Each card renders 1 frame at a time, so an individual card can only use it's half of the memory to draw the frame, while the other card uses it's half to draw the next one. No individual card can benefit from the memory present on the other card.
In SLI configurations, you have a master and a slave card. When the slave is done rendering, it passes the framebuffer to the master for output.
This is assuming "alternate frame rendering" is used. "Split frame rendering" divides the workload in half and passes one half to the slave card. When the slave has completed it's task, the framebuffer is passed to the master and combined into a complete frame for output.
Hope I got my terms right.
Heck, hope I got it right at all.
* linky:
http://www.slizone.com/object/slizone_ask_mmm013.html
Quote:
|
The graphics memory is NOT doubled in SLI mode. If you have two 128MB graphics cards, you do NOT have an effective 256MB. Most of the games operate in AFR (alternate frame rendering). The first card renders one full frame and then the next card renders the next frame and so on. If you picture SLI working in this way it is easy to see that each frame only has 128MB of memory to work with.
|