CB>Since there are so few 386SXs left, you could aim for a generic 386, and
CB>the cached 386 would just run it faster, because everybody KNOWS a cache
CB>always makes a program run faster.  Wrong.  There are many cases where
CB>cache thrashing can cause a cached computer to run slower than a
CB>non-cached system.  And a good solution for a non-cached system can very

CB>Once you get beyond general algorithmic improvements, there are very
CB>very few optimizations that are general across multiple platforms and
CB>CPUs.  Every optimization of any sort has a price, and that price will
CB>be paid by somebody else running the program on a platform different
CB>from yours.
ahahahahahahahahahaha
I don't think anyones stupid enough to release something non-optimized 
because it reacts differently on different machines.
I fail to see why not to include several versions of the optimized executable 
specifically compiled for a machine..
This pretty much sums up most of what you said, you must either think
1) i am some kind of stupid idiot, who doesn't know what cache is
2) that this is an opportunity to show off what you believe to be true.
I am very aware of the effects of chache-thrashing and that different 
platforms execute code differently than others, but please don't claim that 
optimizing makes sloppy code because it reacts differently on platforms.
... Computers all wait at the same speed!
--- Ezycom V1.48g0 01fd016b
---------------