AMD Radeon HD 7970 Review: 28nm And Graphics Core Next, Together As One
by Ryan Smith on December 22, 2011 12:00 AM EST- Posted in
- GPUs
- AMD
- Radeon
- ATI
- Radeon HD 7000
Total War: Shogun 2
One of our goals with this iteration of our benchmark suite was to throw in some additional non-FPS games, so we’re broadening our horizons a bit by adding in Total War: Shogun 2. Shogun 2 is the latest installment of the long-running Total War series of turn based strategy games, and alongside Civilization V is notable for just how many units it can put on a screen at once. As it also turns out, it’s the single most punishing game in our benchmark suite.
When deciding on what settings to use with Shogun, it required a bit more creativity on our part. 2560 is a true moonshot; everything is turned on, and it takes a minimum of 1.5GB of VRAM to run the game at this resolution. Accordingly performance is rather dismal, though as this is a TBS 30fps isn’t quite as critical as it is in other games. In any case the 7970 comes the closest to hitting 30fps, coming in just shy at 28.2fps, which is 29% ahead of the GTX 580 and 48% ahead of the 6970.
Meanwhile for 1920 we turned Shogun’s settings down to Very High, and yet we still had to disable MSAA to make it work with 1GB cards (did we mention that Shogun loves VRAM?). At these lower settings performance rockets up, and at the same time so does the 7970’s lead over the GTX 580. Here it’s 36% ahead of the GTX 580, which will be the greatest lead among all of our gaming benchmarks. As for the 7970 compared to the 6970, it’s still 48% of the 6970 showing us just how similar these video cards are at times.
Finally at 1680 the overall performance goes up yet again, but the 7970’s lead remains. It’s ahead of the GTX 580 by 30%, and 48% ahead of the 6970.
Facebook
Twitter
RSS
292 Comments
View All Comments
Iketh - Thursday, December 22, 2011 - link
As mentioned several times in the article and in the comments, time was an issue. You can rest assured that follow-up articles are in the works.Ryan Smith - Thursday, December 22, 2011 - link
Indeed it is.Malih - Thursday, December 22, 2011 - link
dude, awesome in-depth (emphasizing on depth) review, thank you very much for the excellent work Ryan.Esbornia - Thursday, December 22, 2011 - link
After reading a half ass misinforming review full of errors and typos, I think you didn't read it to say something like that.Iketh - Thursday, December 22, 2011 - link
It is full of typos, but that has nothing to do with in-depth. It was certainly in-depth and a joy to read despite the typos.I'd like to know what you believe is misinformation though.
SlyNine - Thursday, December 22, 2011 - link
He probably couldn't understand alot of it and though they were all typo's.WhoBeDaPlaya - Thursday, December 22, 2011 - link
Sod off you wanker. Go and read Walmart reviews for this cart - they're probably more at your level ;)Marburg U - Thursday, December 22, 2011 - link
Does Eyefinity Technology 2.0 allow me to launch an application within Windows ON WHICH MONITOR I WANT?NikosD - Thursday, December 22, 2011 - link
It seems that nobody noticed but where is FP64 = 1/2 FP32 performance that AMD said back in June when they first introduced CGN architecture ?I copy from Ryan's June article:
"One thing that we do know is that FP64 performance has been radically improved: the GCN architecture is capable of FP64 performance up to ½ its FP32 performance. For home users this isn’t going to make a significant impact right away, but it’s going to help AMD get into professional markets where such precision is necessary."
The truth is that FP64 is 1/4 of FP32 eventually!
Big Loss in GPGPU community even if 7970 is capable of 3.79Tflops of FP32 compared to 2.7Tflops of 6970
R3MF - Thursday, December 22, 2011 - link
it says 1/2 in the architecture article, but 1/4 in the consumer product review, is this AMD taking a leaf from Nvidia's (shitty) book of using drivers to disable features in non-professional (price-tag) products?