Remember that benchmarks are arbitrary. They measure certain types of performance that may or may not reflect the way you work.
The computer was tested on 2000.10.08 under System 7.5.5 using both an external 160 MB hard drive and the same internal 80 MB drive used with Speedometer 4.02. Results are relative to an 8 MHz Mac Classic, which scores 1.0.
cache CPU graphics disk math 32 KB 4.25 3.71 2.44 6.70 64 KB 4.24 3.71 2.43 6.63 128 KB 4.25 3.69 2.45 6.65 256 KB 4.25 3.71 2.44 6.70
Disk tests for the internal drive were slightly lower at 2.15 (32 KB), 2.25 (64 KB), 2.25 (128 KB), and 2.14 (256 KB).
The cache setting should have little influence on non-disk tests, which these numbers bear out. Disk tests are close as well.
The system was tested on 1999.06.17 and 2000.11.21 under System 7.5.5. Results are relative to a Quadra 605, which rates 1.0. Numbers are rounded off to two decimal places.
These numbers compare performance at different cache settings.
cache CPU graphics disk math 32 KB 0.26 0.16 0.77 0.96 64 KB 0.26 0.16 0.70 0.98 128 KB 0.26 0.16 0.77 0.97 256 KB 0.26 0.16 0.76 0.97 ST2.1S 0.26 0.16 1.46 0.97
The fifth test scores are with our benchmark drive, an external Quantum ST2.1S with System 7.5.5 installed and extensions off, which scores nearly double the internal drive in the disk test. This test demonstrates that the Quantum LP80S inside the SE/30 falls well short of this computer’s SCSI potential.
The cache setting should have little influence on non-disk tests, which these numbers bear out. As above, with this particular setup, cache size makes a small amount of difference – the drop is only about 10% between the highest and lowest disk scores.
Compared with the IIcx, which has the same CPU and FPU, the SE/30 offers virtually identical performance except for graphics. There, because the SE/30 uses RAM for video while the IIcx has a separate video card, the IIcx benchmarks roughly 50% faster (0.25 vs. 0.16).
Hard Drive and Memory Speed
The newest addition to our benchmark suite is TimeDrive 1.3 (available here), which measures drive throughput. This can test a floppy, Zip, hard drive, or RAM Disk. TimeDrive is fairly primitive; the benefit of that is being able to run it on very old Macs.
The Quantum LP80S is an older, slower SCSI hard drive that was common when the SE/30 was on the market. We also tested our default external drive, a Quantum Fireball ST2.1S. Unfortunately, we were unable to run our ramBunctious RAM Disk benchmark, since the SE/30 bombed every time we tried to run it. Numbers are in KB/sec.
drive write read Quantum LP80S 1,180K 1,194K Quantum ST2.1S 1,497K 1,850K
Tests with other Macs show the ST2.1S runs faster then 4 MBps, so the 1,497K write and 1,850K read ceilings seems to be a limit on the SE/30’s SCSI bus. The write performance is 28% better than the Classic II, the less expensive model that replaced the SE/30. Read performance is an impressive 35% higher than on the cheaper Classic II.
These tests also show why the SE/30 was a popular network server. Even with the old Quantum LP80S, with a throughput of approximately 9.25 Mbps (average of read and write times 8), it could move data to and from the drive as fast as it could move date over an ethernet network. With any faster drive, the 10 Mbps network becomes the bottleneck.