We'll now show you some tests we have done on overall power consumption of the PC. Looking at it from a performance versus wattage point of view, the power consumption is really not bad. Our test system is a Core 2 Duo X6800 Extreme Processor, the nForce 680i SLI mainboard, a passive water-cooling solution on the CPU, 2GB memory, DVD-ROM and WD Raptor drive. Have a look:

The methodology is simple: we look at the peak wattage during a 3DMark05 session with hefty IQ settings to verify power consumption. It's a good load test as both GPU and CPU are utilized really hard here. Please do understand that you are not looking at the power consumption of the graphics card, but the consumption of the entire PC.

So let's analyze. We had a total system wattage peak at roughly 360 Watts with the HD 2900 Pro card, which is really quite a lot. We simply place a wattage meter in-between the PSU and power socket. It's not the most objective way to test as you have to consider PSU efficiency as well, but it's the best thing we can do, though.

In my view the Radeon HD 2900 XT requires you to have a 500 Watt power supply unit at minimum if you use it in a high-end system, and I think that's barely on the safe side. Also recommended is 35 AMP's on the 12 volts rails for stable power distribution.

Notice that the card uses one 6-pin power connector and one 8-pin, I'd like to strongly advise a PSU with dual 12-volts rails here. You can also connect 2x 6-pin connectors yet overdrive (overclocking) will be disabled. If you wan to do it really right; check out this PSU review please.

Why does the Radeon HD 2900 Pro have two 6/8-pin power connectors? Well, the PCI Express spec allows for 75W from the edge connector plus 75W from each external 6-pin power connector. Then 75 Watts from the PCI-Express bus. The trick here is that the 8-pin connector can draw 150 Watts.

So mainboard 75 Watts, 6-pin connector 75 Watts, 8-pin connector 150 Watts. That's a total of 300 Watts at your disposal for a card that (non-overclocked) peaks at 215 Watts.

If you have dough to spend and opt the guru path of righteousness by doubling up towards two cards in your system -> Crossfire, then you should end up with a 800 Watt or better PSU with a 45 AMPs 12 Volts rail. Again, definitely check out this PSU review, please.

There are many good PSU's out there, please have a look at our many PSU reviews as we have loads of recommended PSU's for you to check out in there. What would happen if your PSU can't cope with the load?:

• bad 3D performance
• crashing games
• spontaneous reset or imminent shutdown of the PC
• freezes during gameplay
• PSU overload can cause it to break down

Les températures de la carte graphique

As with any graphics card we test, we also take a look at heat build-up in the GPU. We measured at a room temperature of 22 Degrees C.

Anyway, the Radeon HD 2900 Pro is running with quite normal temperatures. In idle expect roughly a 50 Degrees C temperature and at maximum we notice a peak of roughly 70 Degrees C.

It does create a lot of that heat though, but the beauty of the cooler design is that this heat is pushed outwards of your PC as exhaust at the connector side.

Noise Levels coming from the graphics card

When graphics cards produce a lot of heat, that heat usually needs to be transported away from the hot core as fast as possible. Often you'll see massive active fan solutions that can indeed get rid of the heat, yet all the fans these days make the PC a noisy son of a gun. I'm doing a little try-out today with noise monitoring, so basically the test we do is extremely subjective. We bought a certified dBA meter and will start measuring how many dBA originate from the PC. Why is this subjective, you ask? Well, there is always noise in the background, from the streets, from the HD, PSU fan etc etc, so this is by a mile or two not a precise measurement. You could only achieve objective measurement in a sound test chamber.

The human hearing system has different sensitivities at different frequencies. This means that the perception of noise is not at all equal at every frequency. Noise with significant measured levels (in dB) at high or low frequencies will not be as annoying as it would be when its energy is concentrated in the middle frequencies. In other words, the measured noise levels in dB will not reflect the actual human perception of the loudness of the noise. That's why we measure the dBa level. A specific circuit is added to the sound level meter to correct its reading in regard to this concept. This reading is the noise level in dBA. The letter A is added to indicate the correction that was made in the measurement. Frequencies below 1kHz and above 6kHz are attenuated, where as frequencies between 1kHz and 6kHz are amplified by the A weighting. 

We start up a benchmark and leave it running for a while. The fan rotational speed remains constant. We take the dBA meter, move away 75 CM and then aim the device at the active fan on the graphics card.

There's no other way of saying this ... the card is noisy. With the fan in idle it's really okay yet when you are heating up that graphics card at certain intervals you can expect a 53 dBa volume level coming from the PC. This is very noisy for a graphics adapter.