He's obviously talking about the two benchmarks that he specifically said: hyperlight and galaxy. Those two show exactly what he described, which is why he asked. I don't even know why you responded in this way...
In any case, it seems that it's something specific to either the benchmark itself, or the OS, or some other odd dependency. If you take notice in Galaxy, the GTX285, RV790 and RV870 all score identically. These obviously are not all identical cards; they have different ROP, raster, texture, shader and bandwidth performances. So how are they all identical in performance? Says to me that something else is bottlenecking them at a systemic level.
Notice that a second card (in both NV and ATI formats) scales almost exactly 2x compared to their "base" card. I can't say what that tells us, but the fact that both vendor single cards are precisely bottlenecked to the same score, and then both vendors X2 cards are precisely bottlenecked at 200% of the same score has to be an obvious pointer to something...
Hyperlight is the same story in single card mode (with a tad bit of variance), but look what it does to the X2 configs. Whatever made Galaxy go to "ludicrous speed" is obviously face-planting Hyperlight.
Overall, I agree with IXBT's stance that these two benchmarks seem to be of little value, as they don't show enough data to really tell us what's going on.
