AMD VEGA confirmed for 2017 H1

[humbug]

Well it is the only thing that you can use to guess, there is no information or benchmarks of Vega.. But judging from the performance / watt etc. of polaris, which is the only info available, it is not looking great.

You're comparing an Orange to an Apple that doesn't even exist yet.

 

[JediFragger]

You're comparing an Orange to an Apple that doesn't even exist yet.

But I think everyone can agree that Polaris is pretty damn far from impressive (in regards to perf/watt). Let's hope Vega can knock it out of the park, they need a HELL of an improvement for it too!!

 

[Final8y]

But I think everyone can agree that Polaris is pretty damn far from impressive (in regards to perf/watt). Let's hope Vega can knock it out of the park, they need a HELL of an improvement for it too!!

In comparison to NV no, in comparison to AMD previous cards then yes.

All i care about is price to performance.
60fps at 500w is no more enjoyable at 60fps at 100w, that game plays the same.

NV and AMD use different arcs thus different power requirements, rarely will they ever be equal.

 

[LoadsaMoney]

Im not waiting for Vega, only 1x choice now, and thats Nvidia, AMDs done.

 

[NirK]

You're comparing an Orange to an Apple that doesn't even exist yet.

Could be the same as a fury X but with another 25% MHZ, similar to RX480 vs 390x.

 

[JediFragger]

Im not waiting for Vega, only 1x choice now, and thats Nvidia, AMDs done.

Waiting for Volta or selling both Fury's for a 1080 man?

 

[Kaapstad]

It would be unwise to write off AMD just yet.

NVidia are making plenty of marketing mistakes that will eventually catch up with them.

 

[old gamer]

Yep I agree with kaap, too early to write amd off, I laughed at a vid I saw quite a while ago about amd's master plan but with there console deals, dx12 and the 480 doing quite well I'm beginning to think amd actually know what there doing and will slowly come together for them. Time will tell

 

[maonayze]

Im not waiting for Vega, only 1x choice now, and thats Nvidia, AMDs done.

Then what's the hold up....go for it. Get rid of those Fury Tri-X'es and pig out on a Titan XP.....fantastic efficiency and power in one not so cheap package. The grass has got to be greener over there.

 

[humbug]

Then what's the hold up....go for it. Get rid of those Fury Tri-X'es and pig out on a Titan XP.....fantastic efficiency and power in one not so cheap package. The grass has got to be greener over there.

Its very very green over here..... too green.

 

[D.P.]

But makes all the difference when it comes to a point scoring debate, which is pretty much the only place where it "matters". Some people on here would be on the whinge if an amd card used 5 more watts than an nvidia equivalent and spin it as a massive negative, yet if the same were true for team green that would be quickly brushed under the carpet. Gets old, fast.

What ridiculous garbage is this? Nonsensical junk like this gets old fast.

 

[Doom112]

The biggest problem AMD will face is efficiency issue. Even if it is 10% faster then GTX 1080 it wont matter after a year and it will be more disappointing that if it takes 50 to 60% more power to beat GTX 1080 ,which is more likely to be the case. AMD will not get enough profit from desktops only ,therefore, they require to win OEM contracts and laptops, which can only be done by beating nvidia on efficiency + performance and it is likely that wont happen.

 

[Gerard]

What ridiculous garbage is this? Nonsensical junk like this gets old fast.

Much like most of your thinly veiled biased waffle

 

[D.P.]

Much like most of your thinly veiled biased waffle

So on a discussion of architectural efficiency discussing the measurable output of architectural efficiency, i.e. performance per watt, gets old? Really, perhaps you are int eh wrong forum.

 

[humbug]

So on a discussion of architectural efficiency discussing the measurable output of architectural efficiency, i.e. performance per watt, gets old? Really, perhaps you are int eh wrong forum.

Its flawed before it even starts, when your comparing Polaris to an architecture unknown and none existent at this point.

It is hot air waffle.

 

[Kaapstad]

Its flawed before it even starts, when your comparing Polaris to an architecture unknown and none existent at this point.

It is hot air waffle.

It looks like Polaris and Vega are going to be about a year apart, this in it's self is a huge clue that they are going to be very different from each other.

 

[humbug]

It looks like Polaris and Vega are going to be about a year apart, this in it's self is a huge clue that they are going to be very different from each other.

The clue is in their own argument, Polaris doesn't scale, its a die shrunk Tonga, never designed for high clock rates, the 1400Mhz is about as far as it will go and uses a bucket load of watts to maintain it because its way outside its comfort zone, clock it where the architecture is efficient and it will use about as much power as it will when it lands in the XBox Scorpio and PlayStation Neo.

Its a GPU designed for the new consoles and released to market for some income.

AMD have not yet finished (The real thing) thats Vega.

 

[TheRealDeal]

The clue is in their own argument, Polaris doesn't scale, its a die shrunk Tonga, never designed for high clock rates, the 1400Mhz is about as far as it will go and uses a bucket load of watts to maintain it because its way outside its comfort zone, clock it where the architecture is efficient and it will use about as much as it will when it lands in the XBox Scorpio and PlayStation Neo.

Its a GPU designed for the new consoles and released to market for some income.

AMD have not yet finished (The real thing) thats Vega.

+1

Even with this in mind it's still good competition for the 1060 performance wise. I would take one over a 1060 due to dx12/Vulkan performance. In dx11 there is not much in it either. Vega's performance/watt should be way better but might not match Pascal in this department. I reckon in overall performance/£ it will win easily. I doubt it will be cheap but it won't be hard to beat Nvidia's pricing atm.

 

[humbug]

+1

Even with this in mind it's still good competition for the 1060 performance wise. I would take one over a 1060 due to dx12/Vulkan performance. In dx11 there is not much in it either. Vega's performance/watt should be way better but might not match Pascal in this department. I reckon in overall performance/£ it will win easily.

Its a great little card, and its still pretty efficient in its own right, its solid, and yes high feature level DX12 and Vulkan performance is excellent.

 

[maonayze]

The biggest problem AMD will face is efficiency issue. Even if it is 10% faster then GTX 1080 it wont matter after a year and it will be more disappointing that if it takes 50 to 60% more power to beat GTX 1080 ,which is more likely to be the case. AMD will not get enough profit from desktops only ,therefore, they require to win OEM contracts and laptops, which can only be done by beating nvidia on efficiency + performance and it is likely that wont happen.

Efficiency has always been AMDs biggest problem and still is. However, as far as their own products are concerned they are starting to do better. If Vega is 10% faster than a 1080 then that will mean that it is 15% slower than a Titan XP.....at the right price it will sell by the bucketload even if it takes more power to do so. Again depends on the price/performance and we will not know that for a while yet. I dont think it will be 50% more power to get that 10% performance over the 1080...maybe 20-30%.

Obviously with only a 30% market share AMD will not get enough profit on desktops alone, but they do also have the console contracts and depending on Vega's price/performance that marketshare may (probably will) increase again. AMD continue to increase efficiency in their own products even though it isnt as good as Nvidia's and that is still a good thing.