andrew@frip.gwd.tek.com (Andrew Klossner) (02/06/88)
>> Basically, I agree that a 68000 probably cannot emulate an 8086 >> without some sort of hardware assistance ... > Wanna bet? I have in front of me a beast named 'PC-ditto' for my Atari > ST.... runs most IBM PC software about as fast as the original 4.77 > MHz (?) PC did ... The original 4.77MHz PC was based on an 8088, not an 8086, and the note to which you responded quoted an emulator that is faster than the original PC in the very next paragraph. -=- Andrew Klossner (decvax!tektronix!tekecs!andrew) [UUCP] (andrew%tekecs.tek.com@relay.cs.net) [ARPA]
finkel@TAURUS.BITNET (02/09/88)
> >> Basically, I agree that a 68000 probably cannot emulate an 8086 > >> without some sort of hardware assistance ... > > > Wanna bet? I have in front of me a beast named 'PC-ditto' for my Atari > > ST.... runs most IBM PC software about as fast as the original 4.77 > > MHz (?) PC did ... > > The original 4.77MHz PC was based on an 8088, not an 8086, and the note > to which you responded quoted an emulator that is faster than the > original PC in the very next paragraph. > > -=- Andrew Klossner (decvax!tektronix!tekecs!andrew) [UUCP] > (andrew%tekecs.tek.com@relay.cs.net) [ARPA] Eeexxxcccuuussseee mmmeee!!! According to what I know, PC-ditto runs 5 to 20 times slower than a normal PC. From my experience with a similar program for the Amiga called The Amiga Transformer, which is also about 5 times slower, it is possible to run programs like kermit 2.29, which is IO bounded, and doesn't have a lot of computations, but is rather limited by the speed the user can type. The program can ahndle 1200 baud easily. realtime 8088 simulation is available on Unix with the Soft PC product. I have also read about a translation program from MS-DOS to Unix. It's a very intelegent program which analyses the program flow and decides which flags should be computed for each instruction, thereby gaining more speed. the translator also detects all the instruction that write to IO areas, and can replace such operations with the corresponding system calls. I forgot the name of the program, but I cann look it up. it's available for 68020 Unix machines, and for SPARC ( if I'm not wrong ). Udi Finkelstein finkel@math.tau.ac.il or finkel@taurus.BITNET ( which is the same ).