I just replaced a fanless 512M 8600GT (MSI, DDR2 version) with a fanless 512M 9600GT (ECS w/Accelero2). Here's the power usage comparison, as measured full-system at the wall by Kill-a-watt.
Idle at Windows desktop: 76W -> 92W
Gaming, high poly count, low framerate: 96W -> 125W
Gaming, low poly count, high framerate: 102W -> 145W
Performancewise, the new card gets about 50% better frame rate under low poly / high frame rate situations (60 fps -> 90 fps) and 100% better frame rate under high poly / low frame rate situations (11 fps -> 22 fps), all settings the same (all settings maxed, AA 4x, anisotropic filtering on, 1920x1200). I'm reading 60C temp on the GPU under max energy usage, 45C idle, ambient fairly warm (82F).
The power usage difference is more than I expected, but then so is the performance difference.
The closest I do to gaming is Second Life, so that's where those numbers came from. It's an environment where content comes from players, so it isn't optimized like most games. There is high polygon count and way too many large textures.
Here's the system:
E8400 3.0 GHz, undervolted
Ninja rev B, fanless
Gigabyte GA-P35-DS3L
G.Skill 4GB (2 x 2GB) DDR2 800
WD GP 500 GB HDD
Gigabyte i-RAM w/ 4 x 0.5G DDR 400
M-Audio Audiophile 24/96
Samsung DL DVD burner
Antec Sonata with Nexus 120mm exhaust and
the Ninja's 120mm Scythe fan intake
Sparkle 250W 80+ PSU
WinXP Pro
power consumption 8600GT vs 9600GT
Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee
power consumption 8600GT vs 9600GT
Last edited by amyhughes on Mon Jun 02, 2008 9:06 pm, edited 1 time in total.