Ghost Recon Advanced Warfighter Tests
And the short story is that the patch released by AGEIA when we published our previous story didn't really do much to fix the performance issues. We did see an increase in framerate from our previous tests, but the results are less impressive than we were hoping to see (especially with regard to the extremely low minimum framerate).Here are the results from our initial test, as well as the updated results we collected:
There is a difference, but it isn't huge. We are quite impressed with the fact that AGEIA was able to release a driver so quickly after performance issues were made known, but we would like to see better results than this. Perhaps AGEIA will have another trick up their sleeves in the future as well.
Whatever the case, after further testing, it appears our initial assumptions are proving more and more correct, at least with the current generation of PhysX games. There is a bottleneck in the system somewhere near and dear to the PPU. Whether this bottleneck is in the game code, the AGEIA driver, the PCI bus, or on the PhysX card itself, we just can't say at this point. The fact that a driver release did improve the framerates a little implies that at least some of the bottleneck is in the driver. The implementation in GRAW is quite questionable, and a game update could help to improve performance if this is the case.
Our working theory is that there is a good amount of overhead associated with initiating activity on the PhysX hardware. This idea is backed up by a few observations we have made. Firstly, the slow down occurs right as particle systems or objects are created in the game. After the creation of the PhysX accelerated objects, framerates seem to smooth out. The demos we have which use the PhysX hardware for everything physics related don't seem to suffer the same problem when blowing things up (as we will demonstrate shortly).
We don't know enough at this point about either the implementation of the PhysX hardware or the games that use it to be able to say what would help speed things up. It is quite clear that there is a whole lot of breathing room for developers to use. Both the CellFactor demo (now downloadable) and the UnrealEngine 3 demo Hangar of Doom show this fact quite clearly.
Facebook
Twitter
RSS
67 Comments
View All Comments
segagenesis - Wednesday, May 17, 2006 - link
I feel so tempted to bring up the old cliche "The message is clear..." when you word it like that :)Really why is there not more "WTF" here? A better analogy to what you describe is the old "Hardware Decelerators" that say the S3 Virge was. And for $300? Damn, next thing we know they will be sub-licensing Patty-On-Patty technology from Burger King with a dual core physics processor for only $600! *groan*
They have the right idea here but this is some of the poorest execution possible in convincing people you need this product.
Magnadoodle - Wednesday, May 17, 2006 - link
Calling this a physics decelerator seems just perfect. I wish anandtech would use some biting humour now and then. But that would mean degraded relations with Asus and BFG.Oh well, let's just get nostalgic about the days of unconstrained journalism and reread those old 6% Pcgamer reviews.
abhaxus - Friday, May 19, 2006 - link
When I got my original voodoo 1 card, the first thing I did was plug it in and run a few timedemos in GLquake... surprise surprise, it was actually a few FPS slower than I was running in software mode. Of course, I was running software mode at 320x240 and GL at 640x480 and the game looked incredible.I haven't seen a PhysX card in person but the trailers for cellfactor look very impressive. With PhysX being taken advantage of throughout the design and coding process I can't wait to see what the final results are for new games... of course, new drivers and a PCIe version will help too.
That said... I really think that this card will eventually turn out to be only for people that don't have a dual core CPU. Seems like most everything could be done by properly multithreading the physics calculations.
Nighteye2 - Wednesday, May 17, 2006 - link
It's perfectly possible to remain be critical while remaining polite. Biting humour is unnecessarily degrading and does not add any value. Even 6% ratings can be given in perfectly polite wording.DerekWilson - Wednesday, May 17, 2006 - link
We certainly aren't pulling punches, and we wouldn't do anything to preserve a relationship with any company. If we make someone angry, we've still got plenty of ways to get a hold of their product.I hope we were able to make it clear that CoV giving similar results to GRAW gave us pause about the value of PhysX when applied to games that just stick in some effects here and there. We also (I hope clearly) stressed that there isn't enough value in the product for consumers to justify a purchase at this time.
But we weren't overly hard on AGEIA as we could be for a couple reasons. First, CellFactor and HangarofDoom are pretty interesting demos. The performance of them and the possibilities presented by games like them indicate that PhysX could be more useful in the future (especially with its integration into UE3 and other game engines). Second, without more tools or games we just can't determine the actual potential of this hardware. Sure, right now developers aren't making practical use of the technology and it isn't worth its price tag. But it is very premature for us to stamp a "decelerator" label on it and close the case.
Maybe we will end up calling this thing a lemon, but we just need more hard data before we will do so.
Magnadoodle - Wednesday, May 17, 2006 - link
Yes, I understand your point of view, and I don't think you're pulling any punches or being biaised. In fact, a biting review would be more biaised than anything. I was just remarking that this would have made a perfect occasion to have a bit of fun with AGEIA and drag them through the dredges. I nostalgically recalled the quite biting and humorous style PC Gamer put into their 6% reviews. PC Gamer never was a pantheon of game reviewing, but they didn't have to be nice to nobody (actually to "nobodies", because they had to be nice to big corporations). My point was more about the lack of wits and style in web publications these days than about anandtech being biaised. Not that anandtech has bad writers, just that it's more scientific than sarcastic.Anyway, good review Mr. Wilson and keep up the good work.
Seer - Wednesday, May 17, 2006 - link
Im also wondering about this claim that the driver update increased framerates. In all but two of the tests, the avg fps was either the same or a decrease. The largest increase was 1 fps, totally within the margin of error. (I'm talking about the GRAW tests). So, um, yeah, no increase there.