Our Experience with UVD and R600
It's easy to point to the presentations and say, "look, this is slightly misleading." This doesn't get across the even more convincing interactions we had with AMD personnel over the course of reviewing the HD 2900 XT. Hopefully this will help to illuminate the issue further. I'll have to put a disclaimer on this and say that I don't have a photographic memory and I forgot to bring my diary to the press briefing. This isn't normally the kind of stuff I take notes on for product launches.
At the AMD press event in Tunis, Tunisia (we still can't figure out why we went to Africa to talk about a graphics card), myself and a couple other reviewers had the good fortune of being able to see HD Decode first hand. We watched a side by side comparison of an R6xx based part and a G8x based part with interesting results that mostly pointed out the long way Intervideo and CyberLink have to go in refining their player software. We were under the impression that this was R600, though clarification was not provided.
We talked about UVD and PureVideo, and we were asked if a video decode comparison would be made against NVIDIA hardware for our HD 2900 XT launch article. Upon replying in the affirmative, we got the impression that we should expect the R600 to outperform its competition (the 8800 GTS) in video decode tests. This doesn't seem as certain without UVD, but our marketing speak filters failed to pick up on anything out of the ordinary.
Moving on, we were initially told to expect a driver to enable video decode acceleration shortly before launch. AMD also provided us with a copy of PowerDVD which was said to support AVIVO HD. We did get a newer driver near launch, but in testing it we saw very bad video decode performance. It turns out that we couldn't test video decode at all, as hardware acceleration was not enabled in our press driver. This indicates to us that, even though we saw decode acceleration working in Tunis, AMD was having more trouble than expected in the QA department.
We would expect QA issue with new UVD hardware. If we've got the same video decode hardware as a previous generation part, it seems a little odd to think that we wouldn't have a driver or software to enable the feature.
In going back through our slides for this piece, we noticed that it is explicitly stated that UVD is included on the Radeon Mobility HD 2300. This is an R5xx based part retrofitted with UVD. Of all the parts not to have this new feature, the DX9 class Mobility HD 2300 would have been the one we thought of first. Surely if this part has the new feature, the HD 2900 XT should also include it. The argument for not is, again, that higher end GPUs tend to be paired with CPUs that can handle the workload.
The reason we offer time and again for the fact that G86 and G84 include full video decode while G80 does not is that the slower parts had 6 months more development time. In NVIDIA's case, it makes complete sense that G80 would lack the more refined video decode hardware. For AMD to launch a top to bottom product line and leave a feature like this out of their highest end part is very strange.
So strange, in fact, that combined with all the other experiences we've had with this rather surreal product launch we still just can't help but doubt what we're hearing. To us, the most sensible explanation for all this is that the hardware must be there but is somehow broken - perhaps broken to the point that it will never be fixed on current silicon (i.e. via drivers) and a new spin would be required to address the problems.
In spite of what makes sense to us, AMD has indicated that UVD hardware is not at all present on R600. Their "official" statement has yet to appear, but we'll update this article when/if we get it.
It's easy to point to the presentations and say, "look, this is slightly misleading." This doesn't get across the even more convincing interactions we had with AMD personnel over the course of reviewing the HD 2900 XT. Hopefully this will help to illuminate the issue further. I'll have to put a disclaimer on this and say that I don't have a photographic memory and I forgot to bring my diary to the press briefing. This isn't normally the kind of stuff I take notes on for product launches.
At the AMD press event in Tunis, Tunisia (we still can't figure out why we went to Africa to talk about a graphics card), myself and a couple other reviewers had the good fortune of being able to see HD Decode first hand. We watched a side by side comparison of an R6xx based part and a G8x based part with interesting results that mostly pointed out the long way Intervideo and CyberLink have to go in refining their player software. We were under the impression that this was R600, though clarification was not provided.
We talked about UVD and PureVideo, and we were asked if a video decode comparison would be made against NVIDIA hardware for our HD 2900 XT launch article. Upon replying in the affirmative, we got the impression that we should expect the R600 to outperform its competition (the 8800 GTS) in video decode tests. This doesn't seem as certain without UVD, but our marketing speak filters failed to pick up on anything out of the ordinary.
Moving on, we were initially told to expect a driver to enable video decode acceleration shortly before launch. AMD also provided us with a copy of PowerDVD which was said to support AVIVO HD. We did get a newer driver near launch, but in testing it we saw very bad video decode performance. It turns out that we couldn't test video decode at all, as hardware acceleration was not enabled in our press driver. This indicates to us that, even though we saw decode acceleration working in Tunis, AMD was having more trouble than expected in the QA department.
We would expect QA issue with new UVD hardware. If we've got the same video decode hardware as a previous generation part, it seems a little odd to think that we wouldn't have a driver or software to enable the feature.
In going back through our slides for this piece, we noticed that it is explicitly stated that UVD is included on the Radeon Mobility HD 2300. This is an R5xx based part retrofitted with UVD. Of all the parts not to have this new feature, the DX9 class Mobility HD 2300 would have been the one we thought of first. Surely if this part has the new feature, the HD 2900 XT should also include it. The argument for not is, again, that higher end GPUs tend to be paired with CPUs that can handle the workload.
The reason we offer time and again for the fact that G86 and G84 include full video decode while G80 does not is that the slower parts had 6 months more development time. In NVIDIA's case, it makes complete sense that G80 would lack the more refined video decode hardware. For AMD to launch a top to bottom product line and leave a feature like this out of their highest end part is very strange.
So strange, in fact, that combined with all the other experiences we've had with this rather surreal product launch we still just can't help but doubt what we're hearing. To us, the most sensible explanation for all this is that the hardware must be there but is somehow broken - perhaps broken to the point that it will never be fixed on current silicon (i.e. via drivers) and a new spin would be required to address the problems.
In spite of what makes sense to us, AMD has indicated that UVD hardware is not at all present on R600. Their "official" statement has yet to appear, but we'll update this article when/if we get it.
Facebook
Twitter
RSS
53 Comments
View All Comments
Goty - Monday, June 4, 2007 - link
http://www.driverheaven.net/reviews/r600reviewz/in...">http://www.driverheaven.net/reviews/r600reviewz/in...Goty - Monday, June 4, 2007 - link
...soooo the HD2900XT doesn't support UVD. Who cares? I still accelerates HD video play back and can even beat its competition at the feat, so what's the big deal? You guys missed something, get over it.Roy2001 - Tuesday, June 5, 2007 - link
ATI card makers cares.drebo - Monday, June 4, 2007 - link
Yet another pointless anti-AMD "article".Look, we realize that Intel buys a shitload of advertising from you, but could we atleast get a shred of real review material please?
This is bollocks. AMD never said UVD would be on the 2900 XT. nVidia didn't include full decode on the G80. Whatever the reasoning, it doesn't matter. My feelings are that those high end parts don't NEED the help. Your feelings are that AMD is deceptive and nVidia is god. We get it. You don't like AMD (or Intel pays you not to). Get over it already. I'd have rather seen an actual comparison of the video decode features of the 2900 XT, the 8800GTS, the 8600GT and the lower-end AMD parts.
You know, something that would actually mean something, other than this useless drivel, specualtion and namecalling.
Roy2001 - Tuesday, June 5, 2007 - link
Yeah, AT and many sites dislike AMD, and AMD partners are also dislike AMD so they pretend to know nothing about lacking of UVD and thus printed their box with UVD support.Is that your theory?
SilthDraeth - Monday, June 4, 2007 - link
Drebo you are hilarious. Anandtech certainly is anti AMD, and back before core 2 they where anti Intel.stfu.
chrispyski - Monday, June 4, 2007 - link
I can understand if review sites like Anandtech or Tom's get things like this messed up as they have to rely on the information given to them from the PR guys whose job it is to make there merchandise appealing. But when board partners are being mis-informed and then mis-labeling their cards accordingly, then you begin to wonder what the hell is going on with their tactics.AMD was deceitful, if not to the review sites, then to their board partners. Either way is just bad business. Plain and simple.
tuteja1986 - Monday, June 4, 2007 - link
Yeah... Like i read around 3 or more reviews like firingsquad article that told it readers that 2900XT didn't UVD.These review were published on the 1st of 2900XT launch and they didn't make fuss about it. Also i have read article that compares 8600GT playback vs 8800GTX playback and it show that Image quality is better on 8800GTX than 8600GT playback.PrinceGaz - Monday, June 4, 2007 - link
Exactly, some first day reviews of the HD 2900 did mention that it did not include dedicated UVD hardware but that it would instead be implemented using its shaders/stream-processors (because the 2900 has sufficient of them to do the task just as well as the dedicated hardware on lower models does).Some people rely on just one review (possibly AT for many readers here); that is always a bad idea regardless of the site you choose. If you feel you know enough about an important new product by doing anything less than reading at least three (preferably more) comprehensive reviews, then you are likely to miss small but significant details. Reading more than one reduces the likelihood you overlook something, and increases your chance of being told about something another site missed such as details of the UVD support in the HD 2900.
I'm not saying any site is good or bad; the site where I first learned on launch day that UVD on the HD 2900 would be provided by its stream-processors rather than dedicated hardware, is the same site which not long previously had posted a review of the 8600GT/GTS launch which was at odds with just about every other respectable tech site. All sites have their ups and downs, so never trust what any site alone says, including AT, unless others support their findings.
Creig - Thursday, June 7, 2007 - link
http://www.bjorn3d.com/forum/showthread.php?p=7888...">http://www.bjorn3d.com/forum/showthread.php?p=7888...