AMD: Southern Islands (7*** series) Speculation/ Rumour Thread

But Cayman wasn't the highest performing part of the 32nm product line. It's likely that the top model would require 8+6pin configuration anyway.

 

Cypress engineering samples had 6+8 pin connectors and even the final reference version had solder points for 6+8 pin (with two 6 pin connectors soldered to it). That's what I meant when saying ES may have an oversized power circuitry (what does not mean that the HD7970 reference version won't be 6+8 pin).

 

btw board is REV 1.B, do you think there will be another revision ? board and cooler seems to be at latest state to me..

 

Cayman ES boards were also with red PCB, but the cooling assembly was pretty much the same as the final product.

 

From the pictures I know, it could be also "1.0" (you know, the zero with a kind of backslash through it to distinguish it from an O).
And that early boards get fitted with the cooling solution of Cayman for instance doesn't appear too surprising to me.

 

well it looks more like 1.B than 1.0/O..
phk says about 7900 card have finished and confirm production.. also 7800(pitcairn) will be released @Q2

http://bbs.expreview.com/thread-48285-1-1.html
http://bbs.expreview.com/thread-48649-1-1.html

 

This is a "best-case" scenario, but I don't know how things have progressed in the driver department in hybrid crossfire between APUs and discrete GPUs.
Maybe it's more, maybe it's less, hence the "if".
Nonetheless, if the notebook is paired with 1600MHz DDR3, the jump in the iGPU performance should be considerable, as should be the hybrid crossfire results.

 

Well there are quite a few GTX560s....

If Tahiti is ~210-220w, that is pretty similar to the GF110 GTX560s.
I'm thinking it is Pitcairn that will have the similar power consumption to the other GTX560's, probably around 150-160w.

 

if powertune is implemented then I dont think the power input connectors indicate anything about average power draw or max board power - both can be managed and constrained. logic suggests that you dont fit 6pin and 8pin inputs on a board that needs less than 225w max board power...

 

you might if it is close (every board isn't going to be the same exact power draw), the cost from 6+6 to 6+8 is going to be pennies at most

 

I just see a yellow/grey blob there.

 

Zoomed out its a little clearer, its either 1.B or 1.3.

 

Looks like 1.3 to me.

Though it has to be said, the 3 doesn't look much like the other 3's.


Could it be a 9?

 

I hope the flagship single GPU card blows away my OC'd GTX580. My money is yours ATI! Just prove yourselves.

 

9 revisions, heh, I hope not.

 

7950 5% after a 580gtx?
7970 30-40% more than 580?

 

If this is the case, consider me sitting out for another round of gfx cards.

 

What did you expect? If GTX 580 is anywhere between 15 - 30% faster than 6970, then 30% faster than 580 would be 50 - 70% over 6970. Not to shabby, is it?

But then again, I still doubt that 30%.

 

Not in gaming it isn't.
At 1920x1200 GTX580 averages 11.88% faster, 2560x1600 only 8.49% faster than HD6970 ( http://www.techpowerup.com/reviews/Zotac/GeForce_GTX_560_Ti_448_Cores/27.html - using TPU as source as they have the largest list of games tested )

 

A 580 is maybe 10-15% faster than a 6970. And a 6970 is maybe 50% faster than a 5870. And a 5870 is maybe 10% faster than my 4870x2. So, yes, only ~100% faster than my 4870x2 three years later is pretty shabby.

I want a return to the days of 9700pro being 5x faster than 4600Ti, after just 1 generation.

I want 90% of 6990 speed from a 7970, realistically.