To ensure that all reviews on Overclock3D are fair, consistent and unbiased, a standard set of hardware and software is used whenever possible during the comparative testing of two or more products. The configuration used in this review can be seen below:
CPU: Intel i7-870 @ 4GHz
Motherboard: MSI P55-GD80
Memory: 2x2GB GSkill Trident @2000MHz 9-9-9-24
HD : Hitachi Deskstar 7k160 7200rpm 160GB
GPU: XFX Radeon HD5850
Graphics Drivers: Supplied by XFX
PSU: Gigabyte ODIN 1200w
As you can see, I decided to run the setup at 4GHz to prevent any possibility of CPU bottlenecks. The 2000MHz bandwidth afforded by the GSKill Trident should also ensure that there are no hold ups in our test setup, allowing the cards to stretch there silicon legs.
During the testing of the setups above, special care was taken to ensure that the BIOS settings used matched whenever possible. A fresh install of Windows Vista was also used before the benchmarking began, with a full defrag of the hard drive once all the drivers and software were installed, preventing any possible performance issues due to leftover drivers from the previous motherboard installations. For the 3DMark and gaming tests a single card configuration was used.
To guarantee a broad range of results, the following benchmark utilities were used:
3D / Rendering Benchmarks
• 3DMark 05
• 3DMark 06
• 3DMark Vantage
• Far Cry 2
• Race drive: GRID
• Call of Duty IV
• Unreal Tournament III
Power consumption was measured at the socket using a plug-in mains power and energy monitor. Because of this, the readings below are of the total system, not just the GPU. Idle readings were taken after 5 minutes in Windows. Load readings were taken during a run of Furmark.
The power consumption of the smaller HD5850 was excellent, consuming just 149w when idle and 259 when placed under full load (system draw at the socket).
Temperatures were taken at the factory clocked speed during idle in Windows and after 30 minutes of running Furmark with settings maxed out (2560x1600 8xMSAA). Ambient temperatures were taken with a household thermometer. As we use an open test bench setup consideration should be given to the fact that the temperatures would likely increase further in a closed case environment.
Temperatures were pretty much in line with the HD5870. Because of the lower clockspeed of the HD5850, the temperatures were slightly lower than the HD5870. Throughout the testing the fan on the redesigned, full cover shroud rarely span up at all, testament to the excellent cooling properties of the heatsink and fan assembly hitting just 30c at idle and 67c on full load after 15 minutes of Furmark (2560x1600 8xMSAA).
For our overclocking tests I used the Catalyst Control Centre utility which worked perfectly with our setup. I was however once more let down by the very conservative overclocks the overdrive utility allowed, reaching just 775MHz on the core and 1125 on the memory with both respective sliders maxed out.
Not happy with that, I decided to push things a little higher and used AMD's new GPU overclock utility which allowed much better results.
The card would actually clock higher but after an hour of looping 3DMark 06 I started to encounter some glitching and tearing so lowered the clocks a little which gave rock solid stability. With scorching overclocks such as these one can only wonder why the white coats at ATI do not allow higher overclocking with the Overdrive utility. Were I being cynical I would suggest they don't want you to overclock the card past the more expensive HD5870 clockspeeds, thereby saving you a wad of cash! Either way, whatever your perspective, the XFX HD5850 can certainly take the overclocks and offer the end user much more over stock form so expect to see some 'overclocked' editions hitting the shelves in the near future.
Here's how COD 4 looked with overclocked vs stock settings:
As you can see, it's well worthwhile overclocking the HD5850 with some quite dramatic gains to be had, even at the highest resolutions. As always, the usual disclaimer applies when overclocking any card: You may not get the same results yourself or you may even be lucky enough to get higher overclocks. Each card is likely to overclock slightly differently from the next but the above should give you a good indication of what most XFX HD5850's are capable of.
After returning the card back to it's default speed, I ran our standard set of GPU benchmarks and compared it against the HD5870 and GTX285 2G...