A few questions pls:
1. I am using Win 7 as my music sever is it worthwhile to migrate to Linux with Ubuntu 10.10 and use it as a music server as well as transporter,will there be any sonic improvements?
or
2. Move to using squeeze touch as my transporter.
If Linux is still yr preferred platform, as I am new to Linux can you direct to a website I can learn the introduction e.g. to use the script. Thanks.
1. You can easily install Ubuntu and Windows side by side.
Though I recommend to go for Linux Mint 10, which is
based on Ubuntu. Mint comes whith certain audio codecs
pre-installed. The look and feel I really prefer over Ubuntu.
2. Installing a Linux is IMO not waste of time. I'm using it as
my base system. I do have Windows sitting in the back.
3. Sorry. But I can't help you with Linux Basics. There are hundreds
of Wikis maintained by the Ubuntu community. You'll find a
solution or at least an answer for most of the problems you'll encounter.
4. Before you start the Linux journey. Make sure your soundcard is
supported. That's IMO one of the most critical issues under Linux.
Good luck.
Though I recommend to go for Linux Mint 10, which is
based on Ubuntu. Mint comes whith certain audio codecs
pre-installed. The look and feel I really prefer over Ubuntu.
2. Installing a Linux is IMO not waste of time. I'm using it as
my base system. I do have Windows sitting in the back.
3. Sorry. But I can't help you with Linux Basics. There are hundreds
of Wikis maintained by the Ubuntu community. You'll find a
solution or at least an answer for most of the problems you'll encounter.
4. Before you start the Linux journey. Make sure your soundcard is
supported. That's IMO one of the most critical issues under Linux.
Good luck.
@phofman ( or other Alsa specialists)
are you there?
I read about ""WASAPI Event Style" as being implemented by J.River 15. The newest J.River is supposed to even outperform apps like XXHE.
The main issue behind it is that the sounddevice fully controls the dataflow. The managing and flushing of buffers etc. from the PC side becomes obsolete. ( like an extended asynchronous USB mode ???) Here we are again
"Buffer Issues". The same story continues. Obviously "audiophile" people are raving about the performance gain with WES.
Did you anything hear about it? Is there an equivalent solution/discussion in Linux/Alsa land about it???
Cheers
are you there?
I read about ""WASAPI Event Style" as being implemented by J.River 15. The newest J.River is supposed to even outperform apps like XXHE.
The main issue behind it is that the sounddevice fully controls the dataflow. The managing and flushing of buffers etc. from the PC side becomes obsolete. ( like an extended asynchronous USB mode ???) Here we are again
"Buffer Issues". The same story continues. Obviously "audiophile" people are raving about the performance gain with WES.
Did you anything hear about it? Is there an equivalent solution/discussion in Linux/Alsa land about it???
Cheers
I am not going into the buffer discussion with you again as you have refused flat to learn the principles. IMO "wasapi event style" (based on the description) or ASIO (being an equivalent non-MS technology) is how alsa operates, without any dmix layer.
Thanks for your reply.
I understand you are very busy with the Squeeze Touch mod.. I am willing to experiment using Linux to investigate whether the sound quality is superior to that offered by Win 7 using J. River's Wasabi Event(this combo does sound good to me).
Under Linux OS it still needs some "mods" to the OS in order for it to run at its optimal, those mods I am not really familiar with, besides I am not familiar with the OS. Until a "GUIDE" comes along on how to implement a "complete system" on Linux I will working in the dark. I may have to wait for you to complete your guide.
BTW I am sure you realize there many commercial systems that charge $1000's for their transport system using their s/w and h/w. I think your idea of using a modified Linux OS running off the shelf music s/w(free?) will compete with these systems in terms of sound reproduction. The complete system would costs less than $500 running under netbooks hardware. If it turns out to be true then your reputation will increase.
In the meantime I may start to install Ubuntu in my hardware and I play around with it however I will NOT get the optimal performance from it until your guide comes along.
Advise us when you have completed the guide. Cheers have a good day.
Ttan, as your first post stated, this thread has over 1700 posts, and having a noobie's linux journey starting at 1740+ is just going to add another few hundred posts which obscure some of the other issues, ie. linux audio, not just getting going on linux in general.
Soundcheck's recommendation of linux mint is good, based on my experience with ubuntu and mint. I'm using ubuntu simply because I've invested too much time into getting it running ok for me.
IMO you _have_ to get a linux pc up and running (could even be a dual boot of your current system), and get to use a couple a music players, and get reasonable audio quality out of that, before you consider putting your collection on linux, modding an SBTouch, etc.
Please consider starting your own thread, "Getting started with linux audio" or something like that, so that others may support you on your journey. Google is your friend, as are forums for ubuntu and/or mint. "Studio" editions of linux may not be for you, as they are oriented to audio/video editing.
How long did it get you to the point that you have a satisfactory W7 music server? Well, you will need some fraction of that time to get mint or ubuntu expertise to that level, whether it's 10% (you are fast!!) or 50% or 200%.
Good luck. Start with figuring out how to load mint or ubuntu without thrashing your W7. It's much easier than even 4 years ago, but to be safe and have some idea what's happening as you install takes time to search and read, and this forum will not be the place to get the help until you're fairly advanced, talking about audio formats, resampling, jitter reduction, streaming protocols, etc.
Soundcheck's recommendation of linux mint is good, based on my experience with ubuntu and mint. I'm using ubuntu simply because I've invested too much time into getting it running ok for me.
IMO you _have_ to get a linux pc up and running (could even be a dual boot of your current system), and get to use a couple a music players, and get reasonable audio quality out of that, before you consider putting your collection on linux, modding an SBTouch, etc.
Please consider starting your own thread, "Getting started with linux audio" or something like that, so that others may support you on your journey. Google is your friend, as are forums for ubuntu and/or mint. "Studio" editions of linux may not be for you, as they are oriented to audio/video editing.
How long did it get you to the point that you have a satisfactory W7 music server? Well, you will need some fraction of that time to get mint or ubuntu expertise to that level, whether it's 10% (you are fast!!) or 50% or 200%.
Good luck. Start with figuring out how to load mint or ubuntu without thrashing your W7. It's much easier than even 4 years ago, but to be safe and have some idea what's happening as you install takes time to search and read, and this forum will not be the place to get the help until you're fairly advanced, talking about audio formats, resampling, jitter reduction, streaming protocols, etc.
Skipping 1000 posts, so forgive me if this was covered. The only real-time kernel I remember was an option in True64 Unix. That was so long ago it was still Dec. That said, any stripped down Unix-like, LINUS, FreeBsd, OpenSolaris or Apple should be able to give you run priority and not the occasional hic-up you get from MS.
OpenSolaris is actually my favorite as it is more "enterprise" ready than Linux no matter what RedHat says. Being a cheap SOB, I think I will prototype on Windows and build a stripped down, no internet ubuntu for the production server. Unix-like is so much easier than Windows to strip down and control, but that is coming from an old former SA.
OpenSolaris is actually my favorite as it is more "enterprise" ready than Linux no matter what RedHat says. Being a cheap SOB, I think I will prototype on Windows and build a stripped down, no internet ubuntu for the production server. Unix-like is so much easier than Windows to strip down and control, but that is coming from an old former SA.
Here: Main Page - Linux Mint Have a Mint..
This does happen to be a "Linux Audio" thread, maybe you should read up on it.. Linux that is..
Distro watch is a good place to start.. DistroWatch.com: Put the fun back into computing. Use Linux, BSD.
Where have you been the last 4 years?
Perhaps you better read the thread first. Or you read the meanwhile
other 500Million threads and posts about OSses (beside PC-HW and apps and drivers) impacting the data stream.
For sure - at this place - you are at the wrong place to blow such a bold
statement in the air.
Cheers
Where have you been the last 4 years?
Perhaps you better read the thread first. Or you read the meanwhile
other 500Million threads and posts about OSses (beside PC-HW and apps and drivers) impacting the data stream.
For sure - at this place - you are at the wrong place to blow such a bold
statement in the air.
Cheers
Show us a single post out of those "500million" threads where the author presents a comparison test between two playback tools/OSes (does not matter) which includes the two indispensible parts - proof of bit-perfection of the chain and properly conducted listening tests. I have not seen any, just claims never supported by corresponding proofs. And a lot of shoulder tapping by followers grateful for any free work done for them. Sometimes a lot of shoulder self-tapping too.
Why don't you ask authorities who really know how the stuff works, such as Takashi Iwai, Jaroslav Kysela, Mark Brown on alsa-devel mailing list? I will not be making a fool of myself
Where have you been the last 4 years?
Dear Soundcheck,
your arrogance is annoying. It seems you want to show a knowledge on computers, operating systems (and audio in general) you actually don't have. There is no such a comparison. (dot)
The fact that you are ignorant about how Linux works emerges from your arguments about RAM loading and buffering and I truly suggest people to not follow what you wrote on this thread. Most of your arguments are just science fiction.
I use Linux mainly for scientific computing and I am not really into the computer audio, but RAM is RAM (dot). Reading your posts about your RAM player (just a simple script actually) I guess you should really go back to a good book and practice with some serious programming under Linux/Unix systems before to make (with such arrogance) silly arguments.
Best Wishes
Pietro
Next time you better skip "Dear"
How do you know? You don't know anything about my background!
Man - where have you been the last 4 years. You might have a look at what's going on different operating systems since quite some time.
Having more knowledge and experience in a certain area shouldn't be interpreted as arrogant and ignorant. Responding to st-pid and aggressive posts like yours, might make you feel that way. Then that's pretty much intended.
Before posting your nonsense better get into the subject first.
I am one of the very few ones over here who NEVER posts or claims something without giving people (e.g. with above script) the opportunity
to check it out.
All others are just blowing hot air (or their personal opinions) in the air.
(I'd rather like to see that other people get involved into a constructive process - this current discussion over here I consider destructive!)
phofman had to learn it too. ( Remembering the buffer issue on SB Touch)
Even if others confirming the changes phofman and people like you call me ignorant. You guys twisting reality!
But again: To experience those OS related improvements, the ones I am usually refering to, it requires a bit more then a Logitech PC speaker.
And you call others arrogant and ignorant. Not a clue about computer audio
and accusing me of being arrogant. Can't believe it.
Yep. It's a simple script. So what!?!? Scripting is the most
powerful way to handle a Linux system. There are thousands
of scripts used inside a Linux installation.
My script lets you very conveniently start a track through the file manager. This simple script beats everything else soundwise you'll find in Linux land (if you have a realtime kernel installed - go a couple of posts back)
I use it every day.
If you don't like it. Come up with a better one!! And we all will be happy.
But please: I invested some of my limited time to write that script.
At least that should be respected.
People like you are the reason that people loose their motivation to
release anything (free) into public.
My advise to you: First think and do some research (at least that you should
have learned during your "scientific" work on a PC), than write.
Just to wrap it up: I modified a Squeezebox Touch "realtime" Linux. I applied most of my optimizations. With those mods I even made it into "Stereophile"
magazine. I even presented those modifications on the biggest german
DIY fair 2 month back, while being part of a great highest quality audio
system (Bastanis/Silvercore).
That's it! (Once more)
Please talk for yourself. There was only one confirmation by a guy who praised your mods on this thread too, I would not dare using the word "uncritically". Just use a search function. I still stand firm that lowering DMA RAM buffer to improve sound quality is a nonsense and is in direct opposite to your effort of reducing CPU/whole device activity. Just as using RT kernels, lowering latency, using a specific bit-perfect playback tool compared to another offering the same characteristics (ecasound vs. sox vs. e.g. aplay). No reason to discuss it here again. Those who know how the chain works understand, those who just copy/paste/follow will praise you.
Presenting something on a DIY fair does not say anything about objective results of the mod. People like tinkering with their stuff and do not ask questions when getting something free they do not understand.
and so... what should we check? That your (supposed) moods sound better than anything else.
strange enough, I have the same feeling when I read your posts.
I don't have PC speakers, I think my system is acceptable and I know what I am listening... well I don't have the Batanis but ...
See you have a big Ego, and really you don't understand that what you hear on your system and what you think about your own sound is of little interest for us. I can say that my system sound much better with my personal groundbreaking moods, there would be any point in that? NOOOOO! Try to understand once in a while.
The only way to prove something is to present empirical evidence (i.e. measurements), when possible... otherwise what you think is a discovery... is your personal feeling which for several reason I don't trust... I believe you gave us proof of the fact that you don't understand how RAM and Buffering works (dot). Since you started this thread (which then became interesting thanks to other guys) you made tons of unsupported statements trying to impress people with unsupported arguments.
it's exactly the other way around. The *guru* like you discourage people to use a computer to listen to music.
Ohhhh Whaooo! Really impressed. Anyway, let me finish here (double dot) ..
Pietro
Where have you been the last 4 years?
Perhaps you better read the thread first. Or you read the meanwhile
other 500Million threads and posts about OSses (beside PC-HW and apps and drivers) impacting the data stream.
For sure - at this place - you are at the wrong place to blow such a bold
statement in the air.
Cheers
You see, I am both an audio person and an engineer. I actually know what an operating system is and does. I know what a driver and codex do. I built my first computer on a wire-wrap board 35 years ago. I make my living as a systems engineer and deal with DSP issues all the time.
The OS is only a framework that provides the ability to send a bitstream to port where a DAC can read it. It must deliver the bits in proper sequence and proper timing. That's all. If it can do that (they all can), I stand by my " ain't no way". Proof? Well, I am typing this comment. What is on the screen is exactly a representation of the bits I intended to put there. Guess even Windows works, just a lot harder to maintain than Linux.
Could an implementation of a CODEX on one OS vs. another cause an issue? Sure, but that would be a CODEX problem, not an OS. Could a specific configuration cause the OS to not service the port as required? Sure, but that is an implementation problem, not the OS. Is one OS harder to configure than the other? Absolutely. I have maintained OSF, BSD, LINUX, SunOs, Solaris, AIX, IRIX, VxWorks, Novel, and Windows. Give me a UNIX-like over Windows any day. (you are right, I never maintained HPUX)
When someone suggests digital audio does not obey the laws of physics, run, do not walk to the nearest exit. Out of the 500 million threads you reference, please point me to one that is actually objective, measurable and has technical backup why it is possible. I have both a job and a life, so I do not have time to read all 500 million subjective posts about perceptions of uncontrolled test environments.
This is huge thread with many good points and those i don't have tested but end of the thread is just flaming. Like i know things better than you. i am engineer with ... years experience. Thread title says linux audio. Maybe someone who has time and knowledge in this thread can make summary of point discussed so we can get clean start.
Fzaad,
I agree, but sometimes I don't like being attacked by someone who in their comments demonstrate they do not even know what an OS is or the basics of digital processing. I do. My apologies if my rebuttal offended you.
So, let us move to discuss inaccuracies in various CODEX implementations or DSP tools in the LINUX environment. Maybe stack buffer allocations and if it impacts timing for NFS mounted file stores in a audible way. I might suspect older implementations of RH 4.2 on a server supporting many streams could have issues. Let us leave undisclosed tweaks behind and limit discussion to objective measurements and theories that do not include magic.
I agree, but sometimes I don't like being attacked by someone who in their comments demonstrate they do not even know what an OS is or the basics of digital processing. I do. My apologies if my rebuttal offended you.
So, let us move to discuss inaccuracies in various CODEX implementations or DSP tools in the LINUX environment. Maybe stack buffer allocations and if it impacts timing for NFS mounted file stores in a audible way. I might suspect older implementations of RH 4.2 on a server supporting many streams could have issues. Let us leave undisclosed tweaks behind and limit discussion to objective measurements and theories that do not include magic.
I don't think so. I see it more as an offer. One can make use of it or leave it.
There is nothing discouraging in this. I myself find the "no matter what you do, it's all the same" style a lot more frustrating, especially when I realize, that this
is a PC Audio/Linux thread about something supposed to be a hobby.
When people don't want to try anything at all, when they completely know the truth about PC Audio, have seen it all, then why still be here?
If not, and if one is still participating in the "Linux Audio" hobby, one might want to give it a listen. So how could soundcheck discourage me? Simply saying "your work is crap, you don't have a clue and you are wrong though i never tried it because i don't need to" seems a bit thin to me, considering the amount of personal attacks that come with it for free in some cases.
I always liked the discussions between i.e. phofman and soundcheck a lot, seemed to push the thread forward though they didn't find the same opinion too often, but there is a lot of bashing from others in the meantime. The fun is fading.
just my five cents again, though.
greetings & a happy new year!
andi
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.