3D Nature and Partners Declare ATI drivers Unsuitable for Professional Visualization

Farid

Artist formely known as Vysez
Veteran
Supporter
Press Release said:
May 4th, 2006, Morrison Colorado

3D Nature, a leader in photorealistic landscape visualization, together with a prominent group of 3D visualization developers and users, have taken the drastic step of publicly declaring ATI's Catalyst OpenGL display drivers as unsuitable for professional realtime visualization needs.

All past and current ATI Catalyst drivers for all ATI Radeon series display cards contain several critical bugs that render the hardware unusable in professional 3D visual simulation applications. ATI has been aware of the problems since 2005 but has not fixed the most critical issues, has not announced an ETA for fixing them, or even acknowledged them publicly in their Known Issues for current releases.

The two critical failures are referred to as "VSYNC spinlock" and "compressed SGIS_generate_mipmap failure". Both are documented extensively on the OpenSceneGraph web site complete with screengrabs and sample code to reproduce the problem.

VSYNC spinlock refers to the situation where the display driver continuously runs the CPU during a wglSwapBuffers() function call when vertical retrace synchronization (VSYNC) is on. This leaves the CPU unavailable to perform other work during this period when it should normally be idle, drastically impairing the performance of modern multi-threaded visualization applications. Other modern display card hardware drivers (3D Labs, NVidia) leave the CPU available during this operation.

Compressed SGIS_generate_mipmap failure described how the very-common mip-map generation extension, "GL_SGIS_generate_mipmap", is completely broken when using texture compression. Most modern applications generate mipmaps using SGIS_generate_mipmap to improve texture anti-aliasing in 3D scenes. Texture compression is also commonly used to reduce the memory consumption of textures in the scene in order that more or more detailed textures can be resident in the display card, improving scene detail and performance. It is very common to use both of these techniques together in professional visual simulation applications to extract the maximum visual quality and framerate from a given graphics subsystem. The failure of this combination of features on ATI cards means that texture compression is unusable, severely impairing scene performance. Again, other manufacturers do not suffer this problem.

ATI has been aware of these these issues for quite some time, and has internally filed them as "EPR #151824".

Joining 3D Nature in their protest of ATI's neglect are numerous luminaries of the realtime 3D visualization industry:
Fredrik Ahl, Scalo AB, www.scalo.se
Don Burns, Andes Computer Engineering, www.andesengineering.com
Carlo Camporesi, Institute of Technologies Applied to Cultural Heritage, Italy
Jan Ciger, OpenGL developer
Ben Discoe, project lead, "Virtual Terrain Project" www.vterrain.org
Murray G. Gamble, Director of Modeling & Simulation, Aerospace and Cognitive Engineering (ACE) Laboratory, Carleton University
Roger James - CTO - Virtual Outlooks Ltd.
Gert van Maren, Head of Development, K2Vi Virtual Reality Software, www.k2vi.com
Robert Osfield, Project Lead, OpenSceneGraph, www.openscenegraph.org
Gordon Tomlinson, Consultant, www.3dscenegraph.com
James A. Zack, President, Xtra-Spatial Productions, www.spatialexperts.com
Collectively, we call upon ATI to address these issues immediately, in order that their hardware may be made suitable for professional visual simulation uses.

Current information about this issue will be found at: http://www.3DNature.com/ati.html

About 3D Nature:
Since 1992, 3D Nature has developed and marketed professional photorealistic 3D landscape visualization software including World Construction Set, Visual Nature Studio, Scene Express and NatureView Express. http://www.3DNature.com



I didn't alter the original title of this Press release but I did emphasize the important terms of this PR: Catalyst OGL Drivers and Professional needs.

Why did I do that? Simply because Ati professional cards (FireGL) do not use Catalyst drivers...
And according to what I heard, these issues, shown in this PR, do not occur with the FireGL parts and their drivers.

So, in other words, Ati's professional solutions are suitable for professional needs.
The gaming parts (cheaper than the professional cards...) are not... Obviously. But they never been advertised as being suitable for professional needs in the first place.
 
hm.... would they just be silently promoting nvidia cards then? (do their gaming cards suffer from the same thing as Ati's cards in the workstation arena?)

Maybe you should email them about their erroneous use of gaming cards as professional cards. ;)
 
Vysez said:
I didn't alter the original title of this Press release but I did emphasize the important terms of this PR: Catalyst OGL Drivers and Professional needs.

Why did I do that? Simply because Ati professional cards (FireGL) do not use Catalyst drivers...
And according to what I heard, these issues, shown in this PR, do not occur with the FireGL parts and their drivers.

So, in other words, Ati's professional solutions are suitable for professional needs.
The gaming parts (cheaper than the professional cards...) are not... Obviously. But they never been advertised as being suitable for professional needs in the first place.
So basically the list of "professionals" who signed this are most probably anything but? :LOL:
 
So the measure of how professional one is is how much you spend on your graphics card? :)

I too was wondering if other consumer cards have similar problems. It seems ATI/NV might both choose to limit their consumer cards to make "professionals
buy the expensive cards. As you should well know many cards in the past were identical but the professional ones had a huge price tag one them.

If I were a professional and could use a gaming card and acheive the same results for 50% of the price I think that I would.
 
Sxotty said:
If I were a professional and could use a gaming card and acheive the same results for 50% of the price I think that I would.
I appreciate that, but there's a fine line between trying to make some savings by using gaming cards for professional task and publishing misleading Press Release about how the "Drivers", in a general sens, of a company aren't suitable for professional needs, whne these Drivers are gaming/home drivers that have never been advertised as supporting professional applications.
 
Well, I would like to hear from our ATI guys that this does work as expected on FireGL.

But if it does, then yeah, it was at least moderately sleazy for them to write the thing the way that they did.

Sure, everybody likes to get a deal. The professional cards exist for a reason tho.

Edit: Tho, the interesting point, is why bother? Why don't they just buy 7900GTX for their needs? A protest of this sort might imply there is something else about X1900 they'd like to have that is missing from the competition, if this issue was addressed.

EditII: Oh. Come to think of it, maybe what they are after here is expanding their market by being able to sell their professional software to folks who want professional-class results from consumer-priced cards. . .that makes more sense for why they'd be a bit on the shrill and somewhat disingenuous side; more financial impact than their own viddy card needs.
 
If you install the FireGL drivers on a standard ATi card does this problem still happen?

I know you can install the FireGL drivers on a regular ATi card, if that solves the problem I don't see a problem.
 
digitalwanderer said:
If you install the FireGL drivers on a standard ATi card does this problem still happen?

I know you can install the FireGL drivers on a regular ATi card, if that solves the problem I don't see a problem.
you have to flash the bios to get the id tag to get the Cp options for the Pro/fgl driver options
But you cant do that anymore with the 1xk group... so hence the whine from the guys who used to alter there vanila cards to pro...
 
arjan de lumens said:
http://www.openscenegraph.org/osgwiki/pmwiki.php/Tasks/OpenGLConformance appears to be the main source of documentation on these bugs. At least one of the bugs (the "compressed SGIS_generate_mipmap failure" bug) is listed as appearing with FireGL drivers.

The page also lists bugs seen on Nvidia and 3dlabs cards (albeit much fewer than with ATI).
From this site:

http://www.openscenegraph.org/osgwiki/uploads/Tasks/ATIRadeonOmegaVersion.png
ATIRadeonOmegaVersion.png


Omega Drivers? They're kidding, right?

Screenshots made with ATI Radeon Mobility 9700, OmegaDrivers.net Catalyst version 5.10a (6.14.10.6575):
The extra lines show when glPolygonMode is set to GL_LINE used in combination with display lists. To disable display lists in the cow.osg model, open it in a text editor and change line 49 from "useDisplayList TRUE" to "useDisplayList FALSE". Confirmed on Mobility Radeon 9600, driver 6.14.10.6549 and Radeon 9800, Catalyst 05.11.
point sprites disappears when GL_POINT_SMOOTH is enabled (Radeon 9800, Catalyst 05.11).

And the only mention of FireGL is this one:
Tested on Linux Mandriva Cooker, with the ATI proprietary driver 8.19.10, the graphic card is FireGL? T2 (IBM T41p Thinkpad).
Why is there's a question mark there, by the way?

Ati Linux Drivers are indeed rubbish, this much I can attest, but most of their complaints are directed toward gaming/home parts, and mobile ones for some reasons. And they're using unsupported Ati drivers (Omega Drivers) to support their claims, in the first place...

While, I wouldn't mind if they were bad mouthing Ati OGL support in an article or a presentation, I think it's quite unexpected to see them go as far as publish a Press Release to claims that Ati gaming drivers aren't suitable for any professional work.
 
3D Nature responds [long]

Hi there, Chris from 3D Nature here. The guy that wrote the press release.

I guess you can imply that somehow our bunch of folks are clueless, or unprofessional. After all, I've only been working in 3D since about 1988, and OpenGL since 1996. But you probably shouldn't impinge on the reputation of people like Don Burns, Robert Osfield, Ben Discoe and Gordon Tomlinson. Don's been involved in GL since before it was Open, back at SGI, and is responsible from some important stuff that's still used in Performer today. Robert is the head of the OpenSceneGraph project, the most successful Open Source scenegraph on the planet, available for Windows, Linux and MacOSX. Gordon Tomlinson worked on Multigen-Paradigm's Vega scenegraph, and pretty much anyone who has ever done realtime CG terrain knows who Ben Discoe is. Together they've forgotten more about 3D than most people ever knew.

Most of us don't own ATI hardware anymore. We got smart. I'm stuck with one because when I bought my laptop, ATI was smacking NVidia around in the laptop market and everything you could buy was Radeon Mobility based. But it's not us we're concerned about. It's our users who get burned by this. They get sold a system with an ATI graphics card in it (probably a great value) and then they find out it doesn't work as it ought. Frequently, they're upset with us, the developer, because they think it's our fault -- because ATI couldn't possibly be neglecting serious bugs for so long, could they? So, we're sick of it. We tried asking, and then pressuring ATI into getting these issues resolved, and 6 months later, they're still broken. So, we decided we needed to publicly state our position and let ATI start taking the flack for it instead of us. We've told them for months we were going to go public with it if they didn't get it fixed, and here we are.

Yeah, Radeon isn't the top-of-the-line pro hardware from ATI. But that's not the point. NVidia cards that are comparable in price to the Radeon work just great. And, outside of a few rare exceptions, you don't find FireGL in laptops, period. Not everybody can justify a pro-level card anyway -- the goal of many of our companies is to bring professional 3D to the masses -- anyone heard of Google Earth?

As far a not using GL_SGIS_generate_mipmap -- GLU provides a (slower) software implementation for mipmap generation and always has -- and in fact most software automatically falls back to software mipmapping if the extension isn't available. The problem is the ATI drivers claim it IS available, so we try to use it, and find it's broken. If they'd just admit that it was broken and make the driver not offer the extension, everything would be great. If you're gonna claim to be able to do the job, then you better live up to the task, or go home. And, as noted elsewhere -- having a FireGL _doesn't_ solve the problem -- according to our tests GL_SGIS_generate_mipmap is broken there too!

Shrill? Yeah, I guess we come off a little shrill. We're sick of being ignored and the only way to get matters like this resolved is to make it apparent to ATI that there _is_ a downside to ignoring the problem. Otherwise they'll just keep ignoring it.

We do use the pro-level cards when the application calls for it. The latest quad Opteron to make an appearance in our company's office sports a pair of NVidia 7800GTX's.

I'm not an NVidia shill. I don't have any particular reason to pimp them, but at the moment, I _have_ to recommend their hardware to my users because ATI's isn't cutting it and 3D Labs is DOA. I'd _like_ to see ATI get this fixed because we find that our users (and therefore our company) _benefit_ from the competition between the manufacturers. ATI has a good price/performance ratio -- if the cards would just work right we'd be able to recommend them again.
 
Vysez said:
From this site:
Omega Drivers? They're kidding, right?
Ati Linux Drivers are indeed rubbish, this much I can attest, but most of their complaints are directed toward gaming/home parts, and mobile ones for some reasons. And they're using unsupported Ati drivers (Omega Drivers) to support their claims, in the first place...

Yup, I'm using OmegaDrivers on my laptop -- which is where that screengrab came from. Why? Because no official ATI driver will install on most laptop Mobility Radeon systems and the most-updated official driver is way older and buggier than the Omega driver. But you can't pin the blame on Omega for this -- while I was the one who confirmed that issue and I do use the Omega driver, it's also been cofirmed by many other users on the whole spectrum of ATI hardware and drivers -- and ATI knows it. We didn't feel it necessary to junk up the page with the comprehensive list of every ATI card/driver combo that failed -- because they ALL do.
 
Now I like Omegadrive drivers too so please don't take this wrong, but what you're basically saying is that your non-ATi drivers for an enthusiast level card aren't capable of doing what their professional line-up can do and you're miffed about it....right? :-|
 
Vysez said:
So, in other words, Ati's professional solutions are suitable for professional needs.
The gaming parts (cheaper than the professional cards...) are not... Obviously. But they never been advertised as being suitable for professional needs in the first place.
If this argument is correct, it would in case indicate that ATI is maintaining product differentiation by more-or-less-intentionally retaining or introducing bugs in their products (as opposed to e.g. holding back performance or advertised features, like Nvidia seems to be doing).
 
Well, first let me welcome you to the Beyond3D Forums, Chris.

Let me also add that I, for one, didn't discuss your or your colleagues, professionalism on this issue.
If any of my post did come off as disrespectful, then understand that it wasn't meant to be as such.

This said, after reading the reasons motivating this Press Release of yours, I still cannot fully appreciate your action in this case.

Is Ati commendable for not addressing the problems you brought forth to them these last months? Probably not, I agree.
But then again, the main point of my first post stand still, the Radeon series and its drivers, the ones named in the PR, is not a Professional line at all.
So, pointing out the fact that something that has never been advertised, nor sold as suitable for professional work is, indeed, not recommended for such tasks sounds strange to me, if not somewhat redundant.

Now, you emphasized the fact that this PR had two purposes, tell your customers that their potential problems aren’t your fault but Ati’s and also you did expect that Ati would finally take notice once this issue is made public.
And I can understand that, especially (only?) if the FireGL drivers are still bugged as you said.


In addition, about the FireGL bug with the GL_SGIS_generate_mipmap extension, as you said in this day and age this particular extension isn’t specially crucial or anything, but you hold a grudge on it, just because Ati claimed that it supports it while in reality you said that it’s not working?
It’s a reason, like any other, for sure, but you’ll agree that it can be corrected quite easily by the software programmer. It would be a trivial thing to implement to permit your customers using Ati’s FireGL parts to use your software.
I know you might argue that you shouldn’t have to write any supplementary code line since Ati should do it in the first place. To what I’ll reply that if you can provide fully compatible software to a large part of your customers by doing some trivial extra effort, then you probably should do it.

The problem is that the other issues listed can’t be fixed as easily as the FireGL’s
GL_SGIS_generate_mipmap… Ati folks are the only ones who can/should correct theses. But, the thing is, theses issues do only concern the gaming cards...

Why should Ati provide support for something that they never advertised in the first place?
Because Nvidia's gaming parts have a better OGL support? That’s hardly a reason, and if it is one, release a PR stating that Nvidia gaming cards are suitable for professional tasks.

I understand that you or your customers can’t all buy expensive professional cards, but at least you should tell them, upfront (not with a PR ;) ), that your software requires Nvidia (Pro&Home) cards, 3Dlabs’s ones or Ati professional cards only.

To write off all the FireGL line because of a bug with one particular extension bug seems really like an overreaction to me.
 
arjan de lumens said:
If this argument is correct, it would in case indicate that ATI is maintaining product differentiation by more-or-less-intentionally retaining or introducing bugs in their products (as opposed to e.g. holding back performance or advertised features, like Nvidia seems to be doing).
That's a different debate all together, really.

If you want to judge, or discuss, the ethics behind this type of marketing practices, very well, but please remember that this is not the topic of the thread.

Althought, quickly, I'd say that while I do not condone these type of methods, one could argue that the software, the drivers, are an integral part of the sold package.
So, as long as the product behaves as publicized one cannot really complain about software unabled features that were advertised for another product.

Well, from a moral point of view, you might emit concerns on the matter, indeed.
 
There's no excuse for failing to comply to the OpenGL specifications :!:

Drivers bugs are one thing, but repeatedly failing to correct them is something to be ashamed of.
Programmers have enough work fighting their own bugs not to have to fight drivers also...

Really this is pretty wrong, even though the little bugs I reported to ATI have always been corrected within 2 drivers release. (I doubt I was the only one reporting them.) :smile:

And the way you are reacting people is pretty wrong, no matter what the card is meant for, it MUST comply to the specs, end of story. :devilish:

I can only wish that ATI is working on some major OpenGL driver rewrite and couldn't take those issues into account for now, but will correct them in a short future. :???:
 
Back
Top