R700 Inter-GPU Connection Discussion

We actually do full shadowing on all light sources (each object blocks light from all light sources correctly), but majority of our light sources have rather limited distances (we have only distance limited spherical point lights and spot lights), so shadows are not that long usually. Meaning that majority of the shadow geometry is culled off nicely before sending it to graphics chip.

With unlimited shadow distances, all the objects have to occlude light (be "shadow casters"), or you get ugly looking artifacts from shadows that "go though geometry" (for example opponent behing a small hill on outdoor scene if the hill does not block the light correctly, or uses some alternative "fake" shadowing technique).

What about ambient occlusion ? For next game ?
 
At least Sampsa Kurri from Muropaketti told yesterday that his copy of R700 is in the mail. I would guess that he's not the only one who is soon visited by a courier :)

And here it is:

0707ati.jpg
 
AFAIK, AMD went for a preview with a couple of websites on the 14th of July and then a launch 2/3 weeks later.
 
I hope something leaks out soon.

I really want it to be something special - they figured out a way of, idk, violating the space-time continuum to transfer infinity data to and from each die, every ns.

But I know that it's just going to be silly little bus with a few GB/s...
 
I really want it to be something special - they figured out a way of, idk, violating the space-time continuum to transfer infinity data to and from each die, every ns.
Lol I kind of hope that if someone figured that out they'd have the perspective to know that GPUs - and indeed computation and the universe as we know it - would be fundamentally altered and there would be much more significant consequences than faster gaming ;)
 
Lol I kind of hope that if someone figured that out they'd have the perspective to know that GPUs - and indeed computation and the universe as we know it - would be fundamentally altered and there would be much more significant consequences than faster gaming ;)

See, now that's just crazy talk right there. :cool:
 

Confucious say, "Man who take 'up to' at face value before release probably needs someone else to cut his food for him."


The claim of 50% "on average" is much more interesting looking, but there are no doubt some unspoken caveats there as well --like, "once we've got the CrossFire profile for a particular game released". That's if we're lucky and it doesn't mean "at 8xaa 2560x1600".
 
Guessing at 4800x2 performance is certainly likely to be a bit more accurate than guessing at the performance of a new unknown part. Is anyone really expecting it to be significantly better than 4870s in crossfire?
 
There were some claims of it being up to 15% better than crossfired cards (on what, I don't know), but I don't know if everything else about the cards such as clocks and memory capacity would be equivalent.
 
Does anyone have a good breakdown of just how much bandwidth was being used by crossfire setups in the past? I know the cards are a good deal faster but I don't remember as much of a performance hit when using mobos with less than 16x/16x PCIe configurations.

Could it just be a matter of the interconnect actually having enough bandwidth to support two chips on top of slightly lower latency? With how ATI has been mapping memory on the cards it wouldn't really surprise me if they were sharing memory. I'd think a pair of x2 boards would saturate things fast. Might also explain why they seem to have beefed up the cache a bit.
 
Back
Top