I was able to run some tests on my 486DLC with and without cache enabled. Bus set at 8 MHz, and RAM set at 3 WS. Slow, but safe settings. Resuts before and after as follows.
Cyrix 486DLC L1 disabled ---> L1 enabled
Landmark
ALU, 86 ---> 129 (50% improvement)
FPU, 174 --->185 (6% improvement)
Speedsys
9.64 ---> 10.52 (9% improvement)
3Dbench
15.6 ---> 17.5 (12% improvement)
I suspect that the L1 cache isn't being utilised to its full ability. When I plugged in a Ti486SXL w/8KB of cache, these scores stayed the same. I probably need one of those cache enabling Cyrix interposer boards.
By the way, the autoexec.bat command I use with the program I attached is:
C:\cyrix.exe -e -i1 -f -r
I should probably also run some AMD 386-40 tests to see where the scores come in at.... Edit, and I have done just that. Here are the scores for an AMD 386DX-40 in the same computer.
AMD 386DX-40 ---> Cyrix 486DLC-40 w/1KB L1 enabled
Landmark
ALU, 61 ---> 129 (111% improvement)
FPU, 140 ---> 185 (32% improvement)
Speedsys
6.84 ---> 10.52 (54% improvement)
3Dbench
13.5 ---> 17.5 (30% improvement)
There is an average improvement of 65% when using a Cyrix 486DLC-40 compared to an AMD 386DX-40. If you were getting a Speedsys score of 9.31 with an AMD 386DX-40, imagine what score you'll get with a 486DLC (~14.3)! Expect the 3Dbench score to improve another 20% if using a graphics card better than a Trident, i.e. a Mach64, Cirrus, or Tseng.
@Keropi
Can you upload a screenshot of your Speedsys which indicated a score of 9.31 on an AMD 386DX-40? I've never seen this high of a score on an AMD 386. Be sure to also have a 40 MHz FPU installed as this will alter the score.
Plan your life wisely, you'll be dead before you know it.