Overclocking and Overclocked Performance
A little fiddling in GPU tweak allowed us to increase the core clock to 1139MHz. Anything after this was stopped by driver crashes.
3DMark Fire Strike
This is where things get slightly confusing. During conferences with AMD, the media were assured that the Fiji GPU cores and HBM combination would allow users to overclock much like the previous graphics cards. However, in a strange turn, AMD have locked memory overclocking and the GPU core was limited to around 10%. I believe that AMD are limiting the card through drivers as I didn’t see any obviously artifacts, which is what you normally can see when trying to overclock to the GPU limits.
Thanks for the review.
Why arkham Knight? not to mention no setting shown for it too. Perhaps you guys should benchmark The Witcher 3 instead of arkham knight. I sure hope amd will release a much better driver for fury x. Sad to see a new architecture(hbm) falling behind an old architechture(GDDR5). So much promise yet so much disappointment.
Arkham Knight has been withdrawn by Warner Brothers as the programming is seriously flawed. It is unplayable on DX11. WB will be releasing AK with a DX12 port and patch.
Why AMD why………..
So basically if you are gaming at 1440p or 4K this is worth serious consideration I am a little surprised at the price point, I did expect it to be a little cheaper being an AMD card but things do look promising for team red.
Unfortunately for me all of my gaming is done at 1080p so I personally would buy the 980Ti. Still I am interested in seeing where AMD take HBM.
When will you run the 3dMark API Overhead Feature Test?
@ Rikki Wright
When you last wrote about DX12 benchmarks you pointed out that Radeon 290x was 33% faster than GeForce 980Ti and as fast as Titan X.
So Fury should be 20% faster than Titan X and 50% faster than 980 Ti.
Right? You know, you ran the benchmarks already yet you aren’t telling us.
Peter Odonnell rote this piece 4 months ago.
www dot eteknix.com/amd-r9-290x-goes-head-to-head-with-titan-x-with-dx12/
Your own site knows that nVidia GPU’s are drastically slower than AMD using DX12 yet you lie to your readers by omitting these very important facts.
Don’t consumers have a right to make educated buying choices?
With DX12 Fury x is 50% faster than GTX 980 Ti and crushes Titan X.
nVidia does not have Asynchronous Shader Pipelines nor do they have Asynchronous Compute Engine IP.
nVidia is great with an obsolete API. But it is absolute trash with Mantle or DX12.