First post, by chublord
From the 686 ultimate benchmark thread (The Ultimate 686 Benchmark Comparison, I noticed that the 586/686 class processors had literally 100x better performance on 3DMark99 compared to 486-class processors, even normalized for the same frequency. The P55c had a score about 250x as high as a DX4 or Am5x86! While obviously 586/686's are a lot faster than 486's, this seems like an excessive difference.
For reference, I tried 3DMark99 on my DX4-100 system and got a score of "12" 3D Marks and "2" CPU Marks. (My Quake1 Timedemo score is 13.6 fps)
Why is the performance so terrible on 3DMark99 on a 486-class machine?
IBM Valuepoint 486 DX4-100, Opti 802G, 50 MHz FSB, Voodoo1+S3 864, Quantum Fireball EX 4.0 GB, Seagate Medalist 1.6 GB, 128 MB FPM, 256k L2