Sapphire Radeon HD4850 512MB Graphics Card Review

Temperature Testing

Unfortunately, it does not seem like the single slot heatsink is able to keep temperatures down much when compared to some other cards on the market. However, it should be noted that that the maximum temperature we reached WOULD not be passed by our card since once it hit this peak, the fan would increase speeds slightly to compensate. Even though the temperatures are well within the normal operating range of the R770 core, what concerns us most is the amount of hot air the HD4850 dumps back into your case. In an optimal situation, any hot air from the GPU should be exhausted immediately so while the single slot cooler may look nice; this is one of its major tradeoffs. We would recommend that you ensure there is sufficient air movement within your enclosure so the heat produced by this card does not make a home for itself around critical components.

Acoustical Properties

It seems that ATI was scared quite a bit by the reviews it received back a while ago with the HD2900XT release which stated that its cooler was louder than a semi downshifting on the highway. Now all of their cards feature heatsinks that happily trade off cooling power for near dead silence and the HD4850 is no different. While we love silence as much as the next guy, there is a fine balance to be struck between performance and a quiet fan but ATI seems to have it wrong here. Let’s put it very simply: the fan on the HD4850 is damn quiet but the core gets too hot for our liking. Even when the fan increases its speed about 20% to keep temperatures under control, the noise it produces will easily be drowned out by any kind of task you are doing. So, at least give us that 20% bump as a constant speed for crying out loud!!

Power Consumption

For this test we hooked up our power supply to a UPM power meter that will log the power consumption of the whole system twice every second. In order to stress the GPU as much as possible we once again use the Batch Render test in 3DMark06 and let it run for 1 hour to determine the peak power consumption while letting the card sit at a stable Windows desktop for 30 minutes to determine the peak idle power consumption.

Please note that after extensive testing, we have found that simply plugging in a power meter to a wall outlet or UPS will NOT give you accurate power consumption numbers due to slight changes in the input voltage. Thus we use a Tripp-Lite 1800W line conditioner between the 120V outlet and the UPM power meter.


Power consumption is actually quite good with this card even though it is about 10% higher than the 8800GT. That being said, the performance per watt based on our game tests is extremely good and unlike the HD4870, ATI’s PowerPlay actually seems to be working this time around.