Gavin.Flower@comp.vuw.ac.nz (Gavin Flower) (05/17/91)
On behalf of The Master of the Arcane, I proudly present... ----------------- Subject: Removing MEMCfaster from FishDemo... > but the FishDemo is all horrible and encoded, so I can't poke the SWI to >a SWINV. Sorry about that... Before I sent a copy on to England on behalf of the Authors, I SQUEEZEd the program, which made it significantly smaller. However, this means that the resulting !RunImage appears to be 'encoded'. It's relatively easy to unsqueeze such programs- Just find the code that is called on entry, and make it stop running when it has finished decompressing the original code into memory... However, I'm not sure if removing the MEMCFasters will help you much, as the demo gets nasty and jerky on Memc 1's (i.e. 310's), and will probably do the same without the Memcfaster... Hope this puts you out of your misery... The Master of the Arcane. ************************************************************************ ************************************************************************ ********************** Lets hope he can get back his own access soon. Thats what we ALL want -YES! <-: < -- The main "user" of well brought up, and educated, children is the community at large. So if you really believe in "user pays", charge the correct users - stop overloading parents with financial penalties. ******* These comments have no known correlation with dept. policy! *******
csuwr@warwick.ac.uk (Derek Hunter) (05/18/91)
In article <1991May17.100004.15705@comp.vuw.ac.nz> gavin@comp.vuw.ac.nz (Gavin Flower) writes: > >On behalf of The Master of the Arcane, I proudly present... > >----------------- >Subject: Removing MEMCfaster from FishDemo... [stuff] >However, I'm not sure if removing the MEMCFasters will help you much, as the >demo gets nasty and jerky on Memc 1's (i.e. 310's), and will probably do the >same without the Memcfaster... Yeah, but the whole point is that the later Arcs (A3000, A410 etc) don't need the MEMC update because they use MEMC1a s rather than MEMC1 s, and the 1a s don't have the kludge that slowed the 1s down. On my machine in particular, it objects to the `speedup' by crashing. I did try to follow the unsqueeze, but I got confused, and gave up. I'll be writing a patch to intercept SWIV and mask out just one MEMC 64,64 call. This wo'n't slow the machine down if it dissociates itself after that one call. I guess I can't just make the SWI a SWINV because that won't be supported in future releases [snigger]. - I don't know, he just phoned up and said something about the goat. - Derek Hunter.