• 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.

Radeon RX 480 "Polaris" Launched at $199

They have released some info as is usual. Can you tell me how it performs? No I didn't think so. Can you say for definite what the core clock is and how the boost works? We know very little. So in essence we don't have all the paper facts and in fact we know less than we actually have confirmed.

I agree, not all specs but it was compared to the 1080 in the SLI test and the 970? in the VR test.

Anyway, it seems certain there's going to be cards for sale on launch day so lets leave it there.
 
We know the 1070 prices but the problem is we don't know the exact prices for the 480 (in Pounds Sterling)
We also dont know how the 480 will perform exactly, let alone that it's going to be a great card for 1440p(meaning 980Ti/1070 level at minimum).

Or that baby Vega will come in October.

That entire post sounded like fanboy wishful thinking to me, but I guess flopper is some known quantity around here who we just laugh off their antics? :/
 
Show me your evidence to back up your claims DM or you have nothing but waffle. Come on show us hard facts !!!

As to Flopper I think his posts are great and although sometimes a bit optimistic, they are very funny and good humoured.

It's funny that your claims never need evidence, but everyone else's you don't believe do.

Why don't you discuss things like an adult.

I've made a point, the gap between cores needs to be big enough to make sense and for Vega to have HBM2 it really needs more shaders to take advantage of that memory bandwidth. Both of these things would lead to a fairly logical conclusion that small vega is bigger than the 1080 and probably by a reasonable amount, it has to be big enough to create a significant gap to RX480, for both itself and salvaged parts to be made from the core. Now be an adult and come up with a reason why small Vega would be only 80mm^2 bigger than RX480 and thus only the same size as the 1080 die. Come up with a reason that attempting to match shader count and die size to HBM2's available bandwidth doesn't make sense. I can think of a few myself.

I reasonably discuss the reasoning behind rumours, I'm not at all always right but I put across an adult discussion that only descends into farce when people like you refuse to discuss rumours in a sensible manor. I think this will happen because X,Y and Z, I don't think that will happen because this that and the other, etc. Not just, it won't happen because I say so and you can't show me the internal design plans for an unreleased product therefore you are definitely wrong.

There ARE leaks that puts small Vega with both HBM2 and 4096 shaders, in fact that 490x rumoured spec has been around for a VERY long time.

But rather than respond reasonably and discuss the merits of my conclusions and either agree or disagree with your own well reasoned counter argument and can almost immediately tell what kind of reply I'll get, why not confound expectations and do something different.
 
Last edited:
It's funny that your claims never need evidence, but everyone else's you don't believe do.

Why don't you discuss things like an adult.

I've made a point, the gap between cores needs to be big enough to make sense and for Vega to have HBM2 it really needs more shaders to take advantage of that memory bandwidth. Both of these things would lead to a fairly logical conclusion that small vega is bigger than the 1080 and probably by a reasonable amount, it has to be big enough to create a significant gap to RX480, for both itself and salvaged parts to be made from the core. Now be an adult and come up with a reason why small Vega would be only 80mm^2 bigger than RX480 and thus only the same size as the 1080 die. Come up with a reason that attempting to match shader count and die size to HBM2's available bandwidth doesn't make sense. I can think of a few myself.

I reasonably discuss the reasoning behind rumours, I'm not at all always right but I put across an adult discussion that only descends into farce when people like you refuse to discuss rumours in a sensible manor. I think this will happen because X,Y and Z, I don't think that will happen because this that and the other, etc. Not just, it won't happen because I say so and you can't show me the internal design plans for an unreleased product therefore you are definitely wrong.

There ARE leaks that puts small Vega with both HBM2 and 4096 shaders, in fact that 490x rumoured spec has been around for a VERY long time.

But rather than respond reasonably and discuss the merits of my conclusions and either agree or disagree with your own well reasoned counter argument and can almost immediately tell what kind of reply I'll get, why not confound expectations and do something different.

Tbh If those charts that AMD showed weren't lies they could conceivably make a 1080 size gpu and put good daylight between itself and the RX480.

Tbh that chart has left me confused because even if vega just had fraction of the p/w improvement that Polaris did at 4096 cores would be offer a ridiculous performance increase over the 480.

Unless AMD plan to release it at a power envelope very similar to the rx480.
I suppose HBM2 would help with that.
 
Last edited:
"ding dong ding"
Tannoy announcement

Please can everybody on station platform 1 who are heading to AMD hype valley get back on the AMD hype train as its about to leave ...the Amd train is hoping to reach speeds of 2500mhz... please all climb aboard and enjoy the hype....

All those debating re nvidia please make there way to platform 2 however the nvidia train is delayed due to a paper launch ... Sorry for the inconvenience it should be arriving shortly..

One way ticket for one please
 
Urgh, if the stock situation on the 29th is bad it's going to undo all of AMD's hard work.

I'm hoping for something that comes with a beefy cooler and overclock already included.
 
It's funny that your claims never need evidence, but everyone else's you don't believe do.

Why don't you discuss things like an adult.

I've made a point, the gap between cores needs to be big enough to make sense and for Vega to have HBM2 it really needs more shaders to take advantage of that memory bandwidth. Both of these things would lead to a fairly logical conclusion that small vega is bigger than the 1080 and probably by a reasonable amount, it has to be big enough to create a significant gap to RX480, for both itself and salvaged parts to be made from the core. Now be an adult and come up with a reason why small Vega would be only 80mm^2 bigger than RX480 and thus only the same size as the 1080 die. Come up with a reason that attempting to match shader count and die size to HBM2's available bandwidth doesn't make sense. I can think of a few myself.

I reasonably discuss the reasoning behind rumours, I'm not at all always right but I put across an adult discussion that only descends into farce when people like you refuse to discuss rumours in a sensible manor. I think this will happen because X,Y and Z, I don't think that will happen because this that and the other, etc. Not just, it won't happen because I say so and you can't show me the internal design plans for an unreleased product therefore you are definitely wrong.

There ARE leaks that puts small Vega with both HBM2 and 4096 shaders, in fact that 490x rumoured spec has been around for a VERY long time.

But rather than respond reasonably and discuss the merits of my conclusions and either agree or disagree with your own well reasoned counter argument and can almost immediately tell what kind of reply I'll get, why not confound expectations and do something different.

So you have no solid facts to base any claims on and at the moment like everyone else you have no idea what performance Small Vega will give.

I will believe and accept the performance when I see reliable tests/reviews, this is the only logical thing to do.
 
i was just going by the definition someone linked her to argue that Nvidia's WAS a paper launch - "A paper launch is the situation in which a product is compared or tested against other products of the same kind, despite the fact that it is not available to the public at the time."

AMD compared the RX480 to a 1080 with a performance benchmark but it was not available to buy at the time. So technically even the RX480 is a paper launch by that definition.

AMD showing one tiny comparison to another card in one very, very narrow instance does not constitute a launch. Reviewers aren't allowed to show off figures, there is no general performance information officially disclosed. AMD have not launched the 480. Conversely, whilst there is some product out there, the 1080 has been largely a paper launch in that neither Nvidia's most people are unable to buy one and Nvidia's own preferred partners are struggling to get chips out of them.
 
AMD showing one tiny comparison to another card in one very, very narrow instance does not constitute a launch.
Putting out reviews of a card that has a very specific later release date arguably does not constitute a 'launch', either.

Please stop arguing over semantics here. It is absolutely meaningless.

Conversely, whilst there is some product out there, the 1080 has been largely a paper launch in that neither Nvidia's most people are unable to buy one and Nvidia's own preferred partners are struggling to get chips out of them.
None of which has to do with the stated definition of a 'paper launch' being discussed by Jono.

If you want to get picky and try and use 'paper launch' as a disparaging term, then it usually means when NO product is available upon actual launch(release date) due to delays. What we're seeing with Nvidia is more of a soft launch, meaning limited supply.
 
Last edited:
AMD showing one tiny comparison to another card in one very, very narrow instance does not constitute a launch. Reviewers aren't allowed to show off figures, there is no general performance information officially disclosed. AMD have not launched the 480. Conversely, whilst there is some product out there, the 1080 has been largely a paper launch in that neither Nvidia's most people are unable to buy one and Nvidia's own preferred partners are struggling to get chips out of them.

lol
 
Booooring

AMD showing one tiny comparison to another card in one very, very narrow instance does not constitute a launch. Reviewers aren't allowed to show off figures, there is no general performance information officially disclosed. AMD have not launched the 480. Conversely, whilst there is some product out there, the 1080 has been largely a paper launch in that neither Nvidia's most people are unable to buy one and Nvidia's own preferred partners are struggling to get chips out of them.

Putting out reviews of a card that has a very specific later release date arguably does not constitute a 'launch', either.

Please stop arguing over semantics here. It is absolutely meaningless

This is like little kids in a playground going:

"nah, nah, nah nah, nah... you did a paper launch... nah, nah, nah nah nah..."​

and then the other kid goes:

"no, stop it! YOU did a paper launch! stop it I tell you!"​

Now queue that for 3 pages worth of posts...
 
The thing that makes me double back on the rumour is that NVIDIA aren't a bunch of dumdums sitting in a factory making 10 GPUs per week - if they knew AMD had a $300 AIB card coming out that can OC to match or overtake a 1070 (and I'm fairly sure they have the resources to gain this information about AMD), I just don't see why they'd price the 1070 and 1080 so high. Perhaps to mug off early adopters before the 480 comes, but I see that as unlikely as it'd be a massive **** you to all the pascal buyers in the past three weeks.

Dude Nvidia does not care - we're talking about company that willing dumped faulty chips on the market to around the tune of 30+ million dollars - which ended up costing the industry; their partners several billion dollars - some estimate 30 billion dollars......this was bumpgate. Do you think they've changed? No - look at 970 memory - we don't care we lied to you; we're not apologizing as its a new feature.........

Seriously Nvidia really is shady as it gets when it comes to business practices. This is coming from personal experience as I've had to work with them; each time - left even more bitter taste in my mouth; (also dealing with several hundred laptops all failing at the same time with nvidia chips cost company I was tens of thousands of dollars and IT manager's job)

Nvidia does make some great hardware - but their business practices....no god no....they don't care who they hurt or **** off; can see that in the mobile industry; console; etc...
 
Back
Top Bottom