UAE CPU module

[pixie]

I've always associated faster programs as of being those compiled for 040/060 for the obvious reason of being made on purpose to take advantage of those specificities present on those line of processors.

My question is, in UAE what would be faster, the same program running the 020 version using the 020 core or the 040 compiled version for the 040 core?

[Karlos]

In theory it makes no difference at all. Despite this, I have found 040+ executables perform a tiny fraction better on the 040 cpu emulation than 020 stuff does. However it's largely academic. Under JIT on a decent x86, everything is so insanely fast that there are more 'mips' in the random difference between two successive cpu tests than your average real 680x0 can manage flat out ;-)

[swift240]

Yep I am runnig AmiKit V1.2.0 and using SYSInfo I read 764 MIPS, on 040 with 8 meg of Chip ram.

Using an 020 2 meg Chip ram configuration I read using SYSInfo 754 MIPS.

My PC is an Athlon 2000+XP Sempron at 1.667 Ghz with 512 DDR ram.
I did use a program that tells you what the Mhz is but for the life of me I cant remember what it was. But I do remember I got 522Mhz out of it using an 040 configuration.

Is this any thing to go by.

Mike.

[Piru]

Some observations:

1. MIPS doesn't tell anything of the real performance. "MHz" of the emulated CPU tells even less.

2. Synthetic benchmarks are useless under emulation.

3. JIT will optimize away most "timing loops".

For any sensible real world results:

A) Use real life tasks, such as compression or decompression of large dataset.

B) Time the operation from outside of the emulation.

C) Run the benchmark multiple times. Take average of 3 best results.