AMD: R8xx Speculation

[Kaotik]

keritto, Juniper has 128bit (or in theory, it could have 256bit too, but that's doubtful I suppose) membus, 192 is out of the question:

 

[Rangers]

September 16 !

So are you trying to say the review/preview NDA is up 9/16 instead of the rumored 9/24?

 

[Kaotik]

So are you trying to say the review/preview NDA is up 9/16 instead of the rumored 9/24?

Wasn't the rumored date 9/22?

 

[Arty]

This is just too confusing that some reviewers might get confused with the dates and press "publish" early and blame it back on AMD.

Sep 8, Sep 10, Sep 22, Sep 24

Its just like lottery.

 

[rpg.314]

Dave's latest sig is

Expand. Accelerate. Dominate.

aka Tessellate, gpu compute, kick nv's ass ?

It keeps getting more and more interesting

 

[seahawk]

Expand - 3 Displays
Accelerate - faster + Tessellate
Dominate - DX11

 

[air_ii]

Dave's latest sig is

Expand. Accelerate. Dominate.

aka Tessellate, gpu compute, kick nv's ass ?

It keeps getting more and more interesting

I'm sure Dave's having tons of fun changing his sig and watching the forum buzz about it :smile:

 

[psolord]

So...what happened in the UK presentation?

I eagerly logged on, only to read discussions about rv740 and display controllers of R200....again!

 

[flopper]

Dominate much likely is due to the speed of the chips running fps in the higher arena.
New features as dx11, all nice, but speed and new functionality as 3 screen support might be a winning move by ati.

The gamer dont see difference between 60fps and 80fps, but the gamer will notice if he can run world of warcraft on 3 screens that alone might make a huge move on Nvidia and allow for a really nice press if it works well....
Potentially 10million people play WoW but even more play other MMOG where the future of PC gaming reside.

 

[leoneazzurro]

I dont get you, do you deliberately miss that RV770 wasn't first ATi's 55nm part they put on market. Six month before RV670 parts based on same 55nm hit the market. And all they did with RV790 was fixing some bugs in current chip to compete with elephantiasis infected nvidia and their new 55nm gt200b chip. Maybe if nvidia responded better we'd saw something more of R700 than just bug fixed chip with huge ring-cap .... or ditch defence.

Yes, I was comparing very similar chip only. RV670 is on the same process but it is a smaller chip and with some changes in the architecture. Also RV790 was specifically targetted for increasing the clock on a proven architecture, but Cypress and other Evergreen products are the first DX11 parts. Of course ATI could have learned much from RV790, but I think they will be conservative with clocks in order to launch OC parts later if GT300 will be much faster.

Why are you all so fixed on SP's. My bet was 1920SPs and only 48TMU/24RBEs somwhere at the middle of this thread. When that German site spoiled it all when they hit us with "sure spec-ula(ifica)tions" ... 1200SPs @950MHz. And 1600SPs on 320mm2 seems that they projected these chip with that huge leakage as default 40nm TSMC process characteristic more than DX11 parts really need to waste so much of die space. And that ludacris (if we look ATis conservative historical approac) 64TMUs and 8-quad RBE seems that might add to that size a lot.

It was only a speculation about how to reach the projectes "2 Teraflops" figure and how to make the HD5870 and HD5850 different enough to justify the price increase.

 

[no-X]

keritto: Display controller (or display pipeline) creates/modifies the image, which is then transformed (by RAMDAC o TMDS) to a form, which can be accepted by the display device. I hope somebody could explain it with a better English ;-)

 

[Kaotik]

So...what happened in the UK presentation?

I eagerly logged on, only to read discussions about rv740 and display controllers of R200....again!

People have so far kept to their NDAs, so that happened.

 

[neliz]

there's enough info, you just have to ask the right entities.

 

[rpg.314]

The last time 4850 went on sale a week before official launch in europe, so somebody must have gotten the shaft this time. I suppose that is keeping everyone in check this time...

 

[w0mbat]

People have so far kept to their NDAs, so that happened.

Sadly

 

[KonKort]

I can confirm Charlie's news that HD 5850/HD 5870 will have 1600 SPs and can add that they will have 80 TMUs.

http://www.hardware-infos.com/news.php?news=3184&sprache=1

 

[mczak]

I can confirm Charlie's news that HD 5850/HD 5870 will have 1600 SPs and can add that they will have 80 TMUs.

http://www.hardware-infos.com/news.php?news=3184&sprache=1

Well if you confirm 1600SP, 80 TMUs doesn't add much since noone expected really different cluster organization .
Ok then guess I need to give up the proposal of 1280 SPs... Guess we expect moderate clocks then with lots of OC potential.
Now, the amount of ROPs... I'd bet on 32 though.

 

[fehu]

"only" 1600?

 

[Lightman]

"only" 1600?

I bet thanks to more sophisticated SCHEDULER it will be utilized a lot better than 800SP's in RV770/790!

Besides. who needs more for playing console ports on PC??

 

[Neb]

I bet thanks to more sophisticated SCHEDULER it will be utilized a lot better than 800SP's in RV770/790!

Besides. who needs more for playing console ports on PC??

I tweak and like to add mods to greatly enhance gameplay and visuals. I need all the power I can get.

Also tomorrow..TOMORROW dear tech warriors!