The Radeon HD 4850 & 4870: AMD Wins at $199 and $299
by Anand Lal Shimpi & Derek Wilson on June 25, 2008 12:00 AM EST- Posted in
- GPUs
Fixing AMD's Poor AA Performance
Now that we have a new architecture from AMD with improved AA performance, it's time again to look at a comparison of all the different AA modes these cards offer. No new modes have been introduced since the R600 and G80 reviews, but AMD has completely rebuilt their ROPs with special attention to hardware based AA resolve. In R600, hardware resolve wasn't much faster than shader based resolve, but this time around, AA runs blazingly fast whether its on the dedicated resolve hardware or on the shader hardware (since their is so much more shader hardware now even shader based resolve gets a giant boost).
The first thing we will want to look at are the MSAA modes. These are the modes we absolutely recommend for use with AMD hardware as all their other filters essentially low-pass filter the entire image by blending in neighboring subpixels. In any case, the results are very impressive for RV770.
The RV670 at 19x12 was limited in some way other than AA (it really couldn't keep up), but at 16x10 we can get a better idea of relative impact of AA. And clearly the RV770 quite improves fall off with increasing AA levels over the previous generation. One special thing to note is that the RV770 does fall off very gracefully to 8xAA. Since the RV670, G80 and GT200 all have sharp drops in performance when moving up from 4xAA to 8xAA, the RV770 really shines here. In fact, the few tests we did with 8xAA paints the 4870 in a much better light relative to the GTX 280. Remember from our earlier architecture discussion that Oblivion is the game where the GT200 had the largest performance advantage over RV770.
While 8xAA performance is all well and good, the image quality difference is just not enough for us to recommend enabling it over increasing resolution (or better yet, pixel density on LCD panels -- hello display makers). For those with panels that don't go over 1280x1024, it would be better to spend the extra money on a large panel than a $300 graphics card. The application where we see 8xAA making the most sense is on 50+ HDTVs used as computer monitors where the pixels are just plain huge. For the majority of desktop users though 4xAA is where it's at.
We did test the performance of all the other modes as well. NVIDIA's CSAA modes are quite good (they actually improve image quality rather than degrade it), but again, stay away from anything but AMD's "box" filtered AA.
The RV770 actually shows a bigger performance hit from enabling their tent filters than RV670. This is likely because the filters are run on shader hardware in both cases while RV770 has faster hardware resolve that can be used for normal AA resolve. If RV670 resolves "box" filtered AA on the shader as well this would explain the flatter performance in that case. Even more so than the image quality question, the fact that they perform lower really should be the nail in the coffin for AMD's tent filter garbage.
Facebook
Twitter
RSS
215 Comments
View All Comments
BusterGoode - Sunday, June 29, 2008 - link
Thanks, great article by the way Anandtech is my first stop for reviews.jay401 - Wednesday, June 25, 2008 - link
Good but I just wish AMD would give it a full 512-bit memory bus bandwidth. Tired of 256-bit. It's so dated and it shows in the overall bandwidth compared to NVidia's cards with 512-bit bus widths. All that fancy GDDR4/5 and it doesn't actually shoot them way ahead of NVidia's cards in memory bandwidth because they halve the bus width by going with 256-bit instead of 512-bit. When they offer 512-bit the cards will REALLY shine.Spoelie - Thursday, June 26, 2008 - link
Except that when R600 had a 512bit bus, it didn't show any advantage over RV670 with a 256bit bus. And that was with GDDR3 vs GDDR3, not GDDR5 like in RV770 case.JarredWalton - Thursday, June 26, 2008 - link
R600 was 512-bit ring bus with 256-bit memory interface (four 64-bit interfaces). http://www.anandtech.com/showdoc.aspx?i=2552&p...">Read about it here for a refresh. Besides being more costly to implement, it used a lot of power and didn't actually end up providing provably better performance. I think it was an interesting approach that turned out to be less than perfect... just like NetBurst was an interesting design that turned out to have serious power limitations.Spoelie - Thursday, June 26, 2008 - link
Except that it was not, that was R520 ;) and R580 is the X19x0 series. That second one proved to be the superior solution over time.R600 is the x2900xt, and it had a 1024bit ring bus with 512bit memory interface.
DerekWilson - Sunday, June 29, 2008 - link
yeah, r600 was 512-bithttp://www.anandtech.com/showdoc.aspx?i=2988&p...">http://www.anandtech.com/showdoc.aspx?i=2988&p...
looking at external bus width is an interesting challenge ... and gddr5 makes things a little more crazy in that clock speed and bus width can be so low with such high data rates ...
but the 4870 does have 16 memory modules on it ... so that's a bit of a barrier to higher bit width busses ...
JarredWalton - Wednesday, June 25, 2008 - link
I'd argue that the 512-bit memory interface on NVIDIA's cards is at least partly to blame for their high pricing. All things being equal, a 512-bit interface costs a lot more to implement than a 256-bit interface. GDDR5 at 900MHz is effectively the same as GDDR3 at 1800MHz... except no one is able to make 1800MHz GDDR3. Latencies might favor one or the other solution, but latencies are usually covered by caching and other design decisions in the GPU world.geok1ng - Wednesday, June 25, 2008 - link
The tests showed what i feared: my 8800GT is getting old to pump my Apple at 2560x1600 even without AA! But the tests also showed that the 512MB of DDR5 on the 4870 justifies the higher price tag over the 4850, something that the 3870/3850 pair failed to demonstrate. It remains the question: will 1GB of DDR5 detrone NVIDIA and rule the 30 inches realm of single GPU solutions?IKeelU - Wednesday, June 25, 2008 - link
"It is as if AMD and NVIDIA just started pulling out hardware and throwing it at eachother"This makes me crack up...I just imagine two bruised and sweaty middle-aged CEO's flinging PCBs at each other, like children in a snowball fight.
Thorsson - Wednesday, June 25, 2008 - link
The heat is worrying. I'd like to see how aftermarket coolers work with a 4870.