AMD Vega 10, Vega 11, Vega 12 and Vega 20 Rumors and Discussion

has anyone done any investigations on this vega fe? comparing dx11 and 12 performance, vulkan and vega in games where a major gain was seen on older hardware? async on and async off to test if that is working? things that will highlight potential dysfunction in the drivers. Or even taking a look in the drivers themselves comparing to fiji drivers, trying to find the date they were compiled or w.e.

anything beyond these meh benchmarks?
 
has anyone done any investigations on this vega fe? comparing dx11 and 12 performance, vulkan and vega in games where a major gain was seen on older hardware? async on and async off to test if that is working? things that will highlight potential dysfunction in the drivers. Or even taking a look in the drivers themselves comparing to fiji drivers, trying to find the date they were compiled or w.e.

anything beyond these meh benchmarks?
There is probably a reason AMD is suggesting not to bother benchmarking. Performs like Fiji, drivers look like Fiji, nothing official from AMD or a gaming product yet. It really seems to come down to AMD wanting an extra month to get drivers in order.
 
The link you have posted shows 10 graphs comparing Vega FE and P5000 results. Vega FE is faster than P5000 im half of them, so both cards are very comparable. Try to compare price / performance.
but Quadro comes with certified drivers, not Vega Fe. Proper price/performance comparison will be with Vega WX
 
Quadro and Vega WX are aimed at professional users while Vega FE is aimed at "Pioneers", and everybody knows that "Pioneers" do not require certified or even functional drivers when they are busy pioneering.
That actually remind be the alien covenant movie where "Pioneers" were so busy pioneering that they forgot to use helmet .
 
Quadro and Vega WX are aimed at professional users while Vega FE is aimed at "Pioneers", and everybody knows that "Pioneers" do not require certified or even functional drivers when they are busy pioneering.

Have you heard about this cool new thing called Sparse Voxel Octrees? I heard it's totaly going to disrupt 3d graphics!
 
That actually remind be the alien covenant movie where "Pioneers" were so busy pioneering that they forgot to use helmet .
Hey, our scanner says the air is breathable just outside the ship, so let's all expose ourselves to all the airborne bacteria and viruses, plus all toxic and/or radioactive stuff while we stroll around.

22nd century men and women sure seem dumb as a door in that movie.
 
Always is a strong word. :)


„The driver“ being which components specifically?
The file quakechamps.ico is dated 4th of june, the VulkanRT installer was signed on april 6th. Some of the OpenCL files were signed on june 25th. So, I think "the driver is from january" is a bold statement.

The OpenGL driver for example is version 6.14.10.13486, the one from the June Catalyst Package 17.6.2 is 6.14.10.13476, the DirectX Universal driver (atiumdag.dll) has version 9.14.10.01280, the 17.6.2-version is 9.14.10.01261. Just for reference.

Beats me, but most likely it's just the original branch upon which the driver is based. The driver installed declares the "Radeon Software Version" as 17.1.1.
 
Beats me, but most likely it's just the original branch upon which the driver is based. The driver installed declares the "Radeon Software Version" as 17.1.1.
I do not think that this is the most likely explanation. But even if that was the case, the driver seemingly has had as much or even more revisions than the official RSCE 17.6.2 in many important aspects and in no way could be called "old".
 
You could use the "old" Fiji PRO branch of the driver, make it running on VEGA FE (without putting in TBR support/activation). Of course you would polish the driver for PRO software to (at least) not crash during the benchmarks... and you polish it for the VEGA FE release day.

I think there are many ways a driver could be called old.
 
If it's a architecture problem, what's next for amd after such a failure ? Navi ? Or could they kind of work on it pretty fast, like nVidia nv30 to nv35 and released another chip ?
 
If it's a architecture problem, what's next for amd after such a failure ? Navi ? Or could they kind of work on it pretty fast, like nVidia nv30 to nv35 and released another chip ?

It might not be the overall architecture but just this (first) implementation of it. It was almost exactly a decade ago when HD 2900 came out and AMD looked like they were in deep trouble. A year later RV770 shook the competitive landscape and they right back in the mix. With multiple teams working semi-independently on succeeding designs, you can usually stem the bleeding with an improved revision within 6 months and then potentially reverse the position entirely within a year with next generation.
 
Maybe it is a bit too early to write off Vega.

Maybe... But right now it's between a 1070 and a 1080, with a lot of power draw, and approx one year later. And it seems that Fiji is more efficient clock for clock... I hope the driver has some "secret sauce", or that some Vega new stuff were not enabled for the FE version...
 
Last edited:
Maybe... But right now it's between a 1070 and a 1080, with a lot of power draw, and approx one year later. And it seems sans Fiji is more efficient clock for clock... I hope the driver has some "secret sauce", or that some Vega new stuff were not enabled for the FE version...

Raja himself has stated that because the architecture is new, they would have a lot of work with the drivers.
Let's wait and see what happens when the RX Vega is released.
 
Back
Top