• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

AMD Polaris architecture – GCN 4.0

Its still a rumor based on nothing. Don't state it as fact.

Point taken, but I personally put a fair bit of credibility in AMD exclusive AIB partners not being too happy right now, even if Polaris is perfectly on track. The last few years has seen a lot of rebrands, cards with a high power draw leading to bigger more expensive PCBs and cooling solution, the Fury line up delayed, poor supply, and then simply not that enticing compared to the 980ti, and Nvidia outselling AMD 4 : 1. And then I think the AMD products often have a lower profit margin

I'm really not trying to knock AMD here at all, but would you rather be nvidia exclusive or AMD exclusive in the last few years? Imagine how much money Nvidia partners made selling 970s, at one stage they accounted for 5% of the installed user base! I would also love to know how many more 980Ti's were sold compared to Fury/nano, and the profit margin on the 980Tis.

The Polaris cards should be fantastic in their market position, but not having anything to combat the 1080 performance still gives the AMD AIB's a marketing and age problem.
 
Oh, so now it's fact, from a clickbait article that most people have dismissed? You were calling it a vague rumour a few days ago and now you are stating it as fact. LOL R7Slayer was right.

Where did I say that the rumour about AMD AIB partners being frustrated was vague?

I said the rumour that the AIB partners wont have anything to show is vague=.https://forums.overclockers.co.uk/showpost.php?p=29484872&postcount=117


These are 2 entirely different rumors and I have 2 entirely different opinions on both rumours.

1) Could AMD exclusive AIB parents be frustrated, yes. I think this is quite credible, but I'm not assigning definite belief, just lots of logical reasons why they might be.
2) Will AMd have nothing to show at computex? A very ague rumour, not very credible, much less likely to be true. But I have no idea what AMD or their partners will be showing.
 
Last edited:
Seems the Validaiton rumour came from Nordic Hardware.

From tweaktown.

Nordic Hardware reached out to us with their report, saying that AMD's partners "won't have any new cards to display at Computex and the only Polaris cards promoted to them from AMD are R9 390/390X performance class but for a mid-range price. Great value but no sign of any GTX 1080 contender".

Which is right, as its been known for a while now, that AMD aren't going to compete with Nvidia, and are only releasing cards for the minimum spec of VR (the 970/290 series), but they'll be better, as they'll be at 390X performance, and priced well, around £200, as thats the market now, and where they'll get marketshare back from, as no ones going to buy a 1070/80, as they are going to be too expensive.
 
Last edited:
lies! lol

Believe what you want. I don't like AMD's marketing approach and I don't like their lack of linux or OpenGl support but my opinion on both AMD and Nvidia are pretty much the same - they are business out to get money. What I don't like is people mindless bashing Nvidia when in my experience Nvidia has always offered everything i needed, solid linux, good OpenGl drivers, i can contact the driver team, excellent CUDA documentation and support and flawless gaming experience.
 
The Polaris cards should be fantastic in their market position, but not having anything to combat the 1080 performance still gives the AMD AIB's a marketing and age problem.

Lets not get ahead of ourselves here, Polaris is a brand new architecture with a lot of significant performance improvements, its also a lot larger than the 390X when scaled up.

On its size alone it will blow the 390X into 2010, the architectural improvements can only make it even faster.

The 1080 is not that impressive, for all we know, P10 could match or beat it easily.
 
I think if AMD had something like that humbug, they'd be shouting it from the rooftops and sticking slides on every hardware site known to mankind.
 
I think if AMD had something like that humbug, they'd be shouting it from the rooftops and sticking slides on every hardware site known to mankind.

And they did, we know what size it is, we know what production node its built on, they even told us what architectural improvements they made to it. < thats all you need and pretty much all Nvidia have told us to date. other than misleading marketing slides on performance.

The problem is no one is listing to them, all people are interested in is utterly baseless click bait.
 
And they did, we know what size it is, we know what production node its built on, they even told us what architectural improvements they made to it. < that's all you need and pretty much all Nvidia have told us to date. other than misleading marketing slides on performance.

The problem is no one is listing to them, all people are interested in is utterly baseless click bait.

Pretty much. We know as much about the 1080 as we do about Polaris. Although we know a little more about the 1080 if you take the DOOM Vulkan demonstration into account. But we have no performance metrics of current cards to compare to till after a few days after Doom drops and tey release the vulkan patch.

The way i see things is that AMD may release three versions of P10, two on smaller PCB's, one being a full die and one cut down. And then possibly a larger pcb version using a full die with a beefier power supply if needed so it can happily run at higher clocks compared to the smaller versions with a larger cooler.
 
Maybe really what's going on here is AMD are waiting for the 17th to get independent review data so they can compare the Polaris cards directly against the 1070/1080 in their launch event :D

Also lets not forget that after all the changes both sides have made, it looks likely they'll end up with the pretty much the same IPC per core (AMD may be mildly ahead if their GCN4 improvements were decent), and since they have the same amount of cores it should be interesting. Though obviously the clockspeeds Nvidia have shown off are impressive, I'd be surprised if 2GHz Polaris is a thing.
 
Its not going to be 390 performance it will be 20/30% more than that. It won't be as fast as a 1080 but probably match the 1070 give or take.
 
Lets not get ahead of ourselves here, Polaris is a brand new architecture with a lot of significant performance improvements, its also a lot larger than the 390X when scaled up.

On its size alone it will blow the 390X into 2010, the architectural improvements can only make it even faster.

The 1080 is not that impressive, for all we know, P10 could match or beat it easily.

Unless a smokescreen, they adress different markets.
if the 1600mhz seems doable it be a great card.
 
Lets not get ahead of ourselves here, Polaris is a brand new architecture with a lot of significant performance improvements, its also a lot larger than the 390X when scaled up.

On its size alone it will blow the 390X into 2010, the architectural improvements can only make it even faster.

The 1080 is not that impressive, for all we know, P10 could match or beat it easily.
It could also easily not be half as impressive as you keep trying to make it sound.

Dont know why you do this. Seems like you're setting yourself up to be disappointed.
 
It could also easily not be half as impressive as you keep trying to make it sound.

Dont know why you do this. Seems like you're setting yourself up to be disappointed.

I can see where the cautious optimism comes from, and I share it.

Thing is, in theory AMD could just stick the 390X design on 14nm and do a 3072 core version running 1300+ MHz in the same size as Polaris 10. Something like that, requiring no arch improvements would match/beat a stock 980 Ti.

So including improvements, I expect the Polaris 10XT should be able to match the 1070.


EDIT: Also who's already forgetting all the hurr durr there's no way a ~320mm2 GTX 1080 could beat the 980Ti, it's just too powerful, etc.?
 
Last edited:
It could also easily not be half as impressive as you keep trying to make it sound.

Dont know why you do this. Seems like you're setting yourself up to be disappointed.

I'm already disappointed, not so long ago Pascal was 10x Maxwell, Jens 'the man himself' said so, then suddenly its only 2x Maxwell, and then we learn its that in a very very rough sense under special circumstance with a lot of things exaggerated.

After all that it now looks like what we actually have is Maxwell x1.25.... from 10x to that, what an anticlimax. so disappointed. eh?
 
I'm already disappointed, not so long ago Pascal was 10x Maxwell, Jens 'the man himself' said so, then suddenly its only 2x Maxwell, and then we learn its that in a very very rough sense under special circumstance with a lot of things exaggerated.

After all that it now looks like what we actually have is Maxwell x1.25.... from 10x to that, what an anticlimax. so disappointed. eh?

And there were people on some forums who actually believed it would be a 10x difference in performance.
 
I'm already disappointed, not so long ago Pascal was 10x Maxwell, Jens 'the man himself' said so, then suddenly its only 2x Maxwell, and then we learn its that in a very very rough sense under special circumstance with a lot of things exaggerated.

After all that it now looks like what we actually have is Maxwell x1.25.... from 10x to that, what an anticlimax. so disappointed. eh?

...and still people are hyped like hell. :rolleyes:
 
Its not going to be 390 performance it will be 20/30% more than that. It won't be as fast as a 1080 but probably match the 1070 give or take.

Would mean it would have to almost match the GTX 980 ti at stock clocks, which I somehow highly doubt it will be able to do. Don't think they want to maximize performance with the Polaris lineup and instead go for a more balanced approach, so I'm guessing 390x performance with much lower power consumption.
 
Back
Top Bottom