![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
-
Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.
One more GPU, how much more performance?
- Thread starter sram
- Start date
PrincessFrosty
Supreme [H]ardness
- Joined
- May 6, 2009
- Messages
- 5,905
There's 2 things here you should understand.
Firstly scaling in multiGPU configurations is not perfect, and it tends to differ between games, some games scale almost perfectly (~100%), CoD MW/MW2 seem to scale incredibly well for example, some other games is might be much lower like only 60-70% extra performance.
Secondly, you need to understand bottlenecking. You're increasing your overall graphics processing power with SLI/Crossfire but with no upgrade to your CPU you run the risk of the CPU becoming a bottleneck as far as frame rates go. Often times multiGPU isn't used to increase frame rate but instead allow you to use higher settings (higher resolution, more AA/AF) while maintaining the same frame rate.
Now if you have a kick ass CPU then naturally you will see a good FPS increase by putting in another GPU, but some games will remain CPU limited, for example most people are CPU limited in BFBC2 and GTA IV, these games specifically seems to be a real CPU hog, multiGPU in this case wouldn't increase FPS in the game.
What it does reliably allow you to do is increase the visual settings with little to no impact in frame rate, so even in the cirumstances in BC2 and GTA IV where you see no FPS increase you'll be able to crank the AA/AF and in game settings and see less (or none) penalty for it.
Firstly scaling in multiGPU configurations is not perfect, and it tends to differ between games, some games scale almost perfectly (~100%), CoD MW/MW2 seem to scale incredibly well for example, some other games is might be much lower like only 60-70% extra performance.
Secondly, you need to understand bottlenecking. You're increasing your overall graphics processing power with SLI/Crossfire but with no upgrade to your CPU you run the risk of the CPU becoming a bottleneck as far as frame rates go. Often times multiGPU isn't used to increase frame rate but instead allow you to use higher settings (higher resolution, more AA/AF) while maintaining the same frame rate.
Now if you have a kick ass CPU then naturally you will see a good FPS increase by putting in another GPU, but some games will remain CPU limited, for example most people are CPU limited in BFBC2 and GTA IV, these games specifically seems to be a real CPU hog, multiGPU in this case wouldn't increase FPS in the game.
What it does reliably allow you to do is increase the visual settings with little to no impact in frame rate, so even in the cirumstances in BC2 and GTA IV where you see no FPS increase you'll be able to crank the AA/AF and in game settings and see less (or none) penalty for it.
Last edited:
- Joined
- Jul 11, 2001
- Messages
- 12,825
Well there are a few things to consider. Is 100% scaling even a realistic goal?
Consider that the 5770, specs wise, is basically half of a 5870. Two 5770s perform almost exactly the same as a 5870. So in that case you are talking about Crossfire scaling with performance the same as if you added the stream processors from the 2nd card onto the first directly. To me that is amazing scaling.
Two 5770s generally won't be 100% faster than one 5770, but a 5870 isn't 100% faster than a 5770 either. If doubling up the resources on a single card doesn't produce 100% scaling why would we expect that from crossfire?
Consider that the 5770, specs wise, is basically half of a 5870. Two 5770s perform almost exactly the same as a 5870. So in that case you are talking about Crossfire scaling with performance the same as if you added the stream processors from the 2nd card onto the first directly. To me that is amazing scaling.
Two 5770s generally won't be 100% faster than one 5770, but a 5870 isn't 100% faster than a 5770 either. If doubling up the resources on a single card doesn't produce 100% scaling why would we expect that from crossfire?