The New Driver Is Out! Meet Oktoberfest!

[Michael]

The New Driver Is Out! Meet Oktoberfest!

AMD 8.41 Display Driver Preview -- Our first article on this new driver are simply the technical details such as noting this driver has been actually in development for over a year, the AIGLX support that is coming next month, and other relevant information.

Read More: http://www.phoronix.com/vr.php?view=10955

ATI R300/400 Linux Performance -- We re-tested the Radeon X300 and X800XL with this new driver and it offers some HUGE performance improvements! The Mobility Radeon X300 was as much as 54 times faster compared to the previous ATI Linux driver. The improvements in these older ATI GPUs are simply astonishing.

Read More: http://www.phoronix.com/vr.php?view=10964

ATI R500 Linux Performance -- With an X1300PRO, X1800XL, X1800XT, and X1950PRO in hand we compared the ATI 8.40 and 8.41 display drivers. The performance improvements for the Radeon X1000 series range from 50% to 90% faster.

Read More: http://www.phoronix.com/vr.php?view=10970

ATI Radeon HD 2900XT Linux Performance -- Four months after the Radeon HD 2900XT was introduced there is finally proper Linux support! Not only is there support, but it offers a pretty darn good level of performance. In fact, the ATI Radeon HD 2900XT offers the best bang for the buck when it comes to high-end graphics cards. There is also more on the R600 support in general inside this article.

Read More: http://www.phoronix.com/vr.php?view=10973

AMD: Accelerating Open-Source Drivers? -- Not today, but in the near future AMD will be making an announcement about accelerating efforts within the open-source community. What does this mean? Open-Source ATI Drivers? We will tell you in due time.

Read More: http://www.phoronix.com/vr.php?view=10977


More to come later in the morning... It's been a very long day. If there are any test requests or other information you would like to know before the driver actually is downloadable, let me know.

[lenrek]

Hey... Finally something interesting to read about.

[Svartalf]

I'll say...

If Michael hadn't already named it...I still like my pet name for 'em... :D

[Svartalf]

If there are any test requests or other information you would like to know before the driver actually is downloadable, let me know.

Can we see some abuse dished out via Viewperf (Yeah, yeah, I know- it's one of the standard torture tests all the same and NVidia AND AMD/ATI use it as one of their benchmarks)?

(EDIT) Oh, can we see if the driver actually WORKS with Hypermemory machines properly in UMA, Dedicated, and UMA+Dedicated modes and what performance values we get from all three modes?

[www1862]

I have a question,what about my X1100 IGP?It belongs to which series?It seemly to be based on X300?

[hobophobe]

If there are any test requests or other information you would like to know before the driver actually is downloadable, let me know. (Emphasis added)

Does it blend? :D

The main question I have is where will AMD go from here. I'm assuming that over the next few months they'll roll out AIGLX (next month), and continue to tweak performance on some of the older cards, do maintenance on the new driver... but does 8.41 mark AMD 'caught up' to linux and able to do more in the future in terms of keeping up with kernel releases and X.org releases?

And the best answer to that I can hope for may be contained in "...what AMD will be sharing in the near future with this announcement..."

Thanks for the good news, Michael.

[hmmm]

WOW - Michael the hype leading up to it 8.41 was well worth it. I made a decision to get a HD2600XT a month ago based on your hype - thankyou and thanks ATI :D

Just have to wait a week or so...

Have there been any changes with respect to CCC and the installation procedure (other then just updating packaging scripts and a new ccc version number)?

Test request - no chance you can see if any major games [hl2, oblivion] supported by wine now "work better"?

Test request 2 - Savage 2 beta old driver vs new driver comparison

Test request 3 - comparison of windows vs linux performance with new driver in doom 3 & q4:et

[hobophobe]

I have a question,what about my X1100 IGP?It belongs to which series?It seemly to be based on X300?

You can find out via 'lspci' or else it's probably printed in your Xorg log somewhere as well (and other places too probably)...

If I do

lspci | grep ATI

it shows

01:00.0 VGA compatible controller: ATI Technologies Inc R420 JI [Radeon X800PRO]
01:00.1 Display controller: ATI Technologies Inc R420 [Radeon X800 PRO/GTO] (Secondary)

[Noneus]

Now let's all hope ATI will continue developing this new driver :) Perhaps I'll buy an ATI next year to replace my 7800GT.
One thing is bugging me though. No word about using the videoplayback acceleration for ATI cards. Someone has a clue?

[hugues]

How nice! You just made my decision easy! I was to go for Nvidia after years with ATI. I recently dropped Windows for Linux and was looking for a performance improvement...

Is there a date of availability for the new driver?

[Malikith]

At last... I can now dust off that x1600XT and retire the reliable yet slow Geforce FX 5500. By the way Michael, I knew it ;).

Originally Posted by Michael
Nope, it's not two weeks. It's ______...

Originally Posted by Malikith
If theres any meaning behind that towards the ATI drivers, count the characters haha..

_ _ _ _ _ _ . . . (9 characters)

S E P T E M B E R (9 Characters)

Hahahaha. Of course it probably doesn't mean anything . But I'm throwing out a good guess out there. Maybe that'll spark you guys up a bit. Definitely is fun. Plus my birthday is in this month so maybe that'll contribute, ATI/AMD can give me a birthday gift.

Yep, I knew it hehe, except I didn't know it was tonight the results would come. This is very exciting. I definately know what my next machine is going to have, thank you ATI/AMD.

Edit:

Oh by the way Michael, hows the anti aliasing performance now? Is it the way it should be now too?

[hobbes]

Now let's all hope ATI will continue developing this new driver :) Perhaps I'll buy an ATI next year to replace my 7800GT.
One thing is bugging me though. No word about using the videoplayback acceleration for ATI cards. Someone has a clue?

I would like to know that too. Fully Functional videoplayback acceleration it is a must for me.

Michael, can you give us a hint on that?

Thanks for the good news.

[ltmon]

Aside from performance, I am looking forward to Xrandr 1.4 and Xorg 7.3 support in general (are my version numbers right?).

As someone who insists on using Linux in the office, being able to move between projectors and secondary monitors quicker and easier would be a huge plus.

Any news on this? Or is it the same story as AIGLX?

At any case, hopefully the new code base makes it easier for the developers to keep up with new features if nothing else.

[wussboy]

Quite frankly I'm just startled you managed to get an x1950PRO running in Linux at all. Maybe I won't have to sell that stupid video card after all.

[Malikith]

Yeah, and I hope the new codebase fixes the whole AGP bus issue that plagues me and Alistair and anyone else with a AGP X1k card.

[Xipeos]

Just my luck...

My hdd failed this night, now I'm using my father's laptop (ugh...IE).

I'm delighted with the new driver (http://www.phoronix.com/scan.php?page=article&item=822&num=1) though :)

I'll be back when I get a new harddisk..

[eatmuchpie]

Hello, GREAT news for us ATI linux peeps, however I'd like to know if AIGLX is supported with this release. I looked through the articles and this thread and didn't see an answer.

Cheers guys!

EDIT. dw, just read the other article saying it isn't in the release ... shame, but now i can play games!! :D

[hmmm]

Hello, GREAT news for us ATI linux peeps, however I'd like to know if AIGLX is supported with this release. I looked through the articles and this thread and didn't see an answer.

Cheers guys!

its on the second page of the 8.41 driver preview

Where's AIGLX? Well, unfortunately due to delays later in their development cycle for the 8.41 driver, AIGLX was pushed out of this month's development and release cycle. The AIGLX support will, however, arrive next month with the fglrx 8.42 driver.

[mile]

what about "desktop performance"....with <8.41 ( specialy in KDE) flickering is very noticeable and very annoying.

Any improvements here?

[notz]

hy, what's about video acceleration? xvmc support, mpeg4, avivo, etc.

[chrisr]

in the mean time FireGL owners will want to stick to the fglrx 8.40 driver.As an owner of a laptop with a FireGL V5250, I really have no choice but to try the 8.41 driver because the 8.40 driver fails when trying to play anything more demanding than "Quake 2". Anyway, I thought that the V5250 was based on the R530. What gives?

[mlau]

Also, what about output device detection/configuration, big desktops, randr-1.2 support?

[Thunderbird]

We (wine developers) are very interested in the glxinfo output of the new ATI drivers. Before ATI didn't report various GL/GLX capabilities correctly which resulted in less features or very ugly hacks. If you could provide us with those that would be great.

[opera]

Also, what about output device detection/configuration, big desktops, randr-1.2 support?

I Very much second to get information about that.

Also, will this piece of shit natively compile on 2.6.22 on x86-64? All I could find in the info was some crap about Fedora. As if everyone in the world used fedora or their kernel packages...

I recommend Michael to start informing about LINUX KERNELS and not some specific RH/Fedora packages.

New features, sound nice, but something (history I suppose) tells me it'll suck, just like the other drivers. Especially for us who don't care about Quake FPS, but rather big desktop / RandR 1.2, new kernel support, x86-64 and other rocket science things ATI seems to find hard to acknowledge.

I'd call this "Script kiddie's FPS goes up"-edition

[Xipeos]

I Very much second to get information about that.

Also, will this piece of shit natively compile on 2.6.22 on x86-64? All I could find in the info was some crap about Fedora. As if everyone in the world used fedora or their kernel packages...

I recommend Michael to start informing about LINUX KERNELS and not some specific RH/Fedora packages.

It actually says it won't work with 2.6.23 (general information), so you can't use it on Fedora 8 (specific information). Maybe I should re-use my signature with "people thinking distros are different kinds of linux"...

New features, sound nice, but something (history I suppose) tells me it'll suck, just like the other drivers. Especially for us who don't care about Quake FPS, but rather big desktop / RandR 1.2, new kernel support, x86-64 and other rocket science things ATI seems to find hard to acknowledge.

Just wait and see. Michael said it'll be available in 1 week wtf.

I'd call this "Script kiddie's FPS goes up"-edition
Then you have no idea what script kiddie means.

[billybore]

Good Good news, very good, I hope!!

I would tell to Michael some additional informations about the new driver:

Anything about Hypermemory? I think it is important for my mobility card (I have 2Gb of RAM, and in WinVista (damn!!) with new driver with full functionality of HyperMemory my graphic performance increase a lot, there's a lot of big differences in the same games between Vista and XP, in Vista they run faster than XP).

Any improvements about graphic card power management and fan speed?

Any notice about suspend/resume functionality? (with 8.40 and F7 i can't suspend correctly)

With new driver and an ATI mobility X300 can I play with FLightgear and can I use Cedega?

Thanks in advance.

Sorry for my english.
:):):):)

[emparq]

Wow, great news (finally)! I'm more interested to read about AIGLX-support and/or any power management-related features, so if you could touch on those that'd be awesome.

Thanks!

[egrmeister]

My xv acceleration works just fine if I add to lines to the "Device" section

Option "VideoOverlay" "on"
Option "OpenGLOverlay" "off"

don't know if I need the second line however

[Michael]

Can we see some abuse dished out via Viewperf (Yeah, yeah, I know- it's one of the standard torture tests all the same and NVidia AND AMD/ATI use it as one of their benchmarks)?

Some SPECViewPerf benchmarks were done about a month ago. I'll see if I can still locate them.

[Michael]

The RS480 and RS690 IGPs have also seen some huge performance improvements, but I haven't personally validated that yet.

[Michael]

WOW - Michael the hype leading up to it 8.41 was well worth it. I made a decision to get a HD2600XT a month ago based on your hype - thankyou and thanks ATI :D

Just have to wait a week or so...

Have there been any changes with respect to CCC and the installation procedure (other then just updating packaging scripts and a new ccc version number)?

Test request - no chance you can see if any major games [hl2, oblivion] supported by wine now "work better"?

Test request 2 - Savage 2 beta old driver vs new driver comparison

Test request 3 - comparison of windows vs linux performance with new driver in doom 3 & q4:et

The only file thing that really changed is fglrx_dri.so is now dead and right now is a dummy file... Other than that, I don't remember any other big file changes.

Some WINE tests are on the TODO list.

[Michael]

Is there a date of availability for the new driver?

Early next week.

[Michael]

Oh by the way Michael, hows the anti aliasing performance now? Is it the way it should be now too?

AA/AF is currently disabled for the R600.

[Michael]

The only desktop performance problems I have had has been corruption around the mouse and watermark (which you won't have) on one specific R500 GPU. Aside from that, everything has been as smooth as silk.

[Michael]

Video support is still done with TexturedVideo.

[Michael]

Also, what about output device detection/configuration, big desktops, randr-1.2 support?

aticonfig has maintained the Dynamic Display Management Options now for a while.

[Michael]

As an owner of a laptop with a FireGL V5250, I really have no choice but to try the 8.41 driver because the 8.40 driver fails when trying to play anything more demanding than "Quake 2". Anyway, I thought that the V5250 was based on the R530. What gives?

The FireGL IDs will be disabled, so you will get no where. AMD just has additional testing they need to go through with their workstation ASICs.

[Michael]

I Very much second to get information about that.

Also, will this piece of shit natively compile on 2.6.22 on x86-64? All I could find in the info was some crap about Fedora. As if everyone in the world used fedora or their kernel packages...

I recommend Michael to start informing about LINUX KERNELS and not some specific RH/Fedora packages.

New features, sound nice, but something (history I suppose) tells me it'll suck, just like the other drivers. Especially for us who don't care about Quake FPS, but rather big desktop / RandR 1.2, new kernel support, x86-64 and other rocket science things ATI seems to find hard to acknowledge.

I'd call this "Script kiddie's FPS goes up"-edition

x86_64 should work with the Linux 2.6.22. Fedora 7 x86_64 is also fixed!

[mile]

The only desktop performance problems I have had has been corruption around the mouse and watermark (which you won't have) on one specific R500 GPU. Aside from that, everything has been as smooth as silk.


???

If you use kde, open Konqueror/dolphin filemanager and open folder with large numner of files. It takes more than one second to draw content. This is not happening with avivo driver or Nvidia card in my other comp....

I dont know, if I missed something or this is "normal desktop performance" for you...

[Michael]

???

If you use kde, open Konqueror/dolphin filemanager and open folder with large numner of files. It takes more than one second to draw content. This is not happening with avivo driver or Nvidia card in my other comp....

I dont know, if I missed something or this is "normal desktop performance" for you...

I don't use KDE so I can't comment for that, but I haven't had any such problems like that with Nautilus and fglrx 8.41.

[nicweb]

aticonfig has maintained the Dynamic Display Management Options now for a while.

Does that mean randr 1.2 is supported or you have to use aticonfig to do that?

[Michael]

Does that mean randr 1.2 is supported or you have to use aticonfig to do that?

Aticonfig needs to be used, since aticonfig had these options long before RandR 1.2.

[Alistair]

The New Driver Is Out! Meet Oktoberfest!

ATI R500 Linux Performance -- With an X1300PRO, X1800XL, X1800XT, and X1950PRO in hand we compared the ATI 8.40 and 8.41 display drivers. The performance improvements for the Radeon X1000 series range from 50% to 90% faster.

Read More: http://www.phoronix.com/vr.php?view=10970


MIchael
Your testbase cards in the article are all PCIE cards..
Any chance you can lay your hands on AGP versions and an agp rig and kick some of these R5xx chips on AGP bus sets?


Currently my 9550 AGP (R3xx) kicks my X1650Pro AGP's (R5xx) ass sideways simply because fglrx_drm does not see the CARD as agp - it identifies the chipset as PCIE and never negotiates an AGP bus speed change. Doing anything that moves large amounts of data to the card (cough THINK about that statement) is sluggish in the extreme since the card is effectively in AGP mode0 (PCI standard speed)

What I'm reading here tho - I may just kick this driver in and give back a speed diff on the R3xx chips ... sounds like there might be a boost in there for me too....

Anyhow -- I"ll file version bump request @ gentoo and be waiting with baited breath for the release - bet yer ass I'm hauling this one down and setting up the ebuild for it in bugzilla.....

[Michael]

All of our AGP maintained systems end with the Radeon 9200 and 9250 chips. I'll see if I can find some more information though on any AGP changes for this release.

[billybore]

why?

i see reply to any question, but not for mine, Why?

[opera]

It actually says it won't work with 2.6.23 (general information), so you can't use it on Fedora 8 (specific information). Maybe I should re-use my signature with "people thinking distros are different kinds of linux"...

I didn't ask about brand new (not even released) 2.6.23. To me, and most others (iirc) with x86-64, you can't use anything after (including) 2.6.20. That's terribly weak.

I don't know what you're trying to say regarding distros, but surely distros do configure and even patch the kernels differently, so you can't compare a certain vanilla kernel version with for instance "Fedora 8". But even if you could, do you think it's reasonable to refer to Fedora releases instead of kernel versions? Am I supposed to know whatever kernel Fedora X uses? To me, that's completely irrelevant. I will never care even remotely whatever kernel version Distro X, version Y uses. I simply don't care, and I shouldn't have too, since I'm not using that Distro damn it!

Just wait and see. Michael said it'll be available in 1 week wtf.

Tell that to everyone else in this thread then? I'm not the only one commenting without actually having the driver. Or perhaps you just thought I was too negative, and thereby I needed to be shut up?
I dunno.

We all speak out of experience with the so far drivers. Many of us see a pattern of not caring about really fixing the sucky drivers, but just fixing a bug a month, and this time, making the driver get a better score on FPS benchmarks.

Then you have no idea what script kiddie means.

That's a terrible loss of knowledge to me then, omfg! I've gotta go back to school and learn these important things, before I give another statement on a forum on the internet. The whole fucking world could collapse!

Anyway. I've got some answers now.
* No RandR 1.2, because ATI has invented their own proprietary tool before RandR 1.2 was invented, and because of that (very interesting reason) they won't use RandR 1.2 even though it's superior.
* 2.6.22 _should_ work. Well, 2.6.20 _should_ have worked (if you ask me) a long time ago. Will they have fixed support for 3 major kernel versions in a row? OMFG. Unbelievable. And highly unlikely :/

Well well, we'll see... Maybe the "open source" driver ATI will release in a couple of decades, and which will have become well tested and patched a few decades after that, will be interesting.

But ATI surely doesn't understand open source development. Release a 50 mb .tar.bz2 and say "we're now open source" is lunacy. Release _early_ and _often_. This way, important and highly skilled people (I'm thinking of X.org developers) can pin-point flaws and comment how things should be interfaced, to work better with X.org.

Otherwise, it might be "open" but it'll take a _long_ time until it'll "work like a charm" with "just about any linux kernel, distro, X.org-version", which the currently support drivers more or less do.

But ATI probably just want being tagged as "open source friendly" instead of really being interested in community development.

Again, I'm speaking ahead of time, but it doesn't take many brain cells to figure out what's going on here. I'm not saying I'm guaranteeing the new ATI driver to suck, I'm just saying it most likely will.

[Alistair]

The New Driver Is Out! Meet Oktoberfest!

AMD 8.41 Display Driver Preview -- Our first article on this new driver are simply the technical details such as noting this driver has been actually in development for over a year, the AIGLX support that is coming next month, and other relevant information.

Read More: http://www.phoronix.com/vr.php?view=10955

ATI R300/400 Linux Performance -- We re-tested the Radeon X300 and X800XL with this new driver and it offers some HUGE performance improvements! The Mobility Radeon X300 was as much as 54 times faster compared to the previous ATI Linux driver. The improvements in these older ATI GPUs are simply astonishing.



Okay Michael I realize that PCIE is the hot new standard -- but there are a ***heck*** of a lot of AGP bus based cards out in the wild still -- can we get tests on AGP bus cards -- The cards in this article are again all PCIE based..... or IGP, which if sanity in chipset design exists are PCIE internally.

[hmmm]

Michael: Will CCC integrate well with the new "Graphical configuration tool for X" coming out in 2.20? It'd be nice to have say an extra tab called "CCC" - sort of like how the windows forcewares work.

Also, I noticed all the benchmarks use x server 1.3 - is there x.org 7.3 /x server 1.4 support in the new driver? I can't find anything on the thread that confirms or denies this.

[Alistair]

All of our AGP maintained systems end with the Radeon 9200 and 9250 chips. I'll see if I can find some more information though on any AGP changes for this release.


heh -- I have the rig..... *grin* I even have disk space -- I can drop an FC install on an lv and give 'er a spin for ya....

[chrisr]

The FireGL IDs will be disabled, so you will get no where. AMD just has additional testing they need to go through with their workstation ASICs.

That's another month wasted, then! I never thought that getting the graphics working (properly) on this laptop would be so frustrating!

Please ask them to test Lenovo T60p wide-screen laptops with V5250 chips and Core 2 Duos, so that the wait is ultimately worth it...

[mile]

I don't use KDE so I can't comment for that, but I haven't had any such problems like that with Nautilus and fglrx 8.41.


Nautilus is not affected with this "feature"...I hope there are improvements on this, because I don't want switch DE because of f***** display driver

[Alistair]

Michael -- I've opened 191374 on bugs.gentoo.org -
http://bugs.gentoo.org/show_bug.cgi?id=191374


As soon as I get a copy of the new driver I'll do what I can to get an updated ebuild working with this driver.....
For what its worth I'm on vacation this week .. and some s.o.b stole our van overnight -- so I'm kinda grounded for the next day or so....

[d2kx]

Rocks, doesn't it? :D

[Svartalf]

As an owner of a laptop with a FireGL V5250, I really have no choice but to try the 8.41 driver because the 8.40 driver fails when trying to play anything more demanding than "Quake 2". Anyway, I thought that the V5250 was based on the R530. What gives?

Considering that all the laptop chips were more poorly supported under the old fglrx (Poor/No Hypermemory support, etc.) it's not surprising, really.

[Xipeos]

Tell that to everyone else in this thread then? I'm not the only one commenting without actually having the driver. Or perhaps you just thought I was too negative, and thereby I needed to be shut up?
I dunno.
You were predicting doom with no knowledge of the driver.

We all speak out of experience with the so far drivers. Many of us see a pattern of not caring about really fixing the sucky drivers,

Since the new driver has a new (and different) codebase, you can't use the old drivers for your predictions.

but just fixing a bug a month, and this time, making the driver get a better score on FPS benchmarks.

According to Michael, this new driver has been in the works for a very long time, not just the 1-month development time allocated to previous drivers.
Also, since the old driver is likely going to be trashed, there was little reason for the dev/financial team to support a soon-to-be-obsolete driver, hence the lack of any real improvement


* No RandR 1.2, because ATI has invented their own proprietary tool before RandR 1.2 was invented, and because of that (very interesting reason) they won't use RandR 1.2 even though it's superior.

Any definitive evidence of randr superiority? Anyway, if the new driver will be open-sourced, you can just hack randr support into it.

But ATI surely doesn't understand open source development. Release a 50 mb .tar.bz2 and say "we're now open source" is lunacy.

When did that happen?! :p

Again, I'm speaking ahead of time, but it doesn't take many brain cells to figure out what's going on here. I'm not saying I'm guaranteeing the new ATI driver to suck, I'm just saying it most likely will.
Again, you can't use the old crappy-driver pattern for the new one.

1 week from now, if the new driver is as bad as you say, we'll all hail you as the new Nostradamus predictor@home. Until then stfu, you have no real basis for your predictions.

[Kano]

Interesting new driver (would like to test it), but compared to NVIDIA I would like if there is something like

Modes "nvidia-auto-select"

To autoselect the highest possible res automatically. Also NV drivers override the H/V sync values until you deactivate that by special options. Is there something like that in there too?

Also for using beryl (and I expect compiz-fusion) too, there is no real need for AIGLX as minimum requirement, but Composite has to work. Is that working? Composite is also used for KDE transparency which showed corruptions on older releases.

[Xipeos]

Interesting new driver (would like to test it), but compared to NVIDIA I would like if there is something like

Modes "nvidia-auto-select"

To autoselect the highest possible res automatically.

That's already in, and has been implemented for a long time.

[Kano]

Ok, then how to use it?

[Xipeos]

Ok, then how to use it?

Just don't specify a resolution in your xorg.conf maybe? I never did, but X always runs at my monitor's native resolution & refresh rate.

Then again, maybe I'm an unique case seeing as I also get some rare bugs..

[mintcoffee]

???

If you use kde, open Konqueror/dolphin filemanager and open folder with large numner of files. It takes more than one second to draw content. This is not happening with avivo driver or Nvidia card in my other comp....

I dont know, if I missed something or this is "normal desktop performance" for you...

I have this same problem with my V5200 on my laptop. Yes, strangely, it may be a KDE issue since my desktop running GNOME on a X1800 doesn't have this problem.

[Kano]

That works, you are right, just for NV you have less things to do when you want to specify a res again (a simple sed line). But using # it would work too...

[Michael]

That works, you are right, just for NV you have less things to do when you want to specify a res again (a simple sed line). But using # it would work too...

It actually has worked before NVIDIA introduced their auto select mode.

[sundown]

The only desktop performance problems I have had has been corruption around the mouse and watermark (which you won't have) on one specific R500 GPU. Aside from that, everything has been as smooth as silk.

OMG... ^_^ if that's what I'm thinking... I'm already liking this new driver :)

[grazzt]

Whats the status of AIW functionality, like video in, etc? if any?

And those R300 benchmarks still look fairly crappy? An 800XL should be much faster than the 6600.

Of course, I guess they have to start somewhere, and this should be better than nothing.

[Michael]

Currently there is no video-in support.

[petercruickshank]

This really should be called the you-nearly-gave-me-a-heart-attack-with-unexpected-news edition.

[Dragonix]

Whats the status of AIW functionality, like video in, etc? if any?

And those R300 benchmarks still look fairly crappy? An 800XL should be much faster than the 6600.

Of course, I guess they have to start somewhere, and this should be better than nothing.

Under win, according to tomshardware.com, a X800XL is little faster (1-2 fps) than an NV6600GT.

[Xipeos]

Under win, according to tomshardware.com, a X800XL is little faster (1-2 fps) than an NV6600GT.

You must compare "apples with apples". Are you sure that the benchmark wasn't using direct x? :)

[grazzt]

Currently there is no video-in support.

Will there ever be video in support? AIW cards are popular ones.

http://www23.tomshardware.com/graphics_2007.html?modelx=33&model1=722&model2=743&chart=318

Dunno where you see 1-2 fps (even select other benchmarks), the X800Xl is 172% faster, even if its a 800 GTO (256 mb, which the phoronix article seems to mention it having), its still 138% faster.

Unless Im reading the wrong benchmarks.

http://lwn.net/Articles/248227 mentions R500 onward being opensourced?

[Michael]

http://lwn.net/Articles/248227 mentions R500 onward being opensourced?

LWN doesn't mention everything... In a few hours the real embargo will expire, and we will spill the beans on what really is going on.

[Regenwald]

ah, can't wait to read it. a few hours, michael... beside, thank you for engagement & your really great articles

[Alistair]

LWN doesn't mention everything... In a few hours the real embargo will expire, and we will spill the beans on what really is going on.

Hmmms...

Yer implying that there's .... More to come that you are still NDA'd on????

blinkblink???

[Michael]

ah, can't wait to read it. a few hours, michael... beside, thank you for engagement & your really great articles

Thank you for the nice comments!

[Thetargos]

When I first read the article, I was ecstatic. I had been avoiding ATi/AMD for graphics on Linux for the longest time. I am a former ATi R300 owner. I totally loved that card (R9500), but I got it just around the time I went 100% Linux and I tried to be patient and held on as much as I could. Finally I went with and nVidia card. These news are great. Finally ATi/AMD is presenting nVidia with some real competition in the Linux ground, and this could prove more beneficial to ATi/AMD than expected, as I've read all over the place the 2900 isn't doing too well on Vista either... Maybe this here is material for another investigative article about the performance Delta in Vista and Linux of the 2900, even with "small talk" about how well does it fair and handle the desktops eye candy effects on each platform and how does it fair in gaming, both with eye candy on and eye candy off.

[Alistair]

When I first read the article, I was ecstatic. I had been avoiding ATi/AMD for graphics on Linux for the longest time. I am a former ATi R300 owner. I totally loved that card (R9500), but I got it just around the time I went 100% Linux and I tried to be patient and held on as much as I could. Finally I went with and nVidia card. These news are great. Finally ATi/AMD is presenting nVidia with some real competition in the Linux ground, and this could prove more beneficial to ATi/AMD than expected, as I've read all over the place the 2900 isn't doing too well on Vista either... Maybe this here is material for another investigative article about the performance Delta in Vista and Linux of the 2900, even with "small talk" about how well does it fair and handle the desktops eye candy effects on each platform and how does it fair in gaming, both with eye candy on and eye candy off.

Dunno bout yer R300 -- but

my
01:00.0 VGA compatible controller: ATI Technologies Inc RV350 AS [Radeon 9550] (prog-if 00 [VGA])
Subsystem: ATI Technologies Inc Unknown device 0402


does just fine with the fglrx drivers from 8.27 on out ... I even get decent framerates at 1440x900 WoW in wine - - not great mind you - - but playable

[ProTech]

Can someone post the output of the xvinfo command on an X1000 series card.

I hope I can finally watch TV under Linux.

[Thetargos]

Dunno bout yer R300 -- but

my
01:00.0 VGA compatible controller: ATI Technologies Inc RV350 AS [Radeon 9550] (prog-if 00 [VGA])
Subsystem: ATI Technologies Inc Unknown device 0402


does just fine with the fglrx drivers from 8.27 on out ... I even get decent framerates at 1440x900 WoW in wine - - not great mind you - - but playable

Well, simply put, I was getting mostly interactive frame rates in most programs, but albeit my friends with nVidia comparable cards of the time (FX series) were getting twice as much as I was, at the very same resolutions.

[Porse]

Sooo, is anyone else constantly reloading this page (http://ati.amd.com/support/drivers/linux/linux-radeon.html) in anticipation? :)

[Michael]

Sooo, is anyone else constantly reloading this page (http://ati.amd.com/support/drivers/linux/linux-radeon.html) in anticipation? :)

You could just subscribe to the Phoronix RSS (http://www.phoronix.com/rss.php) and you will know for sure when the driver is released... along with other great Linux content :)

[carpman]

Yeah, get PLUGed in and turned on. :)

[Michael]

The next Phoronix article (specifications) just got embargoed back twelve hours until noon EST on September 6... So just back twelve hours. Stay tuned for all the details!

[hmmm]

good stuff - couldn't they make it 10 in the morning? [12 Noon EST translates to 2 AM in the morning to me here in Sydney].

[Michael]

FYI, the Radeon HD 2400 and Radeon HD 2600 results will be published this Friday.

It looks like we may have at least one graphics article a day each week day (and possibly weekends) leading up to the official 8.41 driver release.

[Thetargos]

We don't want the good news to be over yet!! :D

[hmmm]

Do you have any plans to do becnhmarks comparing the 1gb & 512mb HD2900xt cards and the 512mb & 256mb HD2600xt?

[Moustacha]

FYI, the Radeon HD 2400 and Radeon HD 2600 results will be published this Friday.

It looks like we may have at least one graphics article a day each week day (and possibly weekends) leading up to the official 8.41 driver release.
Might be too much of an ask, but could you compare with previous gen mid-range cards, of ati and nvidia, and with the nvidia mid-range if you can?

[billybore]

I repeat my questions about new 8.41 driver, anyone can ansewr me?

1- News about Hypermemory? (very important for mobility)
2- News about fan control?
3- News about suspend/resume problem? (specially for Fedora 7)
4- With ATI Mobility X300 and new driver is possible to play with Flightgear and use Cedega? (with Windows Vista I can play with many games (PES6,VSK5, ...) at high quality details)

Please answer me, thanks.

[phreadom]

Personally I'd really like to see a clear answer on the status of support for the AGP X1600Pro.

I have a 512mb card that currently is recognized incorrectly as PCIE, resulting in the 0x AGP mode and the driver only sees 256mb of the 512mb of vid mem.

This effectively makes the card useless for almost any 3d application.

That aside, great news today! I look forward to further news. :) Thanks for all the information Michael!

[opera]

You were predicting doom with no knowledge of the driver.

So people predicting good things are ok, but people predicting bad things are not? They are told to shut up? Are you taking somebody's side here, or what?
And yes, I have the same knowledge of the driver as everyone else. It sucks as of now. So, we can expect it to suck. I'd say, predictions about it still will suck, are more reasonable than the others.

Since the new driver has a new (and different) codebase, you can't use the old drivers for your predictions.

New? You're not talking about the open source driver, are you? Certainly you can't rewrite everything from scratch and keep not only major version (8) but high minor (4)? It will, for sure, include lots of crap we're stuck with today.

According to Michael, this new driver has been in the works for a very long time, not just the 1-month development time allocated to previous drivers.
Also, since the old driver is likely going to be trashed, there was little reason for the dev/financial team to support a soon-to-be-obsolete driver, hence the lack of any real improvement

I still think you're referring to another driver. We're still talking another minor version of fglrx 8.4x.

Any definitive evidence of randr superiority? Anyway, if the new driver will be open-sourced, you can just hack randr support into it.

Oh, my fucking god. Are you serious? You really have to be kidding. Do you know what RandR 1.2 includes? I very much doubt it. It's a solid base meant to be used for all drivers. It has been in the planning before Xfree86 "became" X.org. It has lots of things apart from just "big desktop" support.
* Easy management of monitor layout (one tool for all your computers, whatever driver they have).
* Sub-pixel knowledge for smooth font rendering even with:
* Rotation support
* Auto resizing the desktop on hardware plug-in/out

And it's clear that you've gotten a media-biased and very unrealistic view on open source. You think that just because it's open source, all problems are solved? "you can just hack randr support into it." What?
Let me tell you a few things;
* It's not "just" soemthing to add.
* I've bought an ATI card. Why should I for free make their drivers better? Open source doesn't mean "let other people do the hard work for free"
* If ATI still control the project, instead of being community centered, my (or others) patches might not be merged, and am I supposed to support and release my own versions of the driver? The word "Windows" pops up in my head.
* RandR may very well heavily conflict with what's used in the drivers, so a merge might be very hard. Especially for someone new to the source tree.

Get realistic, will ya?

It's sad to see what might happen when people like you comment on things like this. Imagine in the future, the "open source" ATI driver lacks TV-out, and people complain about this (they bought a card with TV-out support). Will you tell them "it's open source, implement that yourself"? That's just disgusting.

Open source has absolutely nothing to do with the manufacturer's responsibilities. You can't just move the responsibility to the end-user because you give them the source code.

When did that happen?! :p

Never at ATI, but I've seen the same thing on lots of other big corps "going open source", and I see a tendency here. I'm very sorry for sharing my assumptions based on these tendencies.

Again, you can't use the old crappy-driver pattern for the new one.

Yes, since the "new" one (if we're talking about the open source driver, not 8.41), seems (yes) to be developed only at ATI, with the only difference of a big source tree being available. It's a very good thing yes, but the driver could still suck. If they'd work together with the community a heck lot earlier than they are, they could have gotten lots of help into making it work well (portability, no re-using of X.org functionality etc etc). I already explained this though.

1 week from now, if the new driver is as bad as you say, we'll all hail you as the new Nostradamus predictor@home. Until then stfu, you have no real basis for your predictions.

If I'm right, you'll hail me, but until then you tell me to "shut the fuck up", just because my predictions doesn't suite you? Who are you, and what are you afraid of? You a share holder of ATI? You think people will stop buying ATI products like crazy because of what I write?
Come on ffs. I don't see much good at ATI, and I am rather skeptical of the new drivers. Don't I have the right to be skeptical?
Even the better if I happen to be completely wrong. For crying out loud, I'd be happy to be completely wrong.

[Xipeos]

I still think you're referring to another driver. We're still talking another minor version of fglrx 8.4x.

Don't judge a driver by it's version. Someone suggested the 8.41 means the 41 release of the 8th year. Not everyone is using the same versioning system as the kernel devs.


And it's clear that you've gotten a media-biased and very unrealistic view on open source. You think that just because it's open source, all problems are solved? "you can just hack randr support into it." What?
Let me tell you a few things;
* It's not "just" soemthing to add.

Of course you can opera. A couple of dedicated programmers can do it in...two weeks? wtf do you know?! You're just an angry user.

EDIT: you don't have to re-write the whole driver for randr. You can just build an interface that translates randr requests to DDM (like wine and cedega do with their compatibility layer).


* I've bought an ATI card. Why should I for free make their drivers better? Open source doesn't mean "let other people do the hard work for free"

You'd do it for yourself. You can pay yourself if you want to... (but then again, you're just an angry user, not a programmer (?) )


* RandR may very well heavily conflict with what's used in the drivers, so a merge might be very hard. Especially for someone new to the source tree.

Can't take a little challenge, you wuss?


Open source has absolutely nothing to do with the manufacturer's responsibilities. You can't just move the responsibility to the end-user because you give them the source code.

Again with the wussing. As of now, we know very little about AMD's open-source initiative. And even if they shifted all the work on us, it would be much better then what we had so far. They way I remember it, *many* people were asking for the shift (even you - "give it to x.org hackers" remember?!)


Come on ffs. I don't see much good at ATI, and I am rather skeptical of the new drivers. Fine be skeptical, I won't comment on your bitching anymore.

[Michael]

Might be too much of an ask, but could you compare with previous gen mid-range cards, of ati and nvidia, and with the nvidia mid-range if you can?

With the HD 2400 and 2600 we are also doing the NVIDIA 8500 and 8600 and the ATI X1300.

[Michael]

Xipeos and Opera, do I have to tell you two to stop fighting?

[Michael]

I repeat my questions about new 8.41 driver, anyone can ansewr me?

1- News about Hypermemory? (very important for mobility)
2- News about fan control?
3- News about suspend/resume problem? (specially for Fedora 7)
4- With ATI Mobility X300 and new driver is possible to play with Flightgear and use Cedega? (with Windows Vista I can play with many games (PES6,VSK5, ...) at high quality details)

Please answer me, thanks.

I don't know about Hyper Memory right now.

If the fan control is with a Power Play mode, it should work.

Suspend/resume may be fixed depending upon the system.

I have not used Flightgear and Cedega/WINE will run faster but missing some GLX things (see other thread)

[Alistair]

Xipeos and Opera, do I have to tell you two to stop fighting?

Thanks Michael ...needed that -

[Xipeos]

Xipeos and Opera, do I have to tell you two to stop fighting?

It's done. Sorry for the inconvenience.

Thanks Michael ...needed that
Alistair... :(

[Alistair]

I don't know about Hyper Memory right now.

If the fan control is with a Power Play mode, it should work.

Suspend/resume may be fixed depending upon the system.

I have not used Flightgear and Cedega/WINE will run faster but missing some GLX things (see other thread)


I am in the process of collecting all I can on my systems stats right now - I use wine and play world of warcraft religously - Given the context of the OP's questions, and what we see in the reviews so far, I'm suspecting that Flightgear (native linux) will be much more likely to be playable on IGP chipsets. As for Cedega/Wine :

Michael - the excepted GL extension in Wow is specific (Per Tbird's comment) to the way Blizzard has implemented the function in their code.

What Tbird is concerned about beyond that is the mapping method ATI has used to xor the server available extension and the client available extensions into the main available extensions - it in some cases does not result in something that wine can rely on for accurate information, and results in the Wine folks having to use some hacky code to make things work. I can gather from a few other threads I ve seen of this nature that Cedega goes even farther down that path.....

[Alistair]

Personally I'd really like to see a clear answer on the status of support for the AGP X1600Pro.

I have a 512mb card that currently is recognized incorrectly as PCIE, resulting in the 0x AGP mode and the driver only sees 256mb of the 512mb of vid mem.

This effectively makes the card useless for almost any 3d application.

That aside, great news today! I look forward to further news. :) Thanks for all the information Michael!

In the same boat - I'm testing this the instant I can get my hands on the driver and get it built on my system -- I will post results....

[niniendowarrior]

Just dropped in by chance. So, AMD is getting their act together? Okay, great. When do we get this one?

[Michael]

Just dropped in by chance. So, AMD is getting their act together? Okay, great. When do we get this one?

AMD has gotten their act together.

8.41 will be out early next week.

[niniendowarrior]

Can we see some abuse dished out via Viewperf (Yeah, yeah, I know- it's one of the standard torture tests all the same and NVidia AND AMD/ATI use it as one of their benchmarks)?

(EDIT) Oh, can we see if the driver actually WORKS with Hypermemory machines properly in UMA, Dedicated, and UMA+Dedicated modes and what performance values we get from all three modes?

Oh Svartalf, is this the end of our ATi cynical days? :D