OH...MY...GODS! GF 8800 series totally badazz!

Status
Not open for further replies.

Guden Oden

Senior Member
Legend
If you've stopped by dailytech today - and maybe other sites as well - and taken a sneak peek at what the 8800 will be like, I think you're going through equal feelings of giddyness and drool-production overflow!

Just look at the raw base stats... 128 shader processors, 1.3GHz clock... 38Gpix fillrate. Over 80GB/s memory transfer rate. GAH! No wonder we need a small nuclear reactor to power the thing. :)

The future is surely looking brighter than ever for us GPU-lovers. I can't think of one single piece of hardware that has looked as impressive to me at this early stage. Usually, the first generation of hardware typically sucks at using the new features introduced, but somehow I think maybe this won't be the case with this GPU.

Indeed interesting stuff!
 
It say it use's uinfied shaders, but did'nt Nvidia state that there going to SKIP unified shaders for the G80 series?
 
That thread is entirely too big and unwieldy. Surely anyone with half an ounce of sanity realizes that.
 
Surely anyone with half an ounce of sanity realizes that.

Surely anyone with half an ounce of sanity will reserve judgement until all the facts of the cards are on the table. ;)
 
That thread is entirely too big and unwieldy. Surely anyone with half an ounce of sanity realizes that.

And you posted this in the entirely wrong forum. Anyone with half an ounce of sanity realizes that also.

Actually, the above thread is pretty easy to navigate and if you start from post 802 and onward you can jump fully in.
 
One thing to keep in mind is that G80's shader processors are apparently not as wide as G70's (see post #845). Supposedly.

I mean, if we're talking 128 4D ALUs running at twice G71's clocks, at 550MHz that's basically 4x the shader power of a 550MHz 7900GTO. I'm not sure G80 will be quite that bad-asssss. If it's 128 1D ALUs at twice G71's clocks, it's got somewhat more shader power than a GTO, but will apparently be much more efficient in using those ALUs. Combine that with loads more bandwidth, and things are looking good.
 
Status
Not open for further replies.
Back
Top