AMD confirms R680 is two chips on one board

Well, i won't bother as it's too small of a difference to matter much anyway.
However, the positioning of the auxiliar power plugs in this R680 is in fact very poorly thought out.

Grievously so. I wonder what magical reason could there be for their placement and orientation:|
 
Grievously so. I wonder what magical reason could there be for their placement and orientation:|

Pretty obvious if you look at the right pictures...
Fan size.

r680.jpg


Do remember that this is still an early sample/prototype which are also subject to change.
 
Pretty obvious if you look at the right pictures...
Fan size.

r680.jpg


Do remember that this is still an early sample/prototype which are also subject to change.

They should somehow make dual FAN for R680 - otherwise the single FAN on this card will run to load at high RPM to cool 2 GPU's.
 
They should somehow make dual FAN for R680 - otherwise the single FAN on this card will run to load at high RPM to cool 2 GPU's.

How would you go about doing this?
This should work fine, since it effectively looks like the R600 fan, which was more than enough to handle the ~200w of heat.
 
Single-slot design is sufficient for silent cooling of single RV670. So, dual-slot should be sufficient for silent cooling of dual RV670 configuration :) I think this card will be a bit louder than HD3870, but certainly more quiet than HD2900XT.
 
Pretty obvious if you look at the right pictures...
Fan size.

r680.jpg


Do remember that this is still an early sample/prototype which are also subject to change.

I was eliptical...it's not only the placement, it's also the orientation.Why have them perpendicular on the board and not perpendicular on the cooler/PEG slot?This MAY be nice for cases which have the PSU placed at the bottom in a multi chamber design, but if you have a top mounted PSU, it's disadvantageous IMO.
 
You know, I more and more appreciate the idea of external intake & exhaust for graphics HSF solutions, as per NV30. No problems with "negative" or "positive" case pressure affecting the cooling performance of the card...
 
You know, I more and more appreciate the idea of external intake & exhaust for graphics HSF solutions, as per NV30. No problems with "negative" or "positive" case pressure affecting the cooling performance of the card...

You call this "thermal efficiency" ? :p
 
You call this "thermal efficiency" ? :p
No, you drew that bow. Anyhow, what's efficiency got to do with mediocre implementation? :p

However, insomuch as the graphics card's cooling (airflow) was isolated from the rest of the system, it was a good idea. With multiple 120mm+ exhaust case fans, the graphics card's exhaust can stall... Other than a separate chamber, some people are reverting to "positive" pressure case cooling to alleviate GPU overheating.
 
Hmm, is it me, or this 3870X2 board here is a tad shorter than the initial sample shown few weeks ago? :rolleyes:

This is a custom designed board. It's also 2x6 pin instead of 1x8 pin+1x6 pin on the reference boards. Thus, it may be shorter then those, although some ppl suggested that this'll be the length of all 3870X2 PCBs. We'll see soon, I guess.
 
So the 6pin/8pin power plugs will be indeed in that totally awkward position, perpendicular to the PCB... :???:

Didn't someone say in this thread that it was like that in earlier pics because it was merely an engineering sample ?
 
Back
Top