jk@utastro.UUCP (John Krist) (12/28/85)
I have been informed by someone that the IBM 32 bit machine does not use a 80386. But I was assured by the person I talked to from IBM that the computer will run PC software. All I can guess is that maybe it uses a 68020 running an emulation program (probably a good deal faster than an PC or XT). What's going on? John Krist U. Texas Astronomy Dept. jk@utastro.UUCP
tim@ism780c.UUCP (Tim Smith) (01/04/86)
In article <216@utastro.UUCP> jk@utastro.UUCP (John Krist) writes: > > I have been informed by someone that the IBM 32 bit machine does not > use a 80386. But I was assured by the person I talked to from IBM that > the computer will run PC software. All I can guess is that maybe it > uses a 68020 running an emulation program (probably a good deal faster > than an PC or XT). What's going on? > Perhaps a 68000 or 68020 with custom micro-code? -- Tim Smith sdcrdcf!ism780c!tim || ima!ism780!tim || ihnp4!cithep!tim
doon@sdcrdcf.UUCP (Harry W. Reed) (01/04/86)
In article <210@ism780c.UUCP> tim@ism780c.UUCP (Tim Smith) writes: >In article <216@utastro.UUCP> jk@utastro.UUCP (John Krist) writes: >> >> I have been informed by someone that the IBM 32 bit machine does not >> use a 80386. But I was assured by the person I talked to from IBM that >> the computer will run PC software. All I can guess is that maybe it >> uses a 68020 running an emulation program (probably a good deal faster >> than an PC or XT). What's going on? >> > >Perhaps a 68000 or 68020 with custom micro-code? Gross! Yuk, Phooey, bleck ...
baba@spar.UUCP (Baba ROM DOS) (01/05/86)
> In article <216@utastro.UUCP> jk@utastro.UUCP (John Krist) writes: > > > > I have been informed by someone that the IBM 32 bit machine does not > > use a 80386. But I was assured by the person I talked to from IBM that > > the computer will run PC software. All I can guess is that maybe it > > uses a 68020 running an emulation program (probably a good deal faster > > than an PC or XT). What's going on? > > > > Perhaps a 68000 or 68020 with custom micro-code? > -- > Tim Smith sdcrdcf!ism780c!tim || ima!ism780!tim || ihnp4!cithep!tim More likely the machine in question is the much-rumored "PC/RT", which reportedly has a proprietary 32-bit RISC processor plus a 286 processor to handle the PC I/O. The I/O processor could presumably run PC software all by itself, though probably not concurrently with supporting the RISC. Baba ROM DOS
gnu@l5.uucp (John Gilmore) (01/05/86)
In article <216@utastro.UUCP> jk@utastro.UUCP (John Krist) writes: > I have been informed by someone that the IBM 32 bit machine does not > use a 80386. But I was assured by the person I talked to from IBM that > the computer will run PC software. > What's going on? The obvious answer is that it could have an 8088 (or 80286) and some other processor, probably the RISC chip that grew out of the IBM 801 research machine. Of course, IBM always tells *me* what they're doing... :-) Software emulation would work OK too, especially if the processor is fast enough. I suspect a 68020 can emulate an 8088 faster than the 8088 (though I haven't tried it). It would be harder to emulate all the code that "knows" where the graphics screen is and how big it is, how to diddle the serial ports, etc, unless actual PC clone hardware is provided. I think both the Amiga and some microcoded xeroX system emulate the 8088.
kds@intelca.UUCP (Ken Shoemaker) (01/06/86)
> Software emulation would work OK too, especially if the processor is fast > enough. I suspect a 68020 can emulate an 8088 faster than the 8088 (though > I haven't tried it). It would be harder to emulate all the code that "knows" To a point...I doubt that it could emulate the 8088, picking up all the nuances of the flags on arihmetic operations and still be faster, I mean, it would have to do nibble arithmetic! -- remember, if you do it yourself, sooner or later you'll need a bigger hammer Ken Shoemaker, Santa Clara, Ca. {pur-ee,hplabs,amd,scgvaxd,dual,qantel}!intelca!kds ---the above views are personal.
ralphw@ius2.cs.cmu.edu (Ralph Hyre) (01/06/86)
In article <2555@sdcrdcf.UUCP> doon@sdcrdcf.UUCP (Harry W. Reed) writes: >In article <210@ism780c.UUCP> tim@ism780c.UUCP (Tim Smith) writes: >>In article <216@utastro.UUCP> jk@utastro.UUCP (John Krist) writes: >>> >>> I have been informed by someone that the IBM 32 bit machine does not >>> use a 80386. But I was assured by the person I talked to from IBM that >>> the computer will run PC software. All I can guess is that maybe it >>> uses a 68020 running an emulation program (probably a good deal faster >>> than an PC or XT). What's going on? >>> >> >>Perhaps a 68000 or 68020 with custom micro-code? > > Gross! Yuk, Phooey, bleck ... Emulation actually might not be that bad, as long as you don't have to deal with wierd copy protection schemes and other foolishness. I've seen a published rumor (in an Apple ][-related publication) to the effect that the Apple //x will emulate the 6502 processor using a 680x0 processor. If they can get 95% of the //e software and hardware to work with the new setup then this will be a big win, since it will allow the Apple ][ and Mac product lines to start converging. - Ralph -- - Ralph W. Hyre, Jr. Internet: ralphw@c.cs.cmu.edu (cmu-cs-c.arpa) Usenet: ralphw@mit-eddie.uucp Fido: Ralph Hyre at Net 129, Node 0 (Pitt-Bull) Phone: (412)CMU-BUGS
davidsen@steinmetz.UUCP (Davidsen) (01/07/86)
In article <399@l5.uucp> gnu@l5.uucp (John Gilmore) writes: >In article <216@utastro.UUCP> jk@utastro.UUCP (John Krist) writes: >> I have been informed by someone that the IBM 32 bit machine does not >> use a 80386. But I was assured by the person I talked to from IBM that >> the computer will run PC software. >> What's going on? > >The obvious answer is that it could have an 8088 (or 80286) and some other >processor, probably the RISC chip that grew out of the IBM 801 research >machine. Of course, IBM always tells *me* what they're doing... :-) > >Software emulation would work OK too, especially if the processor is fast >enough. I suspect a 68020 can emulate an 8088 faster than the 8088 (though >I haven't tried it). Jan 86 Dr Dobb's Journal had an article on emulating an 8080 using a 68K, and even with a few compromises the emulation doesn't run very fast (usefully fast, but only about 1.4MHz effective speed). There are (at least) two things the 68K doesn't do well, namely decimal arithmetic and segmentation. The lack of the "half carry" flag in hardware requires a bunch of emulation, as does the action of the segmentation. Without getting into the issue of segmentation vs. linear addressing, the 8088 does segmentation *really well*, and a completely correct emulation will probably not run "faster than the 8088". Note that I, too, have not tried it, but I believe that there is an "add on" to the 8088 emulation on the Amiga (don't flame me if it's the Atari) machine. There's no doubt that the 68020 could run the 8088 code fast enough to be useful, and I have a great Z80 emulator in C which will run (eventually) on almost anything, perhaps I could run CP/M under the MS-DOS version of the emulator, under the 8088 emulator... then again I could just use a coprocessor board or get some up to date software. -- -billd seismo!rochester!steinmetz! unirot / \ / ihnp4! crdos1!davidsen \ / chinet! -----------------/ "It seemed like a good idea at the time..."
fnf@unisoft.UUCP (Fred Fish) (01/08/86)
In article <168@intelca.UUCP> kds@intelca.UUCP (Ken Shoemaker) writes: >> Software emulation would work OK too, especially if the processor is fast >> enough. I suspect a 68020 can emulate an 8088 faster than the 8088 (though > >To a point...I doubt that it could emulate the 8088, picking up all the nuances >of the flags on arihmetic operations and still be faster, I mean, it would >have to do nibble arithmetic! The software 8088 emulation mode on the AMIGA is reported to execute IBM PC programs at about 40% of the speed of a real 8088. Given that a 68020 is about 2-3 times faster than a plain 68000, I would expect the equivalent 68020 emulation to run at 80-120% of a real 8088 under the same conditions. Of course this is all just speculation... -Fred =========================================================================== Fred Fish UniSoft Systems Inc, 739 Allston Way, Berkeley, CA 94710 USA {ucbvax,dual}!unisoft!fnf (415) 644 1230 TWX 11 910 366-2145 =========================================================================== -- =========================================================================== Fred Fish UniSoft Systems Inc, 739 Allston Way, Berkeley, CA 94710 USA {ucbvax,dual}!unisoft!fnf (415) 644 1230 TWX 11 910 366-2145 ===========================================================================
rodney@gitpyr.UUCP (RODNEY RICKS) (01/08/86)
In article <1@unisoft.UUCP>, fnf@unisoft.UUCP (Fred Fish) writes: > The software 8088 emulation mode on the AMIGA is reported to execute > IBM PC programs at about 40% of the speed of a real 8088. Given that a > From what I have heard, the Amiga's IBM emulator runs at about 60 to 70% of the speed of an IBM PC. It seems likely to me that IBM's new machine will be the RISC machine mentioned earlier. I don't think they will have an 80386 based machine available soon. I am far from being a good source for IBM info, BUT, I have information that seems to indicate that 1) The PC II does NOT exist. 2) IBM is working on some kind of computer that is NOT based on the 80386 and is probably not 32 bit. Could this be IBM's next try at home computers? 3) IBM is not very big on hiring at this time.
farren@well.UUCP (Mike Farren) (01/08/86)
In article <1@unisoft.UUCP>, fnf@unisoft.UUCP (Fred Fish) writes: > > The software 8088 emulation mode on the AMIGA is reported to execute > IBM PC programs at about 40% of the speed of a real 8088. Given that a > 68020 is about 2-3 times faster than a plain 68000, I would expect > the equivalent 68020 emulation to run at 80-120% of a real 8088 under > the same conditions. Of course this is all just speculation... I've seen a beta version of the PC emulation. Sorry, it ain't 40%, more like 15%. S - L - O - W! The "quick" emulator, if I am not mistaken, will incorporate an 8088 processor, so no comparison there. Having written a number of emulators in my time, I've found that you are pretty lucky if you can get even 50% speed out of one. Hardware just does some things a lot more efficiently than software, and real-time emulation of another piece of hardware is definitely one of those things. -- Mike Farren uucp: {dual, hplabs}!well!farren Fido: Sci-Fido, Fidonode 125/84, (415)655-0667 USnail: 390 Alcatraz Ave., Oakland, CA 94618