ATI Radeon HD 3850 256MB Crossfire
|
Author:
Editor:
Sponsor:
Published:
|
Max Slowik
Beth
AMD
May. 14, 2008
|
|
|
|
|
|
Performance Summary
Again, with anti-aliasing off, these cards did even better than I expected--on average, there was a 20-40% boost in FPS. 3DMark showed that two 3850s are theoretically superior to an 8800 GTX, and when AA is taken off the table, that's totally true.
Unfortunately, things didn't improve as much when AA was turned up. It's by no means a shoddy result, but it only makes 4xAA playable, and doesn't really future-proof the build.
Power and Noise
Two fans didn't really change the noise characteristics I wrote about in the first article:
"At idle, the fan spun only enough to stay spinning, and in an open-air environment, the hard drive was much louder than the video card. Once a game started, the fan speed picked up not long after, but for the most part, it was only a moving air sound. When I tested the cards for overclocking, the fan's motor became audible over the air sounds, but it was still quiet. Inside a case, it would be hard to isolate the fans in any circumstance, and certainly not over game noise."
Doubled up they do become noticable. Not offensive, but noticeable.
Idle (integrated graphics): 70W
Idle (w/ GPUs): 109W
CPU loaded, GPU idle: 159W
CPU and GPU loaded: 279
   Idle power consumption 39W
+ Load power consumption 120W
= Total power consumption 159W
Two cards together shape a different argument. In CrossFire, 3850s draw more power than an 8800 GTX, by about 30W.
Overclocking
Knowing that one card could run at 770 core/ 1040 memory, I just set that speed manually for both cards, and had no problems. I might have gotten lucky, because an extra 20% is nothing if not remarkable.
Add Comment
To add a comment without being a member, you may omit the password field, but you must enter your name (or nickname) along with your comment. * Denotes required fields.
|
Fidgit Oct. 27, 2009 - 11:10 pm
|