Maybe it is also the wrong address here, since it is the distribution of your choice what implements the packages. You could package the unstable driver and provide it to the developers of your distribution.
There are distributions out there with tools that are part of the default package management system to provide unstable stuff easily. But the more unstable you go, the more likely something breaks.
There is a project called oswatershed. It creates statistics on how close distributions are to the actual version of the software...
I'm sure it will be released as soon as the developers think it's ready. That's typically the answer you get in the open source world. Actually, that's typically the response you get from any developer (in the open source or closed source world) that you aren't directly paying. It is also the only answer I've ever seen given by anyone associated with this specific project.
Does anyone actually use something "offically released" in this space? I mean even ditsros use "unreleased" stuff a lot from mesa to drm. People shouldn't be that concerned about stable releases. Consider this gfx "stuff" to be more of a rolling thing like most of the time newer is better. Sometimes things get broken but let distros handle it and let them pick up the usable pieces for you, if you're so worried of breakage.
I guess I still don't understand the problem. The first discussion of this was in late Feb, when agd5f mentioned to Phoronix that it was time for an -ati release now 2.6.33 had locked down, and that he hoped to get to it within a week. The first RC came out a few days later, and a second RC is out now.
Is the issue just that agd5f didn't spell out that there would be release candidates (although there usually are for major releases) and gordboy was expecting the *final* release to come out "within a week" rather than the first RC ?
The problem is that gordboy has an inflated sense of entitlement.
Free software is about giving people the code. The radeon and mesa and drm developers are doing just that.
The source is out there, and it works. If he doesn't want to use it, he can turn to his distro and use their packages. If his distro doesn't package the code, he can change distros.
In any case, it's not the developers' fault that he is refusing to use the code which is out there and which works. It's not even a case of "give me the software for free, NOW", anymore, because the software is there, and he's simply too lazy to download it and use it.
I wouldn't worry about his problems. He clearly has some kind of a mental problem and I doubt that there are very many psychologists or psychiatrists roaming these forums who are qualified to help him.
@gordyboy: take some time off whatever it is that you do during the day. See a shrink. Talk about your issues and get your head in order. Then come back and we'll be happy to have a rational discussion with you.
That one is *really* easy to answer:
Originally Posted by gordboy
IIRWIIR (It Is Ready When It Is Ready)
It will be released once it was tested and the devs consider it stable enough to call it "final" and not "might be ready but we need feedback" (that is what 'Release Candidate' in open source stands for). So if you want your package manager to offer some final version, just wait. There is not much you can do beside creating yourself a binary of the latest RC (RC2 being out), test things thoroughly and report any bugs you observe. Yes, really stabilizing things can take its time!
That sounds the most likely to me, though he still overreacted majorly.
Originally Posted by bridgman
Final version probably won't help you if your distro won't ship it. Use Arch Linux or Kubuntu and you can install this driver very easy - via AUR or PPA.
Still No Release Date, But Plenty Of Hot Air, Lies
Here's the deal, apart from anything else, I maintain lots of machines. And that means using ordinary packages, using stuff from git is simply not feasible. And frankly, the PPA is a joke.
Originally Posted by pingufunkybeat
So I am not "refusing to use code ...". It is completely the other way around. The devs are refusing to release stuff. This situation would not be tolerated anywhere else.
Of course, if the stuff is out there and works, then there could be no possible excuse to delay release. So what is it to be ?
a) the code is not ready yet, it will be released when it is ready.
b) the code is ready and the devs are refusing to release it.
all the indications are b) as far as I can see. Which makes the "Release Candidate" totally inexplicable. People will draw their own conclusions about the timing of the latest RC.
I have another observation to make about some of the elitist crap I have read here. Thanks for all the advice guys, but I really don't need any help from a bunch of lameheads who cannot even conduct a simple argument without shooting themselves in the foot.
I have been a programmer for over 30 years, and I really don't need any advice from n00bs who imagine that gentoo/arch are the zenith of linux. People who hold those views are rank amateurs, and they themselves are very often the only ones who cannot plainly see it.
Can I make a suggestion ? OK, good. If you feel compelled to add your twopence to this thread, please ensure you know what you are talking about. And if you are not directly involved in development, please refrain from the gratuitous insults. I'm quite sure the devs don't need their image tarnished any further.
Oh and one more thing guys :
Could you tell me when the promised 6.13 ddx will be released ?