So, do we know anything about RV670 yet?

AnarchX said:
CF 2.0: seems to be 3-way and 4-way-configurations and the possibility to use all outputs of the cards even in CF-mode.

Sure there can be more, but the this stuff should already much work for the driver-team.
Click to expand...

Link? 3/4-way CF I can see, but multi-display is something neither vendor has overcome yet. I would love to see it happen though.
 
CJ said:
Hehe. Ok... so my next guess is that tomorrow they'll post that RV670 will after all be launched (not a paper launch) in November. First thing I heard earlier this week was mid-November, but today I was told late November.
Click to expand...

Yes, finally... Fuad said the RV670 is coming on November now... RV670 is an R600 in disguise

Fudzilla said:
With 256 bit memory and 55 nm
RV670 will be a mainstream part that ATI plans to offer from November, but there is someething that Nvidia doesn't expect from this chip....
Click to expand...

Sorry for capturing some phase of out there... ;)
 
Quasar said:
...which would be done by...?
Click to expand...

Excellent question, as a total re-design is out of the question I would expect smaller hardware changes and perhaps some driver tweaks that could imrpove AA performance, as ATI is said to claim.

To answer the question how they want to do it, one would need to know which part of the R600 way is the bottleneck. Maybe color compression working again ?
 
ShaidarHaran said:
Link? 3/4-way CF I can see, but multi-display is something neither vendor has overcome yet. I would love to see it happen though.
Click to expand...

It will happen.... for 2 displays at least. This is what my info says about which driver versions should support it:

Extended Desktop
8.45 (Vista only, 2 displays)
8.47 (XP) Dec Jan
Click to expand...
 
CJ said:
It will happen.... for 2 displays at least. This is what my info says about which driver versions should support it:
Click to expand...
Awesome! :) It was high time one of the two IHVs focused on fixing this problem.
 
seahawk said:
Excellent question, as a total re-design is out of the question I would expect smaller hardware changes and perhaps some driver tweaks that could imrpove AA performance, as ATI is said to claim.

To answer the question how they want to do it, one would need to know which part of the R600 way is the bottleneck. Maybe color compression working again ?
Click to expand...

There were rumours of a fault in the R600 hardware causing such a severe AA penalty. Could have been fixed in RV670?
 
Mighty Dave, is there a remote chance the driver team could introduce edge-detect AA with the simpler box filters (x2~x8)? It seems that ED is throwing away a lot of redundant work/data, being passed back to the shader core.
 
msirv670XTX-04.jpg


msirv670XTX-02-1.jpg


msirv670XTX-07.jpg


http://www.expreview.com/img/news/070926/msirv670XTX-03.jpg
http://www.expreview.com/img/news/070926/msirv670XTX-05.jpg
http://www.expreview.com/img/news/070926/msirv670XTX-09.jpg
http://www.expreview.com/img/news/070926/msirv670XTX-11.jpg
http://www.expreview.com/img/news/070926/msirv670XTX-12.jpg
http://www.expreview.com/img/news/070926/msirv670XTX-13.jpg
http://www.expreview.com/img/news/070926/msirv670XTX-14.jpg
http://www.expreview.com/img/news/070926/msirv670XTX-16.jpg
http://www.expreview.com/img/news/070926/msirv670XTX-17.jpg
 
You must log in or register to reply here.

Similar threads

N
Hybrid SLI dying?
2 3
Replies
48
Views
12K
Fox5
F
J
Digital ecosystems Smart TVs Game consoles, cameras the technology and interaction
Replies
18
Views
5K
jeff_rigby
J
N
Nvidia GT200b rumours and speculation thread
27 28 29
Replies
564
Views
117K
neliz
neliz
A
  • Locked
NVIDIA GT200 Rumours & Speculation Thread
130 131 132
Replies
3K
Views
536K
Arun
A
P
AMD: M88 or Mobility RV670
Replies
3
Views
3K
aca
A
Back
Top