Thursday, July 6th 2017
AMD RX Vega Reportedly Beats GTX 1080; 5% Performance Improvement per Month
New benchmarks of an RX Vega engineering sample video card have surfaced. There have been quite a few benchmarks for this card already, which manifests with the 687F:C1 identifier. The new, GTX 1080 beating benchmark (Gaming X version, so a factory overclocked one) comes courtesy of 3D Mark 11, with the 687F:C1 RX Vega delivering 31,873 points in its latest appearance (versus 27,890 in its first). Since the clock speed of the 687F:C1 RX Vega has remained the same throughout this benchmark history, I think it's fair to say these improvements have come out purely at the behest of driver and/or firmware level performance improvements.
The folks at Videocardz have put together an interesting chart detailing the 687F:C1 RX Vega's score history since benchmarks of it first started appearing, around three months ago. This chart shows an impressive performance improvement over time, with AMD's high-performance GPU contender showing an improvement of roughly 15% since it was first benchmarked. That averages out at around a 5% improvement per month, which bodes well for the graphics card... At least in the long term. We have to keep in mind that this video card brings with it some pretty extensive differences from existing GPU architectures in the market, with the implementation of HBC (High Bandwidth Cache) and HBCC (High Bandwidth Cache Controller). These architectural differences naturally require large amounts of additional driver work to enable them to function to their full potential - full potential that we aren't guaranteed RX Vega GPUs will be able to deliver come launch time.
Sources:
Videocardz, 3D Mark's latest 687F:C1, 3D Mark's first 687F:C1
141 Comments on AMD RX Vega Reportedly Beats GTX 1080; 5% Performance Improvement per Month
I blame lack of financial resources on R&D and Raja (just because he was hyped for no good reason).
Hopefully Ryzen will be successful enough, as GPU buyers seem to have less clue about the fruck they are buying.Given its TDP, it is already pushing limits at 1630Mhz. (and what a laughable bump over FE)Oh, please...What about 1.5 times difference in die size? (not that core frequency difference had any meaning across 2 different architectures even if chip size was the same)Not having to pay hefty premium (up to $200) for adaptive sync.Left hand not knowing what right hand is... erm... doing.
Or right hand being too optimistic.
I still remember how Raja said that there are only 2 GPU companies "because it is so hard to write drivers". jeez.
I dint say vega is faster. But considering this one game for the frontier edition not gaming maybe it is capable of performance closer to 1080TI. is that instead a lucky shot for vega? Not in this industry though. That's unlikely. Just looking at the results of the benchmark trying to get something out of it. Don't wanna judge to fast either.
When one Vega is not enough, we have two. When two Vega is not enough, we built 4
And we crossfire 4 of these babes together.
Unlimited POWEEEEEER
2 might be a more reasonable starting point though. :)
I'd like to see a motherboard with an IF direct port where you can clip a vega or two to the MB and clamp a big liquid block on the top like those monoblock CPU coolers. :)
There's another photo-chop project for someone.
Or is that real and it's an MI100? ;)
Interesting to note is that quad Fury X beat quad Titan Maxwell even at 1080p (and more so at higher res)
www.techpowerup.com/forums/threads/an-epic-fury-x-review-quad-fury-x-vs-quad-titan-x.214231/
There were actually "leet" desktops sold out with quad Titan, so it's kinda not completely useless info.
What are you smoking, I want some. You're just creating your own new big disappointment to rage against, exactly like pretty much every video card release ever. Relax and call Vega what it is, a dud. Maybe you'll be pleasantly surprised sometime then instead?
30-40% from drivers isn't going to happen ever, even if AMD cards age better than anything else on the planet. Neither in performance nor power efficiency.
Also explain to me the logic of pushing out gaming benchmarks on non-gaming designed cards, without explaining what market these cards really are for - and above all, explaining how you have found that market that nobody ever knew really existed, you know that grey area between Pro-Semi-Pro-Casual-Pro or something, and still some light gaming - that is at 4K, I guess.. or not. Like... what? There is this thing called product positioning, and then there is Vega FE that really didn't get that at all.
Pile all this AMD PR together and the only image you can distill from it is total panic because this GPU just won't deliver, just like Fury X was never cost effective nor a good seller, it really isn't ready because it should never come to market because the end result is that it only costs AMD money. They are trying to elevate a Fury X rebrand into something more than it really is. Face it.
If it isn't drivers, its the future drivers, or if it isn't the API, its the future API that will fix all of AMD's problems. And if thát isn't true, then surely the totally revamped architecture is the culprit and we'll see it come to fruition in the near future. Meanwhile, we're revolutionizing gaming. Oh yeah, but that benchmark you saw, that wasn't on the final drivers, so don't worry.
Sorry I can't take this shit seriously anymore man
/annoyed
This isn't Fury X release anymore, that was just a big Tonga to be honest. Vega has thoroughly redesigned pretty much everything, so expecting AMD to pull perfect drivers out of the bat is just crazy. They've always have shortages on driver fronts and now everyone is behaving like this is some new fucking shocking thing.
I'm not cluelessly defending AMD just because I'd be a fanboy even though some would like that, I'm defending it because everyone is pissing at them and not being realistic that drivers might actually be holding it back. We're not talking unoptimized drivers but just straight not fully functional. That's a bit of a difference when it's a thing between few percent gains and actually working features that could give it larger gains than just few percent.
TL;DR: I'm not holding my breath but I'm still hopeful that Frontier Edition doesn't really reflect the final product.