About 13% higher than RX Vega 64. Might be something not working with Vega in this particular test.
But I see very high RX Vega 64 results in the FS tests, so this might mean nothing.
https://www.3dmark.com/spy/4051760
-
Competitor rules
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
Radeon VII
- Thread starter DragonQ
- Start date
About 13% higher than RX Vega 64. Might be something not working with Vega in this particular test.
But I see very high RX Vega 64 results in the FS tests, so this might mean nothing.
https://www.3dmark.com/spy/4051760
Seems about right going by what AMD said
Soldato
8700 is low for Time Spy..
I get Graphics Score 7 627 with my Vega 56 (1717 Core Clock)
https://www.3dmark.com/spy/5028536
Something wrong with those results.
I get Graphics Score 7 627 with my Vega 56 (1717 Core Clock)
https://www.3dmark.com/spy/5028536
Something wrong with those results.
This is the thing, 7nm gives them about 25% higher clocks at the same power, but they have disabled 4CU, so you expect final performance around 17%. Only when there is a memory bandwidth limitation will performance increase much beyond that.
Soldato
Oh dear, we are in for a massive dissapointment, or this guy is talking bull on twitter
Soldato
Whats new AMD
Soldato
Well we dont know the clock speeds right now, maybe they are running 25% slower clocks and then doing maths like we do on these forums lol
Soldato
Getting a 15% increase over a pretty overclocked last gen isn't bad when it's also probably operating at lower clocks. Also not factoring in anything over and above what it has been stated to do.
Associate
7600 is about right for a 56 on timespy with those clocks, i think hes expecting to much from v7 in comparison.
- Score 7215, GPU Vega56 @1700/980, GFX Score 7693, CPU Score 5339, CPU 6700k @4.4, Post No.1403, henson0115 - Link Drivers 18.11.2
Last edited:
Soldato
Guess you forgot there is another Vega card between the 56 and the R7..
So the performance from a Vega 64 is gona look even poorer.
That said, theres no clock speed details so its all pointless right now.
My stock vega 56 is about 6500.
They do go another grand with under volting and oc.
Soldato
It looks very disappointing to me, particularly in Timespy.
Soldato
TimeSpy is proven for many years now, doesn't use properly the DX12 API but a heavily Nvidia favoring code optimization.
(i can give you the technical data analysis).
Firestrike on the other hand, doesn't and is the better benchmarks to compare.
Until reviews grab some salt. Someone posted some numbers without further data, and it's perf looks more like my Vega 64 than a Radeon VII
https://www.3dmark.com/fs/16938414
Associate
i didnt forget anything, i simply made a comment on the 56 and v7 i think your quoting the wrong post.
p.s we all knew from the get go that there's no real world performance difference between the 56 and 64.
Last edited:
Associate
yep spot on, my reference 56 undervolted, which increases the clocks by default anyway. actually the lower you drop the temps the higher the core will go, even past whats set in the panel i/e 1500 set, can go to 1700+ and up to 2k with extreme cooling. so which stock are you referring to reference or aib as they are competency different beasts in the cooling department.
anyway its all pointless as benchmark runs in firestrike/timespy whatever do not mean didly squat in terms of real word performance or gaming performance, so not sure why many people arnt simpyl waiting for the real results.
Especially in winter, go outside, leave your computer to freeze a bit and then run the tests at sub zero ambient temperature. The GPU will be very cold.