After all this time they only have a beta driver? Really?
Yup and Crossfire is disabled.
I saw a couple of benchmarks today where the 64 actually beat the 1080Ti. One was WOW, IIRC. So there may be life in it yet.
That will be down GPU utilisation, My 1080 Ti sits at around 36% most of the time in WoW.
But but but AMD were working since like forever on the driver side....or so the fanboys kept crying.
But but but AMD were working since like forever on the driver side....or so the fanboys kept crying.
And as far as I can tell, since January Doom performance hasn't improved by very much. I really don't know what AMD have been doing since Vega silicon was first developed. Increasing clocks? No reason to do that when scaling is abysmal in performance per watt. Producing more chips? Where are they? Refining the process to reduce cost? Apparently prices are set to increase. It's starting to become more clear that it's High Bandwidth Memory that has been teh thorn in AMD's side. It's almost impossible to tell what Fiji and Vega could have been without it, but I'm starting to think it could have been a lot more for a lot less effort.
Maybe Vega is a pipe cleaner product and the real BIG product will arrive with Navi ?
Yeah, I'm thinking that as well, but some were saying that about Fiji/Polaris/Vega. 'Skip Fiji and Polaris; Vega is where it'll be at.' If Navi comes mid-2018, which it won't, maybe Radeon will have a chance in the high-end gaming sector, but I just can't see AMD ever getting back to being competitive in the high-end. I'm strongly considering keeping my Fury card and waiting for Volta. If only Nvidia supported Freesync.
Would be nice but putting the money aspect of it aside, Nvidia do have more control of configuring their G-Sync modules which can help bring latency, Frame rate window that G-Sync works with etc... down to levels that Freesync can't get to so there are plus sides to it.
Maybe Vega is a pipe cleaner product and the real BIG product will arrive with Navi ?
They said the same thing about Steamroller......then Excavator. Piledriver was just a stop gap CPU and Excavator was gonna clean house.
Not saying AMD can't right the GPU ship cause they most certainly did that with the CPU wing of the company but I think they're gonna have to go back to the drawing board to do so just like they did with Zen.
I'm sorry, but I don't entirely agree. Zens success was a result of two factors going in AMDs favour: 1) Their R&D team doing a phenomenal job of figuring out the best way forward (and they really did. Ryzen and Threadripper are awesome products for their price) AND 2) Intel being exceptionally lazy/resting on their laurels and only releasing minor improvements with the intervening chipsets while AMD were not competing.
I'm sorry, but I don't entirely agree. Zens success was a result of two factors going in AMDs favour: 1) Their R&D team doing a phenomenal job of figuring out the best way forward (and they really did. Ryzen and Threadripper are awesome products for their price) AND 2) Intel being exceptionally lazy/resting on their laurels and only releasing minor improvements with the intervening chipsets while AMD were not competing.
The net result of this was Zen not only being the fantastic product that it was going to be anyway, but bowling Intel over and having them scrambling.
Everything that I can see says that NVidia are smarter and more hardworking than this. They delayed Volta - which says that they either are not happy with its respective performance, OR, are deliberately holding it back while tweaking it so that when AMD DOES become competitive again (Navi maybe? Future driver enhancements?) they have an Ace up their sleeve. I have NO DOUBT that they watched what happened with Intel and thought "Nope. Not going to happen to us, if we can help it". So in actuality, it is probably a bit of both. The proof is in the pudding, in that they released the 10xx series while they still ruled the roost (albeit barely, in some cases) with the 9xx series.
What does this mean for AMD? As they did with the CPU side of things, they need to stop making products that "compete" with a given tier of NVidia, and start doing their own product independently of the NVidia product. Failure to do so means that they will always be on the back foot. Doing so means that they will be on the back foot for a while, until they can find the right mix of components that blows NVidia away (or at least, sends them running for the hills).
I might be wrong of course, I don't get paid the big dollars (and this might even be why). But to me, it seems sensible (although, not without risk).