My horrible experience with ATI

[bridgman]

I see. That explains it all . Perhaps your beta testers test it on Windows and not in Linux (btw, do they even download and install it?),

Actually the beta program is Linux only. The beta users report issues and improvements quite regularly, so I expect they are downloading it. There could be a secret pact between the beta testers to send us random responses without ever downloading the driver but it seems unlikely; there does seem to be a high correlation between the beta tester feedback and the new features/issues in each driver build.

and when they are okay, we get the driver for *all OS'es* (which is Windows, Linux, FreeBSD, Solaris, MacOS, Android, RHEL-4.6 ...).

The test coverage I'm talking about is primarily internal QA with coverage spread across all the supported OSes, plus some customer test results. I believe the RHEL 4.x test focus has shifted to 4.7 and 4.8 now that 4.8 has been released and support added.

Looks like your NDA doesn't allow linux beta testers to either download, or install the drivers, or perhaps to report bugs . Mmm, *leverage_test_coverage_across_all_OSes* - sounds good to me .

You may be looking at a different NDA. If you are reading the Beta Test Agreement and see problems please let me know. The beta testers communicate via an ATI/AMD internal mailing list which goes directly to developers and testers, as well as to the other beta testers.

[energyman]

Actually there is a fairly large beta test group which does try out the drivers before you see them.

and I sent a request email to become beta tester several weeks ago without a reply yet.

seriously, monthly releases are fine. monthly stable releases and bi-weekly betas would be even better.

[bridgman]

and I sent a request email to become beta tester several weeks ago without a reply yet.

Who did you send the email to ? I hope it wasn't me

PM me the details and I'll try to find out what's happening with the request.

[energyman]

Who did you send the email to ? I hope it wasn't me

PM me the details and I'll try to find out what's happening with the request.

thanks, I sent it to
[email protected]

25 days ago. I am not panicing yet - it said somewhere that it can take several weeks I panic when there wasn't a response at the end of August So you don't need to have a look at the moment.

Seriously, I don't even expect to be accepted, as a gentoo user with a pretty boring setup (3870 card, single monitor).

[mtippett]

...

1. the monthly release needs to be dropped in favor of a more rational release model, ie: release it when it works. See nVidia stable and beta release method, coupled with timely, well written patches, for an example of a method that works, is proven, and fits how most of the professional software world works.

As has been communicated elsewhere, we are on a monthly release cycle for a number of reasons. This includes at least (but not limited to) leverage of cross-platform release management, predictability for our direct customers, schedulability of new features.

2. stop releasing broken installer scripts in the package. If ATI cannot test an installer script and verify that it works for the distro, it should not be allowed to be in the package in the first place. That anyone even has to say something this basic and obvious is really astounding to me. I cannot understand how AMD/ATI expects anyone to take them seriously when the code they release in their packages, even if they do not write it, doesn't work. Why is this so hard to understand?

The installer packaging scripts included with the installer are provided by third party domain experts, if you use the installer and review the OSes supported by the packaging tools, you can see all the maintainers. If you are concerned about a problem with the packaging you can contact them. All the packagers are part of the beta program and are well and truly aware of upcoming packaging related changes on the driver side. If they are not keeping up on the distro side, then volunteer to help them.

The distributions that should work with out issue are our supported distributions - they are Ubuntu, Red Hat, SuSE and Red Flag. Other distributions are best effort and _should_ work. Even the supported distributions are targetted to have full integration _at the time of release_ of the distribution, so it may mean that karmic will struggle during the alpha's and beta's, but the formal release we should be fine.

I said about 1 year ago that I'd wait to see how ATI was doing last summer before drawing any final conclusions, but the problems all persist, ATI continues to release code that does not work, and rather than either fix it or remove it, they make excuses.

Can you provide references to what "doesn't work"? There are certain feature sets that are at a lower priority and have persistant bugs primarily because we are working on other areas.

Nobody is interested in hearing more ATI excuses I'm sad to say.

I don't believe we make excuses. We may not communicate publicly the full extent of our plans and directions.

On the bright side, for any very low end requirement, ie, server/office machine video card, etc, the cheapest ATI card you can find, plus the free radeonhd drivers, are now an acceptable option, but you are in my opinion throwing your money away if you spend more than $40 on an ATI card for Linux, not until the first two points are addressed would I consider recommending ati cards to anyone, except as basic desktop display if the mobo has no built in graphics, or if the xorg driver does an adequate job for your needs.

After watching ATI for years now, I feel fairly comfortable narrowing the problems down to the top two points, both of which to me indicate ongoing procedural errors in how ATI is developing their drivers. Excuses are not interesting to hear any longer.

Different users have different requirements, fortunately with ATI you have many options - 2 Open Source Drivers, the Proprietary driver or specs to write your own driver.

Regards,

Matthew

[bridgman]

thanks, I sent it to
[email protected]

That might be for the old Windows beta program, will check. The form looks pretty Windows-specific.

Seriously, I don't even expect to be accepted, as a gentoo user with a pretty boring setup (3870 card, single monitor).

Boring configurations are not a problem. I think Gentoo should be OK but I'm not sure; I believe we include unsupported OS users in the beta program as long as we also have a distro packager for that OS in the program as well.

[mtippett]

Why arent fglrx beta's availbale for everyone with stable releases released when they are ready/have some improvement.

I mean who cares that linux and windows drivers get released at the same time? Seems like a really bad and cheap marketing trick.

There are pragmatic reasons for the monthly releases, for the end user you have a number of options as to how to update the drivers.

• Stick with the in-distro one
• Update after reviewing the driver changes or feedback
• Update when you hit a bug that you care about

The monthly reasons do provide benefits for users in small ways. Looking at the monthly release threads on the forums, there are always "great it fixed this bug for me". Due to the structure of the teams and the commonality of code with Windows, we cannot effectively document each and every code change, so we try to summarize the issues that are known to affect end users and have a known fix.

On a different note I agree fglrx has made great progress. And for that ATI should be thanked, I think the better they get, the higher the expectations become...

Thanks, it is appreciated that this is noted

Now can we get some newer kernels supported?

The official support for kernels is more or less settling into the distribution release cycle for us. It means that we may skip restabilizing for a kernel if there are not distros targetting that kernel. For example, Karmic is targetting 2.6.31, and so our work is primarily focused on that kernel (targetted at Karmic's release).

The kernel compatability layer (KCL) is in source form, and users are able to look at the kernel functions and update as needed. If there are comments that are missing that do not communicate the intent of a function sufficiently for a developer to adapt to a new kernel, then please PM a question for a KCL function to me.

Very rarely does a new kernel trigger an architectural change in the binary part of the driver, so realistically the community is at the same level of opportunity as the developers that report to me in updating to a new kernel.

You have the kernel source, you have the KCL source. If the KCL function isn't clear, tell me and I will get the doxygen comments updated.

Regards,

Matthew

[mtippett]

That might be for the old Windows beta program, will check. The form looks pretty Windows-specific.

Boring configurations are not a problem. I think Gentoo should be OK but I'm not sure; I believe we include unsupported OS users in the beta program as long as we also have a distro packager for that OS in the program as well.

I can speak a bit more authoratively on this .

PM me your email address. In general, the beta program grows in a number of ways....

• We see you already contributing (be it directly to the driver, through supporting other users on forums, activity on the bugzilla, etc and we reach out and invite you
• You are recommended to by an existing member
• You can present a case that you will add particular value
• We like your handle .

The hardware doesn't really matter too much. What I tend to look for are individuals who are objective, structured in their thinking and presentation and reporting of issues, balanced enough to understand that your particular issues may not be resolved in a timeframe that directly suits them.

Obviously your Gentoo'ism means that there may be times that Gentoo is too far ahead of the curve and may not get value from betas on occasion.

Regards,

Matthew

[energyman]

yeah. I am using this 3870 for roughly 12 month now - and while there have been disappointments (no .29 support in 9.6. Possible to patch, but dmesg flood is not nice) I see a steady stream of improvements. I am not unhappy that I switched over.

[energyman]

I can speak a bit more authoratively on this .

PM me your email address. In general, the beta program grows in a number of ways....

• We see you already contributing (be it directly to the driver, through supporting other users on forums, activity on the bugzilla, etc and we reach out and invite you
• You are recommended to by an existing member
• You can present a case that you will add particular value
• We like your handle .

The hardware doesn't really matter too much. What I tend to look for are individuals who are objective, structured in their thinking and presentation and reporting of issues, balanced enough to understand that your particular issues may not be resolved in a timeframe that directly suits them.

Obviously your Gentoo'ism means that there may be times that Gentoo is too far ahead of the curve and may not get value from betas on occasion.

Regards,

Matthew

ok, thanks, will sent pm soon. If I do not become beta tester my world will not crumble down. I think I do understand your requirements. So if yes, I will be happy, if not, I can live with that.