You are not logged in.
Pages: 1
Topic closed
I installed mesa-git from the AUR. I used cower, and did a make install in the src directory. It installed fine. Then I tried to install oss-git, and one of the dependencies is mesa. I'm was wondering why its still asking for mesa, even though I installed mesa-git already. So I proceeded anyway, and the install stopped, saying that files already existed. They were the mesa files. Why won't it see that I have mesa installed already?
Last edited by shoober420 (2014-07-27 13:55:16)
Offline
I don't see mesa here: https://aur.archlinux.org/packages/os/oss-git/PKGBUILD
Post the exact command you run and the exact error you got.
Offline
Oh my bad, I actually did a "sudo makepkg --asroot". I got those two commands mixed up. I've been installing alot of packages and forgot which command I used. Whenever I try to do a "sudo makepkg --asroot", it says that it needs gtk2. One of gtk2's dependencies is mesa. I don't understand why it can't see that I have mesa-git installed already from AUR.
Offline
Thy do you run '--asroot' in the first place?
Run 'sudo pacman -S gtk2'. Post the output if you get any errors.
Offline
I didn't know it was such a bad idea. I just read about it, and some packages can break the system if ran as root. I hope I didn't install any packages that would break my system if I install them as root. I'll run makepkg with fakeroot instead for now on. When I try to install gtk2, it tries to install mesa with it. It says a bunch of mesa-libs exist already, and then it just stops with the install. I'm on my Windows machine now, so I can't post the exact output. I don't even have X up and running yet. I'm still in the process of installing packages.
Last edited by shoober420 (2014-07-27 14:58:38)
Offline
You can use https://wiki.archlinux.org/index.php/Pa … in_clients for command output.
https://aur.archlinux.org/packages/me/mesa-git/PKGBUILD
package_mesa-git () {
pkgdesc="an open-source implementation of the OpenGL specification"
depends=('libdrm' 'libvdpau' 'wayland' 'libxxf86vm' 'libxdamage' 'libxshmfence' 'systemd' 'elfutils' 'llvm-libs' 'libomxil-bellagio')
optdepends=('opengl-man-pages: for the OpenGL API man pages')
provides=("mesa=$(_mesaver)" 'libglapi' 'osmesa' 'libgbm' 'libgles' 'libegl' 'khrplatform-devel')I guess''mesa=<some version>' is not the same as 'mesa'.
The errors you get, are there in the form of 'FILENAME exists in Filesystem'? https://bbs.archlinux.org/viewtopic.php?id=130138
Offline
Yes, its number 2.
foobar: /path/to/file exists in filesystem
Is there a way to uninstall the packages I installed with cower? I can't find anything on the wiki about cower. It has no page for itself.
Last edited by shoober420 (2014-07-27 15:14:00)
Offline
Err, it has a man page.
cower doesn't install, pacman does. It can also remove packages.
Edit: Wait a moment, why did you install packages with 'make install'?
Maybe you can run 'make uninstall'. http://stackoverflow.com/questions/1439 … ary-in-lin
Last edited by karol (2014-07-27 15:13:48)
Offline
I see that I can run "make clean" or "make uninstall", but I'm trying to keep the latest mesa-git files. How can I make my system see that I have mesa installed? If I just uninstall the mesa-git package, then I won't be using the latest and greatest mesa files. I thought of something just as I was typing this lol. I'm going to do the "make uninstall", then install gtk2 with the official mesa package. Then, after that, I will install the mesa-git package, so it copies over the older mesa files. I'll go see if that works.
Last edited by shoober420 (2014-07-27 15:16:14)
Offline
Let's start again.
How exactly did you install mesa-git and other packages? What packages?
Offline
Okay, I did a "make clean" and "make uninstall" of mesa-git, and now gtk2 installs successfully. Everything is cool. Now I'm having trouble getting X to start. That's another thread though lol.
Offline
So you were installing manually, outside of pacman. Don't do that. If you want a git build of mesa, use the mesa-git PKGBUILD in the AUR to build pacman packages.
Online
It's not clear what he did. Earlier he wrote that he used makepkg: https://bbs.archlinux.org/viewtopic.php … 4#p1440414
Offline
Okay, I had no idea it wasn't good to install packages manually. So those tar files that PKGBUILD creates are the pacman packages? I think my system is borked. The mesa-libs and nvidia-libs are conflicting. I can't even uninstall mesa, without it depending on something. Its a mess.
Offline
Please read the man pages for pacman and makepkg and the wiki article about AUR before trying to install anything.
Offline
Okay, I think I just borked my system. I'm going to have to reinstall and not install packages manually.
Last edited by shoober420 (2014-07-27 17:00:38)
Offline
And read the wiki.
As you can see, mesa has just been updated a week ago, maybe the stock package is fresh enough? https://www.archlinux.org/packages/?name=mesa
Why do you want to use oss-git?
Last edited by karol (2014-07-27 17:31:11)
Offline
I just love using the latest and greatest. This is what turned me on about Arch Linux. Its the absolute most bleeding edge distribution. I can't think of any other Linux distro that lets you download source packages straight from the git webpage, let alone has package managers built specifically for this. cower is a great AUR package manager, I was just using it wrong. I simply didn't read the wiki on how to deal with AUR packages, and how to build and isntall them properly. Because of this, I have to reinstall Arch, which isn't a big deal, its really not that hard. I don't know why people think installing Arch is hard. The package mananger is simple too, I just didn't know how to handle AUR packages correctly.
As for oss-git, I prefer OSS over ALSA, because it sounds better. It also lets you disable resampling settings that are hidden in ALSA, which further increases sound quality. OSS seems much more, raw. Its the music, sound, games, straight to your ears with no hidden resamplers. I compared OSS and ALSA before, and OSS was for sure the winner. I have an ASUS Xonar Essence STX sound card with Sennheiser HD595s. I can for sure hear the difference between the two sound APIs. OSS is much more cleaner, and crisp sounding. It just sounds of higher quality. ALSA sound muddy and blurry. Its hard to explain, you would have to hear it for yourself on decent sound hardware.
Last edited by shoober420 (2014-07-28 03:17:52)
Offline
As for oss-git, I prefer OSS over ALSA, because it sounds better. It also lets you disable resampling settings that are hidden in ALSA, which further increases sound quality. OSS seems much more, raw. Its the music, sound, games, straight to your ears with no hidden resamplers. I compared OSS and ALSA before, and OSS was for sure the winner. I have an ASUS Xonar Essence STX sound card with Sennheiser HD595s. I can for sure hear the difference between the two sound APIs. OSS is much more cleaner, and crisp sounding. It just sounds of higher quality. ALSA sound muddy and blurry. Its hard to explain, you would have to hear it for yourself on decent sound hardware.
Can you point to those hidden resampling settings? Have you used direct communication for that comparison? What objective measurements have you done to compare ALSA and OSS?
Offline
In the audiophile music player Deadbeef, there are plugins which allow you to disable secret resamplers in both APIs. The thing is, in the osscore.conf, you can disable "cooked mode", which is OSS's secret resampler, without needing a plugin. In ALSA, you have to use some sort of plugin with your audio player to disable ALSA's secret resampler. I know Audacious has a plugin which does the same.
Besides secret resamplers, I used ALSA with the Deadbeef audio player, and watched YouTube music videos with it. I can tell you for sure, OSS sounds more clean, more clear, and more crisp sounding. ALSA sounds muddy and blurry in comparison. I have the ASUS Xonar Essence STX sound card and Sennheiser HD595 headphones, so I can instantly pick up differences in detail between them. If your an audiophile (like me), you should go with OSS.
Last edited by shoober420 (2014-07-29 04:15:55)
Offline
Since this is your thread, I doubt you will mind.
In the audiophile music player Deadbeef, there are plugins which allow you to disable secret resamplers in both APIs.
What makes deadbeef audiophile? What is your definition of audiophile? What makes deadbeef better than, say, aplay playing a wav by addressing the hardware directly?
The thing is, in the osscore.conf, you can disable "cooked mode", which is OSS's secret resampler, without needing a plugin. In ALSA, you have to use some sort of plugin with your audio player to disable ALSA's secret resampler. I know Audacious has a plugin which does the same.
Some sort of plugin? Secret resampler? Do you realize alsas source is open? Let me help you answer my previous questions.
Can you point to those hidden resampling settings?
No, you can not. It is obvious you have no idea what you are writing about.
Have you used direct communication for that comparison?
Also no, you do not realize this option exists, and if otherwise, you do not know how to use it.
In alsa one can address the hardware directly with specified settings like format, sample rate, resolution, channels number etc. The standard setup goes through direct software mixing plugin dmix however, to enable multiple sounds at once. This plugin has some limitations, like a 24 bit resolution and others. Should a resampler be used, it can be specified.
Besides secret resamplers, I used ALSA with the Deadbeef audio player, and watched YouTube music videos with it. I can tell you for sure, OSS sounds more clean, more clear, and more crisp sounding. ALSA sounds muddy and blurry in comparison.
"I can tell you for sure" that you will be laughed at calling yourself an audiophile and reasoning you compared two apis using youtube videos on such places like head-fi.org. This is subjective comparison you should be aware of.
To achieve high fidelity, not only must every element in the signal path be of high quality, but also every preceding element must have at least the same parameters/characteristics as the following one, preferably exceed them. Otherwise that particular element will dominantly decrease fidelity. Shortest path, smallest number of altering elements and so on are a given.
This means that anything else other than lossless input does not deserve the term high fidelity.
I have the ASUS Xonar Essence STX sound card and Sennheiser HD595 headphones, so I can instantly pick up differences in detail between them.
I doubt anyone had a problem reading the name of your sound card the first time. How about providing useful information about it instead? It uses TIs TPA6120A2 and Burr-Browns PCM 1792A DAC, you probably use stock operational amplifiers with it, which should be two JRC2114Ds and one LM4562.
By the way, do you realize that HD595 and HD555 use the same drivers? At least the same impedance models do.
What objective measurements have you done to compare ALSA and OSS?
Apparently none, you missed the hint about objective measurement. Listening to youtube videos or any music is not a measurement. You tested your perception at best. A [double] blind test would be more suited, since presumably you have no means of actually measuring it, even though we are talking about software differences only.
If your an audiophile (like me), you should go with OSS.
Buying expensive hardware does not make you an audiophile. Try to understand how the whole audio system and its subsystems along the signal path work. Learn at least some basic electronics, this will help you understand how OPAs work. You did not even try to use alsa to its full capabilities, alsa wiki should be a good starting point.
Should you provide something tangible for your claim, I will gladly look into it.
Offline
What makes deadbeef audiophile? What is your definition of audiophile? What makes deadbeef better than, say, aplay playing a wav by addressing the hardware directly?
Music players like Deadbeef and Audacious can access the hardware directly. Its not any different then using ossplay to play a wav. If you feel like coverting all your FLAC's and APE's to WAV, just to play them with your API's music player, go ahead. I would hope that you run aplay with the "--disable-resample" though, if you want bit-perfect playback.
Some sort of plugin? Secret resampler? Do you realize alsas source is open?
Deadbeef and Audacious have plugins available to disable the resamplers in ALSA and OSS. In Deadbeef, you simply go to the plugin options and check the box.
Can you point to those hidden resampling settings?
Since the plugins have an option to disable the resamplers, that proves that there are hidden resamplers. Even more so, that aplay has a command line parameter called "--disable-resample". Alot of music players simply ignore this option, and your music gets resampled with you knowing, hence the term hidden resampler.
Have you used direct communication for that comparison?
Yes, it automatically uses direct communication when you use the ALSA plugin with Deadbeef.
"I can tell you for sure" that you will be laughed at calling yourself an audiophile and reasoning you compared two apis using youtube videos on such places like head-fi.org. This is subjective comparison you should be aware of.
To achieve high fidelity, not only must every element in the signal path be of high quality, but also every preceding element must have at least the same parameters/characteristics as the following one, preferably exceed them. Otherwise that particular element will dominantly decrease fidelity. Shortest path, smallest number of altering elements and so on are a given.
This means that anything else other than lossless input does not deserve the term high fidelity.
You purposely ignored the fact that I said I also compared using the Deadbeef audio player. I used redbook FLAC's and even 24-bit vinyl rip FLAC's. I would consider that high fidelity, wouldn't you?
I doubt anyone had a problem reading the name of your sound card the first time. How about providing useful information about it instead? It uses TIs TPA6120A2 and Burr-Browns PCM 1792A DAC, you probably use stock operational amplifiers with it, which should be two JRC2114Ds and one LM4562.
By the way, do you realize that HD595 and HD555 use the same drivers? At least the same impedance models do.
Well, you didn't ask what DAC it used, so why would I even bother mentioning it? Let alone, if you cared so much, you would just simply look that up yourself like you did already. I also do know that the HD595 and HD555 use the same drivers, you just have to remove some sort of foam inside the headphones to achieve the same sound. I personally, don't feel like opening up my headphones, so I spent the little extra and got the HD595s.
Apparently none, you missed the hint about objective measurement. Listening to youtube videos or any music is not a measurement. You tested your perception at best. A [double] blind test would be more suited, since presumably you have no means of actually measuring it, even though we are talking about software differences only.
When people do a blind test, guess what they listen too. Music. I did the exact same thing, except no blind test. I don't need to do a blind test, I can trust my ears. I know for a fact OSS sounds more cleaner and clear then ALSA does, or else I would have stuck with ALSA now wouldn't I?
Buying expensive hardware does not make you an audiophile. Try to understand how the whole audio system and its subsystems along the signal path work. Learn at least some basic electronics, this will help you understand how OPAs work. You did not even try to use alsa to its full capabilities, alsa wiki should be a good starting point.
Actually, buying expensive hardware is one of the things that makes you an audiophile. Not believing this is ridiculous. Just because you know the specs of a sound card, and a little but about a sound API, this won't make your music sound better if your still stuck with an on board sound card.
No, you can not. It is obvious you have no idea what you are writing about.
With all that arrogance, and "know it all" attitude, you sure don't really know alot about sound APIs. As any true audiophile would compare OSS and ALSA themselves, which you seem to have not done. With how big your ego is, I would hope you have a decent audio setup. What sound card do you have? Do you got an external DAC with tubes? If not, I wouldn't go around talking and acting like I do. Okay hot shot?
Last edited by shoober420 (2014-07-30 04:06:56)
Offline
Offline
Pages: 1
Topic closed