Just curious, you have some info on GCN2 when you're expecting (supposed) 768:40:16 GCN2 card to match 1280:80:32 GCN card?

No, I don't have any info. It's just the non-written yet usually true "law" that happens everytime a new gen of cards gets launched. If we look at the last AMD cards we see that the second tier cards, the x8xx ones, usually match the performance of the previous first tier cards, the x9xx ones.

HD 4870  ~ HD 5770

HD 5870  ~ HD 6870

HD6970  ~ HD 7870

And even Nvidia follows the same trend

GTX 285 ~ GTX 460

GTX 480 ~ GTX 560Ti

GTX 580 ~ GTX 660Ti

I agree that this evolution in performance not always translates well into the rest of the cards (the 6870 is faster than the 7770), but it's a good aproximation on how the cards evolve and what to expect from the new cards.

Besides some time ago appeared this table

Excluding the the price, which won't be that (the 8870 at that price would make the 7950 look like a bad joke), if the expected performance increase isn't too far off from the end result and the other cards see the same increase in performance, the 8770 could end with the same performance as the 7870, or very close.

I'm pretty sure that chart is bullshit tho I mean on the same process 600m more transisters at a slightly higher clock rate and it somehow has a lower TDP? I find that hard to belive myself.

Yes, I know and I also dislike that bit. I only mentioned the price being BS but the thermals are also suspicious.

But you know the usual excuses, with the 28nm process being more mature, with the tweaks that AMD/Nvidia can do to improve efficiency, etc. they may have been able to reduce the voltage of the chips as well as reduce the redundant sectors that the first 28nm cards had to improve yields. :-/

With luck, we'll see it in 4 months.

While I mostly agree, here's some comparison of those cards (with Voodoo ratings):

4870vs 5770 - 78VP vs 76VP (VLIW5 800:400:16@750/900; 256bits) vs (VLIW5 800:40:16@850/1200; 128bits)

5870vs6870 - 132VP vs 120VP (VLIW5 1600:80:32@850/1200; 256bits) vs (VLIW5 1120:56:32@900/1050; 256bits)

6970vs7870 - 156VP vs 173VP (VLIW4 1536:96:32@880/1375; 256bits) vs (GCN 1280:80:32@1000/1200; 256bits)

As you can see:

in case of 4870 and 5770 we have VLIW5 with same config, higher clock on 5770, but lower bus width for same performance on average.

In case of 5870 and 6870, we have VLIW5 with more of eveything on 5870, same bus width, for better performance of 5870, lagging only in pixel fill rate

In case of 6970 and 7870, we have VLIW4 vs GCN, and though it seems that 6970 has lot more shaders and TMUs, equilized to 1GHz of 7870, 6970 would be more like 1352:84:28 card - not too different from 7870, but still, considering GCN is newer and quite improved architecture, 7870 wins by some 10%

Now, let's look at why I asked you if you have some inside info on GCN2:

7870 - GCN 1280:80:32@1000/1200; 256bits

8770 - GCN2 768:48:16@unkown clocks; 128bits

I hope you see where I got confused, even if GCN2 is jump as VLIW4->GCN, 8770 with those specs would need (in my guesstimate) to run at 1.5GHz to achieve 7870 level of performance.

Of course, there is a chance that 8770 is not 768:48:16 part, but 1024:64:32, what 8830 is presumed to be,  and 8830 is 1280:80:32 part - in that case I can see how 8770 can reach performance of 7870, if GCN2 is that better than GCN.

@HoloDust: I see your points and I won't even try to refute them. As I said it was a guesstimate from what seems to happen from gen to gen.

About the 8830 thing, is that card for real? Aren't those xx30 cards, the ones that AMD use to sell the chips that aren't good enough to be used on their xx50 or xx70 cards? If so, their specs will vary greatly depending on how good or bad the yields are.

Well, all those specs are guesstimates at this point I think (apart from what that slide was showing, 8870 and 8850, which can be considered rumour) - that't why I said I would expect to see 8770 with 1024:80:32 config (like 7850), and 8830 something lower than what (supposed) 8850 is (1536:96:32), so probably 1280:80:32.

Ok.

Given that some people here have been saying that the next consoles could use the mobility versions of the AMD cards, I thought the reveal of the new HD8000M series of chips would interest someone

full new here (Engadget)

Sadly no power figures are listed, but the specs ofthe chips are these:

No HD8800M chip has been announced, but it will likely be the same as the 8800M at higher frequencies.

There is also this performance comparison between the 8870M and Nvidia's 650M GS (sorry for the small size)

And some comparisons with the HD7000M cards from HotHardware

Not sure have you meant HD8800 (desktop) or HD8900m (mobile) there, but if you meant HD8900m, considering previous generations, HD8970m wiil be mobile version of HD8870 at some 75-85% clock - so, if rumours for 88xx series are true, that will be 1792:112:32 card at around 850MHz core clock - for comparison that's around level of 7950 Boost edition - I would be hugely surprised to see that in any console.