Great news! That means we only have to wait 365 days 'till we see something at decent/consumer prices.

New hardware is always something nice! Bring it on AMD!

7nm... Is it really happening? Sweet!

AMD beat Intel onto 7nm??? (even though it's GPUs not CPUs). I'd love to see AMD competing better in the GPU market like they are now doing in CPU market. Intel/Nvidia have treated us badly for too long!

fatboyslimerr:

AMD beat Intel onto 7nm??? (even though it's GPUs not CPUs). I'd love to see AMD competing better in the GPU market like they are now doing in CPU market. Intel/Nvidia have treated us badly for too long!

Intel already selling 10 nm Cannonlakes. And this is not a gaming GPU. There will be no mainstream GPU from AMD this year. Only budget RX600 series on "12 nm". "7 nm" Navi not earlier of April-May 2019. And it seems likely Nvidia Turning was cancelled because of that. Likely no GTX 1170 and 1180 this year. "7 nm" Nvidia Ampere likely around March-May 2019 as well.

Intel sells low quantity 10nm cannonlakes for laptops or something like that. No mass produced desktop parts spotted anywhere. That said, I think those 7nm is comparable to 10nm from Intel, has to do with measurements and marketing (because 7 is smaller than 10, doesn't mean it's hugely better). But, Ryzen 2 on 7nm will surely be on par with current 14nm Intel chips. As for updates on GPUs, everyone is keeping their mouth shut. HBM2 being so expensive, this new card will have to really perform or it will suck.

this gona be an Draw Power optimization, not performance per clock optimization, but maybe, they can optimize clocks too, well, 7nm is a thing, next step... greetings

Amx85:

this gona be an Draw Power optimization, not performance per clock optimization, but maybe, they can optimize clocks too, well, 7nm is a thing, next step... greetings

Actually, since previous Vega was made on 14nmLPP, that process contributed to higher power draw on the GPU because it was designed for low clocks and mobile parts. Plus, 14nmLPP resulted in lower yields and AMD was forced to increase the voltages on GPU's (which contributed to high power consumption). Also, AMD gave Vega on average 40% more compute units than Nvidia gave Pascal CUDA cores... that also contributed to higher power consumption on Vega. But, 7nm TSMC process (on which this new Vega will be placed) will be designed for high clocks and efficiency... and the benchmarks suggest about 80% increase in performance over existing Vega GPU's with power draw at about 150W - maybe 180W. Essentially, 7nm should allow Vega to be clocked much higher on the core and HBM while simultaneously reducing power draw. I suspect that power draw might remain a bit higher on this new Vega (in comparison to Tesla GPU's) because it will still have a higher number of compute units... but the performance should also likely increase. So it won't just be optimized power draw. We don't know what the yields will look like though, so that's another question, however, the SHOULD be a lot better because TSMC has experience in producing GPU's anyway, and AMD already has a functioning 7nm Vega silicon in the lab with sampling happening later this year, and mass scale availability (probably for consumers) coming early next year (this would be Navi). AMD will likely release Vega on 7nm for professional purposes and AI.

@Hilbert, got some weird typos in the post-- "AMD itself has mentioned the existence of VEGA0 multiple ties away and earlier on mentioning that it would be launching in 218, according to their plans."

deksman2:

80% increase in performance over existing Vega GPU's with power draw at about 150W - maybe 180W.

Some house holds have a curse word jar, i think in the forum we need a hype jar and you sir need to put a dollar in it !

Venix:

Some house holds have a curse word jar, i think in the forum we need a hype jar and you sir need to put a dollar in it !

When you take into account the fact that 14nmLPP was designed for low clocks and mobile parts, produced low yields (Which contributed to higher voltages) and Vega in general had 40% more compute units than Pascal, you easily start to see why people were 'disappointed' with Vega. As an architecture, Vega is actually ok... and achieves parity with Pascal on lower frequencies... but the manuf. process it was made on was not its friend. When clocked to 1200MhZ on core, Vega is quite power efficient. The 14nmLPP is also what prevented Ryzen from clocking much higher than 4Ghz. Had Vega been made on 16nm TSMC process, it would likely be able to clock much higher on core and HBM with much lower power draw and effectively surpass Pascal in performance. Considering that 7nm TSMC process allows 35-40% increase in performance over 16nm TSMC ,you have to bear in mind that for Vega, this would translate to much higher performance gains and far lower power draw. I still expect Vega to draw a bit more power due to having more compute units... but nowhere near the power draw as it did on 14nmLPP even with increased clocks. You need to pay more attention to manuf. processes and how they affect clocks, efficiency, yields, etc. Also, hype has nothing to do with this, because leaked benchmarks of Vega 20 on 7nm suggests 3% higher performance on 7nm while running on much lower clocks in comparison to Vega 64.

deksman2:

When you take into account the fact that 14nmLPP was designed for low clocks and mobile parts, produced low yields (Which contributed to higher voltages) and Vega in general had 40% more compute units than Pascal, you easily start to see why people were 'disappointed' with Vega. As an architecture, Vega is actually ok... and achieves parity with Pascal on lower frequencies... but the manuf. process it was made on was not its friend. When clocked to 1200MhZ on core, Vega is quite power efficient. The 14nmLPP is also what prevented Ryzen from clocking much higher than 4Ghz. Had Vega been made on 16nm TSMC process, it would likely be able to clock much higher on core and HBM with much lower power draw and effectively surpass Pascal in performance. Considering that 7nm TSMC process allows 35-40% increase in performance over 16nm TSMC ,you have to bear in mind that for Vega, this would translate to much higher performance gains and far lower power draw. I still expect Vega to draw a bit more power due to having more compute units... but nowhere near the power draw as it did on 14nmLPP even with increased clocks. You need to pay more attention to manuf. processes and how they affect clocks, efficiency, yields, etc. Also, hype has nothing to do with this, because leaked benchmarks of Vega 20 on 7nm suggests 3% higher performance on 7nm while running on much lower clocks in comparison to Vega 64.

a) yes indeed when undervolted vega is not tragic power efficiency wise b) cuda cores and gcn cores are not comparable 1 to 1 in any case c)traditionally in raw compute performance amd seems to be always packing more compute perfomance ...quite few years now but that never gave advantage to amd since the cards on paper where going toe to toe with much lower "in raw compute perfomance with nvidia" d) you do not have 7nm chips in your hand ... i don't none here does ...we do not know in the end how they will perform and how much better .... stop presenting speculations as something that's guaranty e)the 1000 mhz benchmark that came out and put it close to the vega that is on the market now ... says nothing because when a card is not supported on the benchmark the frequency read is wrong .... so it can be running at 500 ...1000 or 5000 mhz we do not know ... f) you said ( Considering that 7nm TSMC process allows 35-40% increase in performance over 16nm TSMC ) over what ? performance per watt ? still no products out .... but mind you performance /watt ratio is not raw performance increase . and to finish ... i really wish every single thing you said comes true ,hell i hope too i look forward to see the gpu arena on fire once again ! but you have to keep in mind over hyping hurts amd ... most of what you said are speculation and linger on the best case scenario up to wishful thinking