I frequently run a set of my custom autolisp routines that, at times, take quite a while -- up to 3 hours. I consider them to be quite lean thanks to lots of help from this forum.
My machine is an Athalon 900mhz, with 1.25 gbytes RAM on Win2k and I use Acad 2004. The routines are not compiled, since (believe it or not) they run faster as is.
My question has to do with today's hardware. I note there are systems out there that boast 3.4 ghz processors and fast busses (I think mine is 1.33 mhz).
Any guesses as to what kind of performance improvement I could expect by upgrading hardware?
How about converting the code to C? I've taken a few C++ classes a few years ago so consider myself no more than a novice. My routines are some 175,000 bytes long and I've only glanced at Sharp C. If they could be successfully converted would there be a substantial savings in performance?
Len Miller
My machine is an Athalon 900mhz, with 1.25 gbytes RAM on Win2k and I use Acad 2004. The routines are not compiled, since (believe it or not) they run faster as is.
My question has to do with today's hardware. I note there are systems out there that boast 3.4 ghz processors and fast busses (I think mine is 1.33 mhz).
Any guesses as to what kind of performance improvement I could expect by upgrading hardware?
How about converting the code to C? I've taken a few C++ classes a few years ago so consider myself no more than a novice. My routines are some 175,000 bytes long and I've only glanced at Sharp C. If they could be successfully converted would there be a substantial savings in performance?
Len Miller
