Path: utzoo!utgpu!water!watmath!clyde!att!osu-cis!tut.cis.ohio-state.edu!rutgers!apple!bionet!agate!ucbvax!TAURUS.BITNET!finkel From: finkel@TAURUS.BITNET Newsgroups: comp.sys.amiga Subject: Re: Amiga Emulators Message-ID: <8810131004.AA13250@MATH.Tau.Ac.IL> Date: 13 Oct 88 10:04:27 GMT Sender: usenet@ucbvax.BERKELEY.EDU Reply-To: Lines: 48 Newsgroups: comp.sys.amiga Subject: Re: Amiga Emulators Summary: Expires: References: <7321@well.UUCP> <189@kesmai.COM Sender: Reply-To: finkel@virgo.UUCP (Udi Finkelstein) Followup-To: Distribution: Organization: Tel-Aviv Univesity Math and CS school, Israel Keywords: In article <189@kesmai.COM> dca@kesmai.BITNET writes: >In article rg20+@andrew.cmu.edu (Rick Franc >>I've wondered >>why doesn't some one make a program that, instead of trying to emulate all of >>the hardware, doesn't take the software and create 68K machine code, treating >>the software as "source" code of a high level language (in this case one which >>has all the attributes of say an IBM or a Mac...). [ ... ] > >Actually, there is a company out there that does do an IBM emulator by >actually 'compilesing' 8086 code into 68020 code. As it only >generated 68020 code they were supposed to be targeting getting marketing >arrangements with people like Sun but I don't know how well the product >works or if they have had any success. It's been a while since I saw the >article so I don't remember much more about it. > >David Albrecht The product is available for the 68020/Clipper/88000 processors, and it runs on SUN3's, Sun4's, MacII, and on of the new Tektronix computer series. The program accepts a standsrd IBM/PC binary file, and produces a Unix executable file. One benchmark was Drhystones that ran at 331 Drhy/Sec on an XT, and around 2000 on a Sun/3. I forgot the product's name, but it's sold by Hunter systems. I can dig up the details if anyone is interested, just mail me. If I'm not wrong, it's called XDOS, or something smiliar. It uses advanced compiler techniques to map code/data sections, to decide which flags should be computed for each instruction ( it won't computer Half carry for all of them!), etc. ------------------------------------------------------------------------------- Udi Finkelstein | Bitnet: finkel@taurus.bitnet or finkel@math.tau.ac.il Tel Aviv University | Internet: finkel%taurus.bitnet@cunyvm.cuny.edu Israel | UUCP: ...!psuvax1!taurus.bitnet!finkel -------------------------------------------------------------------------------