You can see performance difference in any game if you run 2x6600GT against 1x6600GT - in DAoC's case the difference wont be that big (noticeable at 1600x1200 though). The better your cards are the less benefits you get out of them in the current games.Brunore said:He means the newer nVidia cards that can be run in SLi not duel cpus.
As far as I know the games have to have code in them to make them SLi aware, so no its wont.
Saggy said:You can see performance difference in any game if you run 2x6600GT against 1x6600GT - in DAoC's case the difference wont be that big (noticeable at 1600x1200 though). The better your cards are the less benefits you get out of them in the current games.
No, they don't need to be SLi aware. It does help if the nvidia driver is aware of the game though.Brunore said:He means the newer nVidia cards that can be run in SLi not duel cpus.
As far as I know the games have to have code in them to make them SLi aware, so no its wont.
Danya said:No, they don't need to be SLi aware. It does help if the nvidia driver is aware of the game though.
In theory it should double the performance difference but in practice it doesn't and usually its far far away from it.Quinlan said:You can compare it relativly easy. If you have a new motherboard which supports 2 video cards one builds the lower part of the screen the other one the upper.
Ofcourse technically it aint that simple but for comparison reasons it is.
You can force SLi on in the driver (details are in fact in that hardocp article). What you said was true for some very early drivers, but the more recent versions allow you to force any game to use SLi. It's not always beneficial to do so though.Crookshanks said:No - I don't think this is correct unless numerous sites and magazine reviews are wrong - nvidia MUST know about the game and add it to its driver or there will be absolutely no speed improvment whatsoever as the game will just use the one card - see:
http://www.hardocp.com/article.html?art=NzEx
In fact - as a SLI configuration drops the number of PCI-E channels to each card from 16 (in a single card machine) to 8 on each - you might even find a slight decrease in performance (although tests seem to indicate that its hardly a noticable drop with any current game).
I agree with the above comments though - DAOC is largely CPU (and memory) constrained - not graphics.
Brunore said:He means the newer nVidia cards that can be run in SLi not duel cpus.
If the low end ram can't run @ 1t the performance difference will be huuuuuge on A64 I was about to get 2GB of ram but A64 can't handle more than 1GB @ 1t so I forgot about itDanya said:The difference between high end ram and low end ram in games is typically 1-3%, CPU will make a significant difference, as will the amount of RAM. Not sure on SLi vs normal as I don't have an account to test with, however given it's an AFR (alternate frame rendering) profile for toa the difference could be large. You'd likely need to be running high res with AA/AF on to see any difference though.
Overall, rating a module as 1T is either false advertising or needs to be excused as blatant ignorance as all unbuffered modules are capable of a 1T CMD rate up to four banks per channel, beyond which chipset limitations come into play
Low end ram seems to have problems to run at 1T in Dual Channel and its the CPU that determines it in A64's case (if I'm not mistaken?), not the MB's chipset if that was your point - 333DDR 1:1 @ 1T is faster than 400DDR 1:1 @ 2T. Test it by yourself Done plenty of testing with DFI SLI-D, A64, Winbond UTT / cheap Kingston and the difference between 1T and 2T was something stupid (more than two modules -> 2T, Kingston in DC -> 2T)Danya said:Might want to read this: http://www.mushkin.com/mushkin/pop-up/latencies.htm
Basically any ram can run 1T as long as the chipset is good enough.
Danya said:The difference between high end ram and low end ram in games is typically 1-3%, CPU will make a significant difference, as will the amount of RAM. Not sure on SLi vs normal as I don't have an account to test with, however given it's an AFR (alternate frame rendering) profile for toa the difference could be large. You'd likely need to be running high res with AA/AF on to see any difference though.