Ryan quite clearly states it will be enabled at launch. Launch is in two weeks.
While I don't expect it to make a huge difference the comment that Vega currently has features disabled is supported by AMDs own words.
Ryan quite clearly states it will be enabled at launch. Launch is in two weeks.He says they are not enabled the Frontier Edition, which everyone knows already.
I am talking about RX Vega. Which is supposed to have them, which AMDs slides with performance numbers should already reflect.
That can't be used as an excuse anymore for RX Vega's poor numbers in AMDs own RX Vega numbers.
Ryan quite clearly states it will be enabled at launch. Launch is in two weeks.
While I don't expect it to make a huge difference the comment that Vega currently has features disabled is supported by AMDs own words.
Wondering where did most of Vega's transistors go? Just to increase clock speeds!Talking to AMD’s engineers, what especially surprised me is where the bulk of those transistors went; the single largest consumer of the additional 3.9B transistors was spent on designing the chip to clock much higher than Fiji. Vega 10 can reach 1.7GHz, whereas Fiji couldn’t do much more than 1.05GHz. Additional transistors are needed to add pipeline stages at various points or build in latency hiding mechanisms, as electrons can only move so far on a single clock cycle; this is something we’ve seen in NVIDIA’s Pascal, not to mention countless CPU designs. Still, what it means is that those 3.9B transistors are serving a very important performance purpose: allowing AMD to clock the card high enough to see significant performance gains over Fiji.
The only "currently" for the public, is the Frontier Edition. It is inaccurate to say Vega RX doesn't have these features running.
Launch is when they go public. That doesn't mean they haven't been running for months internally.
If Vega hasn't already been running with these features enabled internally for weeks, if not months, they aren't shipping with them enabled publicly, in two weeks.
Those AMD slides represent Vega RX with it's features running. It is really grasping at straws to think AMD revealed performance numbers that are meaningless this close to launch.
I think they aimed for much higher clock frequency but failed. I guess GCN just doesn't scale.
As a professional/compute card, Vega isn't too bad. Radeon WX 9100 should sell decently; at $2,199 it's only a bit more expensive than Quadro P5000, but far more powerful.
The problem isn't that AMD is not competing in the high-end gaming segment; writing that off is a plausible business decision. The problem is that their marketing team was so flagrantly dishonest about it. I've lost a lot of respect for AMD over this fiasco. Any gamer who bought into the hype has been stuck with lesser products for months "waiting for Vega" when they could have had a superior Nvidia card at a price no higher than Vega will now cost. Worse, prices have become inflated from the mining craze, with the GTX 1070 almost unavailable and the GTX 1080 temporarily back up in price, so anyone who waited is likely to now incur a real financial loss unless they are willing to wait longer.
Well, Vega is a complete failure for gamers. Months ago, I was hoping for at least -10% from a 1080ti. Total disappointment. AMD have regressed in essentially every parameter, perf/watt, perf/mm^2 etc, without solving the essential issues that held Fiji back. The only thing that's holding me back from buying nvidia is Freesync (I haven't bought a monitor yet, I'm waiting for the C27HG70).
All AMD had to do was to remove the bottlenecks of Fiji then add +50% to everything. Think about it. A Vega with 1TB/s mem bandwith 6000+ SP and crapload of raw geometry performance. At ~1400-1500Mhz. Would have destroyed 1080Ti. But what do they do? They go for professional market and spend all transistor budget on features no gamer needs.
Hell. Even 2x Polaris would had been better.
Thankfully, Ryzen is good, because otherwise, this could be company destroying for AMD.
They must have known it was going to be this bad, even before tape out, and yet they just went along with this train wreck. It doesn't bode well for bouncing back.
I really think they need outside talent to straighten out their GPU design, but where would that come from these days?
Vega Technical Overview.
https://www.techpowerup.com/reviews/AMD/Vega_Microarchitecture_Technical_Overview
Welp AMD just solidified their hold on mining. Vega has specific instructions for mining. This explains the reports I've see that Vega was really good for mining.
Vega 56 at $400 and 210W TDP (and going by recent AMD offerings, about 90-95% of Vega 64) sounds like an excellent deal. It reminds me of R9 290. I expect good overclocks as well, getting the card to stock Vega 64 performance levels.
No reviews yet but I've seen reports for Vega 64 being 15% faster than 1080. The Vega 56 should be equal or maybe slightly better for $399. And they're listing TDP at ~220W IIRC. Assuming performance is as I've seen rumored most recently, why wouldn't that $399 card sell really well? Keep in mind that quite a few people prefer AMD's open ecosystem where they aren't getting gouged for every new feature. I'll take Freesync support and the best support for Vulcan and DX12, GTX 1080 performance, for $400.
I haven't looked back to verify those numbers. It's my best recollection. I believe they are correct though.
Vega Technical Overview.
https://www.techpowerup.com/reviews/AMD/Vega_Microarchitecture_Technical_Overview
Welp AMD just solidified their hold on mining. Vega has specific instructions for mining. This explains the reports I've see that Vega was really good for mining.
If they had wanted the pro crowd they would have spent a bit more on the hardware - quad stacked HBM with ECC support - while dropping clocks so it worked more efficiently. Instead of wasting all those transistors on failing to clock high, they would have used some giving decent 32 float support.Well, if there is any good news here, it sure isn't aimed at the gaming crowd, that is more of a thud than anything else.
Vega is squarely geared toward the more professional crowd, and from what we see from the show, that will be their main focus.
At least they will make more $$$ selling to that crowd than the gaming crowd.
That's why "ecosystems" are crap and people should stop buying into them unless there's very good reasons. Variable refresh isn't good enough that I'd let that dictate multiple $300+ purchases. I'm genuinely baffled at how obsessive over this people seem to be, as I think it is one of the biggest shams perpetrated in gaming. It never should have happened either because this never should have been an issue (adaptive sync should have been something that occurred long before now). Hopefully it being part of DP spec will put an end to this particular madness.
AMD always had more direct hardware command calls on their GPUs, they are just continuing their lead in that area. The side effect of this was their improved mining performance for many years. Looks like that may continue.
What reports? Everything I've seen points to ~33mh/s for ETH which is ok until you look at the power consumption to do it which is over 2x what a 1070 pulling ~31mh/s uses.
With 75% of the market and little competition at the high end, they don't really have to care about VRR. They've solidified themselves in consumers minds as the premium product and G-Sync is just that, the premium VRR solution.It is ridiculous that nvidia doesn't support the standard. I wouldn't be surprised if monitor manufacturers liked the extra margins and have told nvidia to take their time moving away from gsync.
With 75% of the market and little competition at the high end, they don't really have to care about VRR. They've solidified themselves in consumers minds as the premium product and G-Sync is just that, the premium VRR solution.
Correct. AMD had 1 instruction that was very useful for mining computations that Nvidia didn't support.
From people writing the mining software.
Monitor manufacturers don't care - they probably get a bigger cut with the more expensive g-sync monitors so if people buy them they're happy. The only people that can change Nvidia's mind are the consumers by not buying Nvidia due to that feature. It also has to be said VRR is not the better solution, it's just the cheaper one, and it's only cheaper because Nvidia charge for g-sync not because it really costs much more for the hardware. VRR has a distinct lack of quality control with half the monitors having bad implementations (e.g. 48-75hz only), all the g-sync monitors work well as Nvidia won't certify anything that doesn't. It's this sort of thing that makes Nvidia appear "premium" and hence people buy them not the competition despite the cost.VRR is awesome and I will not buy a monitor in the future that doesn't have it. The ability to play games at 80-120 fps without ever having to worry about screen tearing or leaving my hardware under utilized is great.
It is ridiculous that nvidia doesn't support the standard. I wouldn't be surprised if monitor manufacturers liked the extra margins and have told nvidia to take their time moving away from gsync.