Tech-Hounds.com

Performance

To measure real world performance, we choose to measure the time it takes for a game to load a level. Although measuring save game performance is valuable from this article standpoint, it's very difficult to do and in games like Quake 4, saving a game involves more than just the hard drive because the system have to make a thumbnail shot of your current view. We choose three games for this test - F.E.A.R, Quake 4 and Serious Sam II. Because of the way the game works (caching some elements in RAM and virtual memory), we ran test for both first time load and reloads.

We chose both F.E.A.R's and Quake 4's opening cinematic simply because these levels are generally much larger than the average levels for both games. For Serious Sam II, we choose the final Mental Institution level for the same reason. Measurements are taken with FRAPS, from the time the game starts loading the level to the very beginning of the cinematic. Since we're measuring load times, the measurements are not frame per second, but load time (in msec).

We'd like to thank Tagan for supplying the additional power supply for this article.

Our test setup
AMD Athlon 64 3500+ socket 939
2 x 1024 MB Kingston KVR 3-3-3 PC3200 DDR-SDRAM
MSI K8N NForce 4 SLI motherboard
Leadtek WinFast GeForce PX7900GS 256 MB
Maxtor DiamondMaxPlus9 80 GBs Serial ATA 8 MB buffer
ASUS E-616 DVD-ROM
Tagan TG530-U15 530 watts ATX/BTX power supply

Windows XP Professional with Service Pack 2 installed
NVIDIA Forceware 84.21 reference driver
NVIDIA NForce 4 6.66 reference driver
Creative SoundBlaster Live! 24 bit 5.12.1.512 driver.
DirectX 9.0c

The graphs are pretty self explanatory, but in case you can't see the text, they're arranged in groups of three colors: green for first test run,  blue for the second test run and red for the third test run. Results in black are the average of three runs. The purpose of including all three runs is to see how much variability these test have.  Results are in milliseconds, for example 25821 really means 25.821 seconds. 

The results:

No doubt the results that's more representative to real world performance are the first run results. The game spent most of its time fetching data for the first time from the hard drive. There are some write operations involved, since it's pretty likely not all the data is written to RAM but instead stored in a cache on the hard drive. The higher read transfer rates from the Hitachi Deskstar 7K160 is paying dividends here. Using a stripe array gives us a shorter load time, though not by much (4 seconds).

Consecutive loads pretty much confirms the trend we see with first run loads. Though there are more variations between runs, its obvious the Hitachi Deskstar 7K160 is still much faster than our Maxtor DiamondMax 10, both as a single drive and in a RAID 0 array. With the Hitachi Deskstar 7K160, we basically spent no more than 20 seconds, while with the DiamondMax 10 we spent more than 20 seconds.

[Previous Page]
[Go to top]
[Next Page]