does resampling per stream, rather than mixing streams of the same rate first, and I don't know that SRC is very optimized), but I've had no major problems with my C2D. It replaces the 'default' ALSA device. You then also get things like multiplexing easily enough, and also per-stream volume controls.

Quote:
Originally Posted by user123456 View Post
The problem is windows, period.
Windows may not be the most secure platform, but users help.
post #22 of 98
Quote:
Originally Posted by cerbie View Post
Yes, sort of. libsamplerate, using Secret Rabbit Code, can do very good sync resampling. Technically, you can use libsamplerate with just ALSA, but ALSA configuration is a PITA.
Once you figure it out, it's not too painful -- I followed the advice in the docs that came with the plugin.

Unfortunately, I can't make it work with dmix, and I don't know why; it also breaks flash, again for reasons unknown. Reading on PulseAudio, it seems its users run into their own problems (like Flash crashing every 2-3 page loads).

Really, I wish two things would happen with Linux audio:
  1. I wish we would standardize on one audio system, whatever it is, and
  2. I wish the developers would focus on correcting the reasons other people want to build 300 different audio systems for Linux.
There's no reason ALSA couldn't be made to do everything PulseAudio does -- ALSA just doesn't bother, so we've grown wrapper after wrapper to work around it.

If we picked one audio system, and made it work, people like the Linux Flash developers might be able to learn how to use it, work out their bugs, and maybe release a stable plugin. Until this happens though, Linux audio will be a mess of tradeoffs and compromises. PulseAudio's existence is an indication we're still headed the wrong direction -- making yet another solution rather than fixing one of the existing ones. Ideally, we wouldn't be able to have a conversation about PulseAudio vs. Jack vs. straight ALSA -- we'd have one way of doing everything, and it'd actually work, and we could just listen to music

Quote:
Originally Posted by cerbie View Post
With Pulse, you set it to use SRC's sinc of medium or best, a default sample rate, and there you go. One problem is that it uses a good bit of CPU (does resampling per stream, rather than mixing streams of the same rate first, and I don't know that SRC is very optimized), but I've had no major problems with my C2D.
Athlon X2 4200+, audacious + libsamplerate: 64% of one core used
Core2 Duo 2.0GHz, iTunes + Leopard CoreAudio's resampler: 5% of one core used

I did a quick comparison, and CoreAudio's resampler in Leopard (not in Tiger!) sounds just as good to me. It also measures pretty well, according to http://src.infinitewave.ca/.

There is definitely room for improvement, performance-wise
post #23 of 98
Why won't they develop a standard interface so all the apps can get a hold of the audio system in the same way? Seems they are just hurting themselves by reinventing all the time. I seriously wish the devs luck. I love open source. I make a living off an open source project but the time wasted on these niggling issues scares off more people than it should.
post #24 of 98
Windows is best. Go with it and just learn to use it.
post #25 of 98
Windows for one reason:

Foobar2000.
post #26 of 98
its all the same =/
all of them break!
none provides enough convenience or obvious benefits to be the clear choice.
post #27 of 98
Quote:
Originally Posted by LnxPrgr3 View Post
Once you figure it out, it's not too painful -- I followed the advice in the docs that came with the plugin.

Unfortunately, I can't make it work with dmix, and I don't know why; it also breaks flash, again for reasons unknown. Reading on PulseAudio, it seems its users run into their own problems (like Flash crashing every 2-3 page loads).
*shrug* I switched to gnash, and had no problems. Flash is a good reason to dislike binaries. I never got resampling working just from ALSA.

Quote:
Really, I wish two things would happen with Linux audio:
  1. I wish we would standardize on one audio system, whatever it is, and
  2. I wish the developers would focus on correcting the reasons other people want to build 300 different audio systems for Linux.
There's no reason ALSA couldn't be made to do everything PulseAudio does -- ALSA just doesn't bother, so we've grown wrapper after wrapper to work around it.
Actually, there are very good reasons. The problem there is that there have been no good reasons that things like ESD, Gstreamer, etc. haven't done everything it does on the desktop, out of the box. ALSA is good at dealing with hardware, and nothing else. That it got to be what applications talk to directly is bad. It should be what sound servers talk to.

It's getting better, but the way ALSA came about and matured has put Linux back quite a ways in audio.

Quote:
If we picked one audio system, and made it work, people like the Linux Flash developers might be able to learn how to use it, work out their bugs, and maybe release a stable plugin. Until this happens though, Linux audio will be a mess of tradeoffs and compromises. PulseAudio's existence is an indication we're still headed the wrong direction -- making yet another solution rather than fixing one of the existing ones. Ideally, we wouldn't be able to have a conversation about PulseAudio vs. Jack vs. straight ALSA -- we'd have one way of doing everything, and it'd actually work, and we could just listen to music
Except for Flash, you have a good point. Flash's problems are Adobe, and Adobe only. They could easily make it work through Gstreamer, or use ALSA directly. However, PulseAudio is the right direction. Have an interface to talk to that abstracts the sound device, rather than talking straight to the sound drivers. It's probably not the best implementation, of course, but that's normal .

Quote:
I did a quick comparison, and CoreAudio's resampler in Leopard (not in Tiger!) sounds just as good to me. It also measures pretty well, according to SRC Comparisons.

There is definitely room for improvement, performance-wise
Much. I recall the Windows FB2K SRC plugin not using more than 20% or so on my old AXP, where my C2D uses 10-15% per audio sending process (I use sinc medium, as I can't tell it from best on anything). For an older Linux desktop, it could pose a real problem.

Quote:
Originally Posted by dvessel View Post
Why won't they develop a standard interface so all the apps can get a hold of the audio system in the same way? Seems they are just hurting themselves by reinventing all the time. I seriously wish the devs luck. I love open source. I make a living off an open source project but the time wasted on these niggling issues scares off more people than it should.
There are several, and they work well (Windows has many, too, BTW...but you only have to worry about resampling, not errors and not working). This is one way in which having money and deadlines involved can actually help, as opposed to the rolling unpredictable FOSS development that is commonly seen. Multiple interfaces and sound servers aren't a terrible thing, but they need to all work together somehow, and they haven't been until very recently.
post #28 of 98
I have been using Windows since 3.1 and last night, I installed Ubuntu onto an old PC - it was surprisingly easy, arguably easier than Windows.

Since I only have one monitor at the moment, it is abit difficult to set up and configure the system, but I sort of intend to use it as an media player in my bedroom. Hopefully it won't be too difficult to get the codecs installed - I'm new to linux!
post #29 of 98
Quote:
Originally Posted by Jubei View Post
....Hopefully it won't be too difficult to get the codecs installed - I'm new to linux!
Right after a new Ubuntu install, the first thing I do is add the Medibuntu repos:
Medibuntu :: Multimedia, Entertainment & Distractions In Ubuntu

Basically, this allows you to install all the stuff that Canonical can't bundle into Ubuntu, for legal reasons. ie. DVD playback, W32 codecs like WMA, Adobe Reader, Flash and SUN Java.

After that, I suggest adding OGG Vorbis ("vorbis-tools" package) and FLAC support ("flac" package), if that's your thing. As for a media player, there's a LOT of them. I use Amarok. myself ("amarok" package).
post #30 of 98
Quote:
Originally Posted by cerbie View Post
PulseAudio is the right direction. Have an interface to talk to that abstracts the sound device, rather than talking straight to the sound drivers.
A sound driver is an abstraction of the audio hardware

PulseAudio certainly is neat. If the open source crowd settled on it, I'd be ecstatic -- we could stop inventing new sound server architectures every week, PulseAudio could mature, and audio might universally "just work" on Linux. Being able to blast Britney Spears on the neighbor's speakers would be an added bonus

PulseAudio, however, is trying to clean up a mess that never should have existed, and I would be equally happy if ALSA simply got its act together. It already tries to do most of what PulseAudio does. If we just made its existing features work well and made them user friendly, all we'd lose is... the ability to blast Britney Spears on the neighbor's speakers.

Right now, though, the Mac has been by far the easiest platform I've ever used to get good, clean, no-fuss audio output. Windows only provides slightly more resistance. Linux, however, has been one wild ride. I really do hope things settle out, and some solution on Linux wins out, and we ignore anyone who tries to re-re-re-invent the wheel past that.

Quote:
Originally Posted by user123456 View Post
I know more about windows than you will do in your entire life, and I'm happy if I can find a better solution than this crappy operative-system.
I'm sorry, but you really do seem to encounter a lot of viruses. I can't say I've had the same experience.
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Computer Audio
Head-Fi.org › Forums › Equipment Forums › Computer Audio › Windows vs Mac vs Linux