Monday, December 12th 2016
AMD "Vega" Demoed in Sonoma, California
AMD's next-generation high-end graphics card, based on the "Vega" architecture, was showcased at an event in Sonoma CA, earlier this week. While the architecture is being debuted with the Radeon Instinct MI25 deep-learning accelerator, a prototype graphics card based on the silicon was exhibited by the company, showing Vulkan API gaming.
AMD was pretty tight-lipped about the specifications of this prototype, but two details appear to have slipped out. Apparently, the chip has a floating point performance of 25 TFLOP/s (FP16), and 12.5 TFLOP/s (FP32, single-precision). On paper, this is higher than the 11 TFLOP/s (FP32) of NVIDIA TITAN X Pascal. The other important specification that emerged is that the card features 8 GB of HBM2 memory, with a memory bandwidth of 512 GB/s. This, too, is higher than the 480 GB/s of the TITAN X Pascal. It remains to be seen which market-segment AMD targets with this card.
This article was updated on Dec 15 to accommodate AMD's request to remove all info regarding the demo system, the shown game and its performance, which has been put under NDA retroactively.
Source:
Golem.de
AMD was pretty tight-lipped about the specifications of this prototype, but two details appear to have slipped out. Apparently, the chip has a floating point performance of 25 TFLOP/s (FP16), and 12.5 TFLOP/s (FP32, single-precision). On paper, this is higher than the 11 TFLOP/s (FP32) of NVIDIA TITAN X Pascal. The other important specification that emerged is that the card features 8 GB of HBM2 memory, with a memory bandwidth of 512 GB/s. This, too, is higher than the 480 GB/s of the TITAN X Pascal. It remains to be seen which market-segment AMD targets with this card.
This article was updated on Dec 15 to accommodate AMD's request to remove all info regarding the demo system, the shown game and its performance, which has been put under NDA retroactively.
120 Comments on AMD "Vega" Demoed in Sonoma, California
And Doom well Meh.
12.5 GFlops 32bit
GP100
NVLink
10.6 GFlops 32bit
5.3 GFlops 64bit (Double-precision)
PCI-E
9.3 GFlops 32bit
4.6 GFlops 64bit (Double-precision)
Titan X (Pascal)
11 GFlops 32bit
i'd say in hectic action/scene it'd be averaging around 60-lower 70s,
Like i said Dooms pretty Meh.
And thats a really disappointing score, unless thats a small vega benchmark or running under OpenGL. Titan XP pushes a good 30% faster in DOOM.
But yes there is that reason as AMD have there name on it but the game is not all that graphical wise.
At least this time it's a mainstream title, so it's actually used in most reviews. We'll be spared the "why don't you test with Ashes of the Singularity" mantra... :P
It makes no sense to use DX11 games or games that are not optimized for gamecode of AMD.
We all know Nvidia uses a different approach when rendering frames. The positive is that it's usually faster in DX11 games but lacks in DX12.
Async compute is what favours AMD, and since we know the future is more aimed for putting a GPU to work the way it should, AMD is on the right side.
Gamedevs are having a less troubled coding of games since consoles use the same GCN hardware. Porting is easyer and it removed the obstacle the PS3 with the IBM Cell had.
Bottomline: Vega is ready.
And if they still use a CPU cooler then there wont be a release anytime soon.
I mean it definitely isn't a production cooler, but as an owner of that case I could see them having some major issues after they tapped almost all of the airflow off...
The GPU placement would have to be closer to the display outputs. Maybe a modified version of this. If they are fins because its almost touching the case i/o covers