About the Benchmarks
These benchmarks are not scientific. I didn't try and clone the drives, or assure that they all had the exact same services and apps running during testing. However, I did attempt to make the runtime scenarios as similar as possible. The one "apples-to-oranges" issue would be that the Phison was formatted FAT32 and the RunCore was formatted NTFS. I used FAT32 on the Phison because FAT32 has a smaller block size and does not use journaling or the other more advanced write-intensive features of NTFS. On a systems with slow write speeds, this seemed to make sense. Both machines contained the following:
-Window XP Professional SP3
-All Windows Updates
-BIOS 1808 (latest)
-"Super Performance" mode a.k.a. full CPU speed
-Latest device drivers
These are the common tasks I chose to look at. Number 3 is a cold startup timed from button press to right after the sound drivers load, but before the network (wireless) drivers load. Number 5 is copying and pasting a 98Mb zip file.
I used PassMark PerformanceTest 6.1 on both drives to evaluate disk speed. All of the other tests in the benchmarking suite, i.e. 2D, 3D, RAM, CPU, etc., were statisically insignficant between the 2 drives. As you can see, the Disk tests yielded far different results.
The results speak for themselves, I think. The bottom line is that the RunCore SSD destroyed the stock Phison disk in every measurable way. Some of the results are so extreme, they're practically laughable. In short, adding a RunCore SSD to your EeePC 901 will turn it into the machine you wished it was out of the box. I'll go so far as to say that it makes the 901 useful for more than just web browsing, which in the realm of today's netbooks is really saying something. RunCore SSDs are available from MyDigitalDiscount.
Not sure what Part 2 of this article will be. Could be Phison NTFS vs. FAT32. Could be Phison SLC vs. RunCore. Ether way, Part 2 will be coming soon. If you have strong feelings about what you'd like to see given what you've just read, feel free to comment.