Power Consumption, Temperature, Noise, Overclockability
We used a nonstandard GeForce GTX 460 1GB to measure the power consumption of GF104-based solutions in our earlier tests, so now we want to recheck the results using the Zotac card and our testbed configured like follows:
- Intel Core 2 Quad Q6600 CPU (3GHz, 1333 MHz FSB x 9, LGA775)
- DFI LANParty UT ICFX3200-T2R/G mainboard (ATI CrossFire Xpress 3200 chipset)
- PC2-1066 SDRAM (2x2 GB, 1066MHz)
- Enermax Liberty ELT620AWT PSU (620W)
- Microsoft Windows 7 Ultimate 7 64-bit
- CyberLink PowerDVD 9 Ultra/"Serenity" BD (1080p VC-1, 20 Mbit)
- Crysis Warhead
- OCCT Perestroika 3.1.0
The new testbed for measuring electric characteristics of graphics cards uses a card designed by one of our engineers, Oleg Artamonov, and described in his article called PC Power Consumption: How Many Watts Do We Need?. As usual, we used the following benchmarks to load the graphics accelerators:
- CyberLink PowerDVD 9: FullScreen, hardware acceleration enabled
- Crysis Warhead: 1600x1200, FSAA 4x, DirectX 10/Enthusiast, "frost" map
- OCCT Perestroika GPU: 1600x1200, FullScreen, Shader Complexity 8
Except for the maximum load simulation with OCCT, we measured power consumption in each mode for 60 seconds. We limit the run time of OCCT: GPU to 10 seconds to avoid overloading the graphics card's power circuitry. Here are the obtained results:
The graphics cards from Nvidia and Palit/Gainward behave in a similar way in desktop mode, having a peak power consumption of 20 watts. And then they differ. Particularly, the reference card has a higher peak power draw when performing such not very heavy tasks as decoding HD video. When it switches into power-saving mode with reduced clock rates, its average power consumption is about 23 watts, which is comparable to the results of the Gainward card, but the latter is quicker to switch to save power.
The reference card from Nvidia also needs more power in modern games but the difference is less than 10 watts and can be neglected. Take note that the first power connector is loaded more than the second one: 7.4 and 3 amperes, respectively. So, the GeForce GTX 460 might have been even equipped with only one 8-pin PCIe 2.0 power connector (with a load capacity of 150 watts), but not all power supplies, especially in the midrange segment, offer them.
Thus, we don’t see anything new in our second test of the power consumption of the GeForce GTX 460 1GB. Its results are quite acceptable for its performance although inferior to same-class solutions from AMD.
Nvidia’s reference cooler is less effective than the cooling system installed on the Gainward/Palit card. This might have been expected, considering the smaller heatsink and the DVI connectors getting in the way of the hot air being exhausted out of the system case. On the other hand, the card was absolutely stable even under very hot weather. The reference cooler is going to be just fine for ordinary users who don’t intend to overclock the card. Otherwise, you need something better.
Considering the above-mentioned downsides of the reference cooler, the GeForce GTX 460 1GB is expectedly loud. It is as loud as the flagship GeForce GTX 480 in 3D mode and not silent in 2D mode, either. Granted, the weather was hot, but Gainward released a much quieter version of this card with higher GPU clock rates, so that’s not an excuse. If you value silence, you may want to replace the reference cooler or even buy a GeForce GTX 460 with a nonstandard and better cooler.
As for overclockability, the GPU of our Zotac GeForce GTX 460 1GB was stable at 820/1640 MHz and its memory, at 1065 (2130) MHz. This is somewhat higher than what we achieved with the Gainward GeForce GTX 460 GS GLH. The difference of 20 MHz wasn’t worth benchmarking the card once again in overclocked mode whereas our readers already know how fast the GF104 is at 800/1600 MHz. So, we will just focus on our main subject which is to benchmark the GeForce GTX 460 1GB in SLI mode.