AMD RX Vega Leaked Benchmark Shows It Ahead Of GTX 1080 – Specs Confirmed, 1630MHz Clock Speed, 8GB HBM2 & 484GB/s Of Bandwidth

[Mr_Troll]

 

Quote

AMD Radeon Vega Specs Confirmed - 1630 MHz clockspeed, 8GB of HBM2 @ 945 Mhz, 484 GB/s memory bandwidth

 

The clock speed represents a modest 30Mhz improvement over the Vega Frontier Edition and the HBM2 frequency is identical to that of Vega FE which translates to 484GB/s of bandwidth. This, funnily enough, translates to 1GB/s of bandwidth per mm² .  In terms of memory capacity as expected this board, which we expect to be the Vega 10 flagship in the RX lineup, comes equipped with 8GB of HBM2, rather than the exuberant 16GB of the Vega FE.

 

To deliver 16GB of memory in two HBM2 stacks in the Vega Frontier Edition each stack has to be 8-Hi. 8-Hi HBM2 stacks are considerably more expensive and difficult to produce compared to standard 4-Hi HBM2 stacks which is why it makes all the sense in the world for AMD to opt for the 8GB option on its gaming Vega card, at least for now.

 

Now, on to the bit most of you have been waiting for, actual performance data for AMD’s upcoming gaming oriented RX Vega.  We’ll only be looking at the graphics score in this test, because that’s what’s actually representative of the GPU’s performance. The combined score on the other hand is influenced, as the name implies, by a combination of GPU & CPU performance.

RX Vega manages to score 31,873 graphics points, roughly four thousand points and some change ahead of the GTX 1080 and double that compared to the GTX 1070. However, the GTX 1080 ti maintains a commanding lead with over 38 thousand points, nearly seven thousand points ahead of Vega.

It’s important to point out that 3DMark 11 graphics scores scale exceptionally well with GPU core counts. This is why we see the GTX 1080 Ti maintaining a solid 39% lead overGTX 1080 here, where in more realistic real-world gaming scenarios it’s ahead by a slightly more modest 34% and 28% at 4k and 1440p respectively.

 

Putting the figures into perspective, the GTX 1080 Ti is ahead of RX Vega by roughly 20%. RX Vega is ahead of the GTX 1080 by 15% and leads the GTX 1070 by 35%. Keeping that in mind, we expect that similarly to the GTX 1080 Ti, RX Vega’s lead over the GTX 1080 and GTX 1070 could shrink in real-world gaming tests depending on the game and the resolution tested.

 

 

Source:http://wccftech.com/amd-rx-vega-benchmark-leaked-specs-confirmed-1630mhz-8gb-hbm2-484gbs-of-bandwidth/

 

If the 3d mark performance translates into games and if it gets the right price tag it might still sell well despite all the dissapointment surrounding the FE card.

[Jurrunio]

Can Vega beat the final boss: the 1080 ti though?

[iRileyx]

Uh, but the 1080 is running at 1607 mhz, and the 1080ti at 1481 mhz? who actually games with them at those speeds.

 

And i imagine Vega won't be able to overclock as much as those two can.

[TOMPPIX]

Let's hope this gpu is terrible for miners.

[Majestic]

3 minutes ago, RKRiley said:

Uh, but the 1080 is running at 1607 mhz, and the 1080ti at 1481 mhz? who actually games with them at those speeds.

 

And i imagine Vega won't be able to overclock as much as those two can.

Those are the standard baseclocks dude. Nothing funky.

[rn8686]

Because WCCF is a reliable source of leaks... 

 

(Big Vega will probably be faster than a 1080 though, just not quite a 1080ti). 

[iRileyx]

Just now, Majestic said:

Those are the standard baseclocks dude. Nothing funky.

I know, but GPU Boost makes them run quite a bit over the base clock anyway, so comparing the base clock isn't really a fair comparison.

[Majestic]

Just now, RKRiley said:

I know, but GPU Boost makes them run over the base clock anyway, so comparing the base clock isn't really a fair comparison.

Who says they weren't boosting during the test?

[Simon771]

4 minutes ago, TOMPPIX said:

Let's hope this gpu is terrible for miners.

It will be since it's using HBM.

[genexis_x]

Let's hope Vega cards comes with an interesting pricing

And...please fix formatting

[Droidbot]

AMD finally got their driver-related shit together, it seems. 

 

Adored's video on the shit that AMD is going through is really quite good: 

 

[Taf the Ghost]

6 minutes ago, Droidbot said:

AMD finally got their driver-related shit together, it seems. 

 

Adored's video on the shit that AMD is going through is really quite good: 

 

It's a good video from Adored, as he's been pretty down on Vega for a while.  Or, more accurately, accurate to where it should line up.

 

The main thing with the Vega FE is that is really seems like they rolled out an utterly Stable driver without any post-Fiji optimizations. Thus, everything runs, but the Gaming Driver for RX Vega isn't in there. So, the Vega FE is the worst-possible performance for the top-tier RX Vega, though it should come in higher with everything turned on. We'll just have to wait and see.

 

Expect it to slot in right between the 1080 and 1080 Ti, if it does they'll put it at 599. If it's closer to the 1080, probably 499. At least for the top SKU. Or maybe they want to screw over Nvidia a bit and we get a 399 monster.

[tom_w141]

Yay on the day my 1080Ti is coming. Didn't think it would be much of a threat but nice to see it won't devalue my purchase  

 

http://www.3dmark.com/3dm11/12139704

[tom_w141]

25 minutes ago, valdyrgramr said:

OCs like a god

As much as I like AMD. Their stuff doesn't OC as well as the competition. With AMD its strong architecture or high clocks, pick one.

[Shreyas1]

The problem here is that even if amd can match the 1080ti, all nvidia has to do is provide a significant increase from pascal.

[tom_w141]

1 minute ago, valdyrgramr said:

OC higher in the CPU department

As much as I love my 1700. That isn't true. 4GHz hard ceiling for me.

[tom_w141]

1 minute ago, valdyrgramr said:

but the FX lineup still holds the world records.  Hence, the lack of performance.

 

7 minutes ago, tom_w141 said:

With AMD its strong architecture or high clocks, pick one.

 

[Master Disaster]

I just don't believe these numbers.

 

Unless someone can explain how halfing the RAM and boosting the core clock by 30Mhz can push performance so far ahead of Vega FE I'm sorry but I just don't see these numbers as being real.

 

Linus said on last weeks wan show, Vega FE & Vega RX are the exact same die, sure there might be a few small changes but essentially Vega is Vega and basic physics tells us you can't magically make something out of nothing which leaves 3 scenarios

 

1) AMD is infact Hogwarts

2) They gimped Vega FE

3) These numbers are not real

 

Which one do you think is more likely?

[MeatFeastMan]

I can smell the 400w tdp coming from a mile off...

[TheRandomness]

5 minutes ago, Master Disaster said:

I just don't believe these numbers.

 

Unless someone can explain how halfing the RAM and boosting the core clock by 30Mhz can push performance so far ahead of Vega FE I'm sorry but I just don't see these numbers as being real.

 

Linus said on last weeks wan show, Vega FE & Vega RX are the exact same die, sure there might be a few small changes but essentially Vega is Vega and basic physics tells us you can't magically make something out of nothing which leaves 3 scenarios

 

1) AMD is infact Hogwarts

2) They gimped Vega FE

3) These numbers are not real

 

Which one do you think is more likely?

FE is designed to work with two sets of drivers (well, one or the other at any given time, not both), so swapping between them would probably cause some issues.

Let's hope this gpu is terrible for miners.

According to some reviewer somewhere, Vega kinda sucks for mining because as already said, mining hates HBM(2). @Ryan_Vickers is getting about 32 MH/s on his Fury, whilst a memory-overclocked 390X can match and surpass that.

[Master Disaster]

1 minute ago, TheRandomness said:

FE is designed to work with two sets of drivers (well, one or the other at any given time, not both), so swapping between them would probably cause some issues.

That still doesn't explain how FE can barely match a 1070 yet RX lands somewhere between a 1080 & 1080ti? That's way to much of a discrepancy to be software.

[TheRandomness]

Just now, Master Disaster said:

That still doesn't explain how FE can barely match a 1070 yet RX lands somewhere between a 1080 & 1080ti? That's way to much of a discrepancy to be software.

Remember, it's AMD's day-one drivers we're talking about here The same trend that lead the 300 series to surpass the 900 series after a few months, just from drive updates.

[cj09beira]

9 minutes ago, Master Disaster said:

I just don't believe these numbers.

 

Unless someone can explain how halfing the RAM and boosting the core clock by 30Mhz can push performance so far ahead of Vega FE I'm sorry but I just don't see these numbers as being real.

 

Linus said on last weeks wan show, Vega FE & Vega RX are the exact same die, sure there might be a few small changes but essentially Vega is Vega and basic physics tells us you can't magically make something out of nothing which leaves 3 scenarios

 

1) AMD is infact Hogwarts

2) They gimped Vega FE

3) These numbers are not real

 

Which one do you think is more likely?

Drivers can make huge difference in how the card performs more so on dx11

[Master Disaster]

Just now, cj09beira said:

Drivers can make huge difference in how the card performs more so on dx11

But from sub 1070 to 1080+ performance just from drivers? Also if it is driver optimisation then its safe to assume the FE will get the same boost, right?

[Taf the Ghost]

1 minute ago, Master Disaster said:

That still doesn't explain how FE can barely match a 1070 yet RX lands somewhere between a 1080 & 1080ti? That's way to much of a discrepancy to be software.

Current Vega FE drivers are Fiji derivatives. They seem extremely stable. We have no publicly available copies of the RX Vega gaming drivers. The Vega FE set appears to have none of the Polaris or Vega optimizations active.  Raja has also mentioned them building an entire new driver architecture.

 

But the biggest thing is the professional testing. In those, the Vega FE comes in just behind the GP102 cards, so the Quadro form of the 1080 Ti. That wouldn't happen if the performance just wasn't there. That's why the gaming performance is odd.  My running theory is not that the Vega FE cards are "gimped", but they are fully stable drivers.  Which means RX Vega will have fully stable drivers for all games.  Then the optimizations can be rolled out with the RX launch and if a game has issues, there's a fall-back driver to work from.

 

With the information currently available, that would logically fit together.