ATI Radeon HD 4890 vs. NVIDIA GeForce GTX 275
by Anand Lal Shimpi & Derek Wilson on April 2, 2009 12:00 AM EST- Posted in
- GPUs
New Drivers From NVIDIA Change The Landscape
Today, NVIDIA will release it's new 185 series driver. This driver not only enables support for the GTX 275, but affects performance in parts across NVIDIA's lineup in a good number of games. We retested our NVIDIA cards with the 185 driver and saw some very interesting results. For example, take a look at before and after performance with Race Driver: GRID.
As we can clearly see, in the cards we tested, performance decreased at lower resolutions and increased at 2560x1600. This seemed to be the biggest example, but we saw flattened resolution scaling in most of the games we tested. This definitely could affect the competitiveness of the part depending on whether we are looking at low or high resolutions.
Some trade off was made to improve performance at ultra high resolutions at the expense of performance at lower resolutions. It could be a simple thing like creating more driver overhead (and more CPU limitation) to something much more complex. We haven't been told exactly what creates this situation though. With higher end hardware, this decision makes sense as resolutions lower than 2560x1600 tend to perform fine. 2560x1600 is more GPU limited and could benefit from a boost in most games.
Significantly different resolution scaling characteristics can be appealing to different users. An AMD card might look better at one resolution, while the NVIDIA card could come out on top with another. In general, we think these changes make sense, but it might be nicer if the driver automatically figured out what approach was best based on the hardware and resolution running (and thus didn't degrade performance at lower resolutions).
In addition to the performance changes, we see the addition of a new feature. In the past we've seen the addition of filtering techniques, optimizations, and even dynamic manipulation of geometry to the driver. Some features have stuck and some just faded away. One of the most popular additions to the driver was the ability to force Full Screen Antialiasing (FSAA) enabling smoother edges in games. This features was more important at a time when most games didn't have an in-game way to enable AA. The driver took over and implemented AA even on games that didn't offer an option to adjust it. Today the opposite is true and most games allow us to enable and adjust AA.
Now we have the ability to enable a feature, which isn't available natively in many games, that could either be loved or hated. You tell us which.
Introducing driver enabled Ambient Occlusion.
What is Ambient Occlusion you ask? Well, look into a corner or around trim or anywhere that looks concave in general. These areas will be a bit darker than the surrounding areas (depending on the depth and other factors), and NVIDIA has included a way to simulate this effect in it's 185 series driver. Here is an example of what AO can do:
Here's an example of what AO generally looks like in games:
This, as with other driver enabled features, significantly impacts performance and might not be able to run on all games or at all resolutions. Ambient Occlusion may be something some gamers like and some do not depending on the visual impact it has on a specific game or if performance remains acceptable. There are already games that make use of ambient occlusion, and some games that NVIDIA hasn't been able to implement AO on.
There are different methods to enable the rendering of an ambient occlusion effect, and NVIDIA implements a technique called Horizon Based Ambient Occlusion (HBAO for short). The advantage is that this method is likely very highly optimized to run well on NVIDIA hardware, but on the down side, developers limit the ultimate quality and technique used for AO if they leave it to NVIDIA to handle. On top of that, if a developer wants to guarantee that the feature work for everyone, they would need implement it themselves as AMD doesn't offer a parallel solution in their drivers (in spite of the fact that they are easily capable of running AO shaders).
We haven't done extensive testing with this feature yet, either looking for quality or performance. Only time will tell if this addition ends up being gimmicky or really hits home with gamers. And if more developers create games that natively support the feature we wouldn't even need the option. But it is always nice to have something new and unique to play around with, and we are happy to see NVIDIA pushing effects in games forward by all means possible even to the point of including effects like this in their driver.
In our opinion, lighting effects like this belong in engine and game code rather than the driver, but until that happens it's always great to have an alternative. We wouldn't think it a bad idea if AMD picked up on this and did it too, but whether it is more worth it to do this or spend that energy encouraging developers to adopt this and comparable techniques for more complex writing is totally up to AMD. And we wouldn't fault them either way.
294 Comments
View All Comments
evilsopure - Thursday, April 2, 2009 - link
Update: I guess Anand was making his updates while I was making my post, so the "marginal leader at this new price point of $250" line is gone and the Final Words actually now reflect my own personal conclusion above.Anand Lal Shimpi - Thursday, April 2, 2009 - link
I've updated the conclusion, we agree :)-A
SiliconDoc - Monday, April 6, 2009 - link
You agree now that NVidia has moved their driver to the 2650 rez to win, since for months on end, you WHINED about NVidia not winning at the highest rez, even though it took everyting lower.So of COURSE, now is the time to claim 2650 doesn't matter much, and suddenly ROOT for RED at lower resolutions.
It Nvidia screws you out of cards again, I certainly won't be surprised, because you definitely deserve it.
Thanks anyway for changing Derek's 6 month plus long mindset where only the highest resolution mattered, as he had been ranting and red raving how wonderful they were.
That is EXACTLY WHY his brain FARTED, and he declared NVidia the top dog - it's how he's been doing it for MONTHS.
So good job there, you BONEHEAD - you finally caught the bias, just when the red rooster cards FAILED at that resolution.
Look in the mirror - DUMMY - maybe you can figure it out.
7Enigma - Thursday, April 2, 2009 - link
Check the article again. Anand edited it and it is now very clear and concise.7Enigma - Thursday, April 2, 2009 - link
Bah, internet lag. Ya got there first.... :)sublifer - Thursday, April 2, 2009 - link
As I predicted elsewhere, they probably should have named this new card the GTX 281. In almost every single benchmark and resolution it beats the 280. In one case it even beat the 285 somehow./Gripe
That said, Go AMD! I wanna check other sites and see if they benched with the card highly over-clocked. One site got 950 core and 1150 memory easily but they didn't include it on the graphs :(
Anand Lal Shimpi - Thursday, April 2, 2009 - link
Hey guys, I just wanted to chime in with a few fixes:1) I believe Derek used the beta Catalyst driver that ATI gave us with the 4890, not the 8.12 hotfix. I updated the table to reflect this.
2) Power consumption data is now in the article as well, 2nd to last page.
3) I've also updated the conclusion to better reflect the data. What Derek was trying to say is that the GTX 275 vs. 4890 is more of a wash at 2560 x 1600, which it is. At lower than 2560 x 1600 resolutions, the 4890 is the clear winner, losing only a single test.
Thank you for all the responses :)
Take care,
Anand
7Enigma - Thursday, April 2, 2009 - link
Thank you Anand for the update and the article changes. I think that will quell most of the comments so far (mine included).Could you possibly comment on the temps posted earlier in the comments section? My question is whether there are significant changes with the fan/heatsink between the stock 4870 and the 4890. The idle and load temps of the 4890 are much lower, especially when the higher frequency is taken into consideration.
Also a request to describe the differences between the 4890 and the 4870 (several comments allude to a respin that would account for the higher clocks, lower temp, different die size).
Thank you again for all of your hard work (both of you).
Warren21 - Thursday, April 2, 2009 - link
Yeah, I would also second a closer comparison between RV790 and RV770, or at least mention it. It's got new power phases, different VRM (7-phase vs 5-phase respectively), slightly redesigned core (AT did mention this) and features a revised HS/F.VooDooAddict - Thursday, April 2, 2009 - link
I was very happy to see the PhysX details. I'd started worrying I might be missing out with my 4870. It's clear now that I'm not missing out on PhysX, but might be missing out on some great encoding performance wiht CUDA.I'll be looking forward to your SLI / Crossfire followup. Hoping to see some details about peformance with ultra high Anti-Aliasing that's only available with SLI/Crossfire. I used to run Two 4850s and enjoyed the high-end Edge Antialiasing. Unfortunetly the pair of 4850's were a too much heat in a tiny shuttle case so I had to switch out to a 4870.
Your review reinforced something that I'd been feeling about the 4800s. There isn't much to complain about when running 1920x1200 or lower with modest AA. They seem well positioned for most gamers out there. For those out there with 30" screens (or lusting after them, like myself)... while the GTX280/285 has a solid edge, one really needs SLI/Crossfire to drive 30" well.