What settings are you using?
everything maxed on 1080p without ubersampling.
Last I can remember I had terrible negative scaling with crossfire. I'll make sure everything is up to date and get back to you on that.
Ran a benchmark of the opening siege. I benchmarked just the part where you fight your way to the ballista and then stopped it after I fired the bolt from it.
This is my result: http://imgur.com/Dc52Vdl
I was not getting any use out of the second card, but everything out of the first. CPU usage was low as well never going over 30%. I feel like the 670 should beat the 7950 at this game, but with Crossfire working correctly it shouldn't come close.
wow thanks a lot man. i was really curious to see the effects of the crossfire because I was thinking about doing dual 7950's and selling my 670. especially with this game because it is my favorite game and I am way too overly excited for witcher 3.
You are very welcome! I would honestly go with the new R9 280X in crossfire. Though the prices of the 7950s are waaaay down. This game has been the only game that really didn't react to crossfire at all. The other games I have benchmarked above were at least using the 2nd card 50%. Witcher 2 just didn't want anything to do with it. I'm hoping that Witcher 3 will be better optimized for Crossfire. I can only cross my fingers and hope, because I am a Witcher nerd as well :)
Just realized that I pulled a huge dumb *** move and ran that benchmark in windowed mode. Crossfire doesn't work in windowed mode.. Ran the same test again and got:
Both cards ran around 60-80% usage on average.