New Performance Benchmarks of AMD's Vega Frontier Edition Surface

[Raevenlord]

You probably took a long, hard read at our article covering a single-minded user's experience of his new Vega Frontier Edition. Now, courtesy of PCPer, and charitable soul Ekin at Linus Tech Tips, we have some more performance benchmarks of AMD's latest (non gaming specific) graphics card.

Starting with 2560x1440, let's begin with the good news: in what seems to be the best performance scenario we've seen until now, the Vega Frontier Edition stands extremely close to NVIDIA's GTX 1080 Ti video card in Fallout 4. It trails it for about 10 FPS most of the test, and even surpasses it at some points. These numbers should be taken with a grain of salt regarding the RX Vega consumer cards: performance on those models will probably be higher than the Frontier Edition's results. And for the sake of AMD, they better be, because in all other tests, the Frontier Edition somewhat disappoints. It's beaten by NVIDIA's GTX 1070 in Grand Theft Auto V, mirrors its performance in The Witcher 3, and delivers slightly higher performance than the GTX 1070 on Hitman and Dirt Rally (albeit lower than the GTX 1080.)







At 4K (3840x2160), the Vega FE trails the GTX 1080 by about 3 FPS (at 57 FPS, just shy of 60 FPS) on Dirt Rally; trails it again (this time with a 7 FPS difference) in Fallout 4, at around 42 FPS; delivers around 66% of the GTX 1080's performance on Grand Theft Auto V, and less than 50% of the GTX 1080 Ti's performance on the same game. In Hitman, the Vega FE delivers around 83% of the GTX 1080's performance (around 50 FPS versus the 1080's 60), and delivers almost the same result on The Witcher 3, barely maintaining a 30 FPS performance towards the end of the run.



Do note that all of these tests will apparently be re-run by PCPer, and the publication is looking to publish their results later today. Also keep in mind the Vega Frontier Edition isn't a consumer graphics card, and isn't officially meant for gaming. Instead, it's meant for professionals or prosumers who do some professional workloads as well as some gaming, and want to have the ability to test their development fruits with the same graphics card they developed with. Power draw was rated at around 280 W while gaming, with only 25 of those being taken from the PCIe slot, which seems somewhat disproportionate.

Apparently, there was some testing done on mining software as well, and performance is reported as disappointing (as in, "very low".) This probably speaks to the HBC (High Bandwidth Cache) and HBCC (high Bandwidth Cache Controller), which probably will require a lot of fine tuning from mining software (remember the GTX 1080 is generally poor in mining workloads compared to the GTX 1070 because of the higher latency of its GDDR5X memory implementation.) Perhaps these news come as a relief, however, since availability of RX Vega cards will likely be limited without miners taking up the whole supply.

View at TechPowerUp Main Site

 

[GhostRyder]

Performance seems all over the place, this is probably a sign that these frontier editions either don't have the drivers ready for games or maybe its the cards themselves. To me its a bit odd that Fallout 4 would do that because that does not seem to be the case in the past in this comparison unless there is something specific about the Vega FE the game really likes.

Who knows, unfortunately we really are not getting much of a taste of this card and that to me worries on so many levels.

 

[ratirt]

I wonder how much better RX gaming Vega will be. I know we have to keep in mind that this card is not for gaming and new drivers architecture. Anyway it keeps up in at least one title around 1080 TI which gives some clue although in all other titles if falls behind by a noticeable margin from 1080 TI. Anyway this gives some indications as if the card was able to keep up with the 1080TI in one game who knows if this wont be the case if the RX gaming Vega shows up with proper drivers. Actually i do hope so since i'm waiting to buy one Xross my fingers for Gamin Vega high performance

 

[phanbuey]

Mining software will most likely need to be optimized to take advantage of the architecture...

 

[ratirt]

Mining software will most likely need to be optimized to take advantage of the architecture...

Well i think that's the case not only for mining but sure it does need optimization.

 

[notb]

Well... I've written on this forum before that new generations of gaming GPUs could be limited in mining and virtually no one believed me.

I expect the same from the accelerated Pascal Refresh.

 

[phanbuey]

Well... I've written on this forum before that new generations of gaming GPUs could be limited in mining and virtually no one believed me.

I expect the same from the accelerated Pascal Refresh.

Why do you think that could be?

 

[_Flare]

It has been told that the Card actually runs on a modified GCN4 or even Fiji Driver.
So there may be no GCN5 specific new Features enabled because the Driver lacks support.
Also there could be ECC used slowing Games down by upto 20% alone, in some cases.


With the first real GCN5 Driver the clould be a 20% uplift easy.

 

[Divide Overflow]

Where are the workstation benchmarks?

 

[os2wiz]

I find it amusing how ignorant people jump on the anti-RX Vega bandwagon the same way they did with Ryzen. Yes the silicon may be the same but the video bios may be considerably different. Since F.E. is not targeted as a gaming card not only are the drivers not optimized for it but the bios may shut off a significant number of registers that are not directly involved withe A.I. the actual target audience for this card. This would be to save power etc. The gaming mode on the card may not entirely resolve this issue. Then the clock speed on F.E. may be 15% lower than RX Vega. Next the state of drivers for F.E. may also incur a 15% penalty over drivers upon RX Vega release.. All told RX Vega is likely to perform 30% to 40% better than Frontier Edition depending on how many factors I mentioned are in play.

 

[cdawall]

I find it amusing how ignorant people jump on the anti-RX Vega bandwagon the same way they did with Ryzen. Yes the silicon may be the same but the video bios may be considerably different. Since F.E. is not targeted as a gaming card not only are the drivers not optimized for it but the bios may shut off a significant number of registers that are not directly involved withe A.I. the actual target audience for this card. This would be to save power etc. The gaming mode on the card may not entirely resolve this issue. Then the clock speed on F.E. may be 15% lower than RX Vega. Next the state of drivers for F.E. may also incur a 15% penalty over drivers upon RX Vega release.. All told RX Vega is likely to perform 30% to 40% better than Frontier Edition depending on how many factors I mentioned are in play.

Or amd could sell them with disabled cores since they are a lower market value and they actual perform worse. Another brand does that with prosumer vs consumer cards.

 

[xkm1948]

I find it amusing people still trying to defend a failed product launch and a bad flagship produxt.

 

[rainzor]

It has been told that the Card actually runs on a modified GCN4 or even Fiji Driver.
So there may be no GCN5 specific new Features enabled because the Driver lacks support.
Also there could be ECC used slowing Games down by upto 20% alone, in some cases.


With the first real GCN5 Driver the clould be a 20% uplift easy.

Where did you read that? RTG dude at B3D mentioned it uses "older" driver. He didn't say anything about the features missing, he even dismissed the insinuation that the FE driver may be gimped in some way compared to RX driver. Next driver revision should be pretty telling i guess..

 

[cdawall]

Where did you read that? RTG dude at B3D mentioned it uses "older" driver. He didn't say anything about the features missing, he even dismissed the insinuation that the FE driver may be gimped in some way compared to RX driver. Next driver revision should be pretty telling i guess..

To be fair an amd product with a garbage performing driver is hardly news or unbelievable.

 

[BiggieShady]

I'm wondering why is Fallout 4 an exception in these games ... I mean it's a game with DX11 shader support tacked onto aging DX9 era engine ... weird because this arch should be very DX12 friendly, and it doesn't shine in any of the DX12 titles.
What makes Vega almost twice as fast as Fury, but only in Fallout 4?

 

[cdawall]

I'm wondering why is Fallout 4 an exception in these games ... I mean it's a game with DX11 shader support tacked onto aging DX9 era engine ... weird because this arch should be very DX12 friendly, and it doesn't shine in any of the DX12 titles.
What makes Vega almost twice as fast as Fury, but only in Fallout 4?

Maybe it can take advantage of the HBC?

 

[ratirt]

I'm wondering why is Fallout 4 an exception in these games ... I mean it's a game with DX11 shader support tacked onto aging DX9 era engine ... weird because this arch should be very DX12 friendly, and it doesn't shine in any of the DX12 titles.
What makes Vega almost twice as fast as Fury, but only in Fallout 4?

I think it is the game driver. it is new here and this vega is not meant for gaming. So i guess this is why. Maybe the Fallout showed the potential of the card?

Maybe it can take advantage of the HBC?

I don't think Fallout 4 is that much memory sensitive.

 

[RejZoR]

I find it amusing people still trying to defend a failed product launch and a bad flagship produxt.

Launch was crappy, but lets be honest, this is a production card, not a gaming one. It's not meant to be released with all the colorful and noisy PR that we expect with gamer cards. Anyone remembers a lot of fanfare around every Quadro card release? I sure as hell don't. When was the last time anyone ran Fallout 4, Dirt, Heaven and 3DMark on them (be it FirePro or Quadro)? I honestly can't remember a single time. It doesn't matter what they say, basing whole RX Vega state on FE card is like evaluating gaming performance of GeForce cards using Quadro. Yes, they can run games, but they usually do it like shit as well.

So, people, don't be stupid. With all this excitement of finally having a first actual Vega core, people forgot they just threw all the logic out the window by testing it as a gamer card. If gaming RX Vega will be rubbish, piss on it by all means, but doing that now just makes everyone look like absolute cretins. Yes, I watched PCPer's live stream, but I still believe the same to be true.

 

[Hood]

I find it amusing how ignorant people jump on the anti-RX Vega bandwagon the same way they did with Ryzen

Some people try to improve their status on forums by spewing negative comments, as soon as any tiny problem arises with a new or upcoming product, acting as though they have "inside information", and making stuff up. Some people are afraid of being seen as a newbie or fanboy by saying they like something, it's easier to put down everything so they seem edgy and cool (at least in their own mind). Nothing new, I've noticed this tendency in bullies and the small-minded since I was in first grade (50 years ago), and now these same losers all have computers and can spew their hate to a much larger group of people. This negativity is a reliable indicator that they can be safely ignored Or, you can read their comments and take away a reinforced positive impression of the product (if this nimrod feels threatened by it, it must be a good product).

 

[the54thvoid]

All this, pro-sumer cards can't game nonsense. While they're orientated at pro markets, it does not somehow mean all that 'identical' hardware can't work with gaming.
I'm on mobile so posting links is too difficult for my 'on holiday's brain. But a quick Google search shows the P6000 Quadro (hothardware I think) beat the Titan X (original Pascal model) in gaming, convincingly.
Yes, far more expensive than Vega FE but the point is, it's a Quadro that games better than a Titan X.
Vega FE doesn't have an excuse unless we beat that worn track, AMD still can't write proficient drivers. And before I'm lambasted for criticising AMD, explain the Quadro gaming performance in the context of Vega FE gaming performance.

 

[ratirt]

Did you guys notice the 2k fallout 4 is at 120FPS while the 4k is 40+ FPS. That kind of a drop is just not possible. Although it's not a gaming card but such a difference?

 

[RejZoR]

All this, pro-sumer cards can't game nonsense. While they're orientated at pro markets, it does not somehow mean all that 'identical' hardware can't work with gaming.
I'm on mobile so posting links is too difficult for my 'on holiday's brain. But a quick Google search shows the P6000 Quadro (hothardware I think) beat the Titan X (original Pascal model) in gaming, convincingly.
Yes, far more expensive than Vega FE but the point is, it's a Quadro that games better than a Titan X.
Vega FE doesn't have an excuse unless we beat that worn track, AMD still can't write proficient drivers. And before I'm lambasted for criticising AMD, explain the Quadro gaming performance in the context of Vega FE gaming performance.

Just because Quadro works well in 1 game and one benchmark, it doesn't mean it's a great gaming card. There are thousands of games out there and they don't make game profiles for cards like this. They do for GeForce. I mean, why do you think Quadro drivers are basically half the size of GeForce drivers (P6000) if the GPU is essentially the same?

 

[dj-electric]

Performance is poor but this is not a gaming card.
Waiting to see the actual gaming card, with its drivers for gaming.

As much as i would like to bash stuff for horrible performance, Gaming VEGA deserve a fair chance just like any other cards.

 

[ratirt]

Performance is poor but this is not a gaming card.
Waiting to see the actual gaming card, with its drivers for gaming.

As much as i would like to bash stuff for horrible performance, Gaming VEGA deserve a fair chance just like any other cards.

I hope it does and i do believe that waiting for the Vega gaming for so long. I'm even one click from buying a 4k display for it tempting Already got my pick just waiting for Vega.

 

[RejZoR]

I'd rather see how Vega FE performs in professional applications like Maya, Blender and the likes... After all, even AMD themselves showcased it doing exactly that during presentation. They ran Prey basically for the lolz because everyone (gamers) expected it. And they probably only optimized drivers for that presentation for that one game. Which, surprise suprise, wasn't tested by PCPer so we can't really know.