rg20+@andrew.cmu.edu (Rick Francis Golembiewski) (10/06/88)
I've noticed that there seems to be a lot of talk of various emulators lately
(ie Should the new amigas have At & PS/2 slots, Is there going to be a
Mac Emulator etc.) Well in any case, I've noticed that in general emulators
are Kludges, sinc they are trying to copy hardware using software (this is
even true for bridge card to some extent, ie. the display etc.). 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...). Granted this is a lot of
work, but then so is an "interpreted" emulator (like our good friend the
transformer). Also, to simplify it could one create specific patches for
specific programs, I should think that given all the efor put into an
emulator package, and the fact that the current emulators either offer
unacceptably slow (all software, ie transformer or C64 emulators) or
more expensive then the machine to be emulated (sidecar/bridge card).
Any comments? (I'm sure that some people are waiting to flame me out of
existance, about how impossibly difficult this would be, but it's just
an idea, so save the energy...)
[Disclaimer: Me? Post that? Impossible I never post anything... ]
[CYouAllLater(Everyone); <--- a "functional" goodbye ]
[ Rick Golembiewski {Pronounciation is 1/2 the battle, spelling the other ]dnelson@umbio.MIAMI.EDU (Dru Nelson) (10/06/88)
... a whole bunch about how most emulators stink.
The Spectre 128 on the Atari ST runs mac software faster than the Mac
SE !! Dave Small, the mastermind of the Spectre and Magic Sac is
going to try and make a similar item for the Amiga! 2+2=5.
So, what was that you were saying about emulators?? ;-)
--
Dru Nelson UUCP: ....!uunet!gould!umbio!dnelson
Miami, Florida MCI: dnelson
Internet: dnelson%umbio@umigw.miami.edudaveh@cbmvax.UUCP (Dave Haynie) (10/06/88)
in article <gXGgcst88k-00=SJ5p@andrew.cmu.edu>, rg20+@andrew.cmu.edu (Rick Francis Golembiewski) says: > > 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...). The cases of IBM and Mac are radically different, for one. The "IBM Emulator" is a true case of an emulator, the PC bridge card software is a partial emulator. For a "Mac Emulator", you're really just implementing the Mac OS on a somewhat different 68000 based machine. There's not all that much to really emulate, some, but not much. Which is why you see slow IBM Emulators, while Mac Emulators on 68000 machines (like the Atari "Magic Sac") can actually run the Mac OS faster than similar Macs can run it. If emulating the CPU was all there is to it on an IBM emulator, you could write an assembly code translator to produce a 68000 version of the same program, and it wouldn't be all that difficult. The problem is that most IBM programs, as C64 programs, write directly to the machine's hardware, instead of going to the hardware through OS calls. So there's no way to implement those function on alien hardware other than by directly emulating that hardware in software. And that slows things down. The other problem is that these older systems also tend to use self-modifying code, which you can't translate automatically. Still, I find the idea of such translators more interesting than the emulators that are popular today. > [Disclaimer: Me? Post that? Impossible I never post anything... ] > [CYouAllLater(Everyone); <--- a "functional" goodbye ] > [ Rick Golembiewski {Pronounciation is 1/2 the battle, spelling the other ] -- Dave Haynie "The 32 Bit Guy" Commodore-Amiga "The Crew That Never Rests" {ihnp4|uunet|rutgers}!cbmvax!daveh PLINK: D-DAVE H BIX: hazy "I can't relax, 'cause I'm a Boinger!"
ewhac@well.UUCP (Leo 'Bols Ewhac' Schwab) (10/07/88)
In article <gXGgcst88k-00=SJ5p@andrew.cmu.edu> rg20+@andrew.cmu.edu (Rick Francis Golembiewski) writes: >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...). [ ... ] I assume he means a program that takes FOO object code (where FOO is a given CPU type) and generates equivalent 68K object code for later execution. The reason this isn't done is because it's impossible (if not impossible, certainly supremely difficult). The two biggest blocks are data segments (How does the interpiler know what's code and what's data? The program object code will no doubt grow bigger, so how do we move the data segments around to compensate, and modify the object code to find them?), and self-modifying code (replacing a 68K opcode with a FOO opcode is going to cause you problems *real* fast). The first case is unavoidable, unless the program still has its symbol table around. The second case crops up all the time in copy protection routines, and some high-performance applications (read: games). Add to all that the "reversed" byte order of words and longwords on the 68000, the differing flag behavior, and the fact that the target program will still be running on foreign hardware, and you have some seriously hairy bananas before you. _-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_ Leo L. Schwab -- The Guy in The Cape INET: well!ewhac@ucbvax.Berkeley.EDU \_ -_ Recumbent Bikes: UUCP: pacbell > !{well,unicom}!ewhac O----^o The Only Way To Fly. hplabs / (pronounced "AE-wack") "Work FOR? I don't work FOR anybody! I'm just having fun." -- The Doctor
ewhac@well.sUCP (Leo 'Bols Ewhac' Schwab) (10/07/88)
In article <gXGgcst88k-00=SJ5p@andrew.cmu.edu> rg20+@andrew.cmu.edu (Rick Francis Golembiewski) writes: >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...). [ ... ] I assume he means a program that takes FOO object code (where FOO is a given CPUexecution. The reason this isn't done is because it's impossible (if not impossible, certainly supremely difficult). The two biggest blocks are data segments (How does the interpiler know what's code and what's data? The program object code will no doubt grow bigger, so how do we move the data segments around to compensate, and modify the object code to find them?), and self-modifying code (replacing a 68K opcode with a FOO opcode is going to cause you problems *real* fast). The first case is unavoidable, unless the program still has its symbol table around. The second case crops up all the time in copy protection routines, and some high-performance applications (read: games). Add to all that the "reversed" byte order of words and longwords on the 68000, the differing flag behavior, and the fact that the target program will still be running on foreign hardware, and you have some seriously hairy bananas before you. _-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_-_ Leo L. Schwab -- The Guy in The Cape INET: well!ewhac@ucbvax.Berkeley.EDU \_ -_ Recumbent Bikes: UUCP: pacbell > !{well,unicom}!ewhac O----^o The Only Way To Fly. hplabs / (pronounced "AE-wack") "Work FOR? I don't work FOR anybody! I'm just having fun." -- The Doctor
rodmar@ga.ecn.purdue.edu (Rodney L Martin) (10/07/88)
It may be true that there is a mac emulator COMING out for the amiga, and that it is faster than an SE on the Atari ST. My question was why the IBM emulator on the ST is so much faster than on the Amiga. Isn't the Amiga a more powerful machine?
fnf@fishpond.UUCP (Fred Fish) (10/07/88)
In article <gXGgcst88k-00=SJ5p@andrew.cmu.edu> rg20+@andrew.cmu.edu (Rick Francis Golembiewski) writes: > 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...). Granted this is a lot of >work... There is, but sorry, I can't recall the name of the company at the moment. I did see a demo when I worked at Motorola, where they took a generic IBM PC program, ran it through their compiler, and then ran the resulting executable on a Sun-3. It ran faster on a Sun-3 than it did on a real IBM PC. -Fred -- # Fred Fish, 1346 West 10th Place, Tempe, AZ 85281, USA # noao!nud!fishpond!fnf (602) 921-1113
dlm@druhi.ATT.COM (Dan Moore) (10/07/88)
in article <716@umbio.MIAMI.EDU>, dnelson@umbio.MIAMI.EDU (Dru Nelson) says: > The Spectre 128 on the Atari ST runs mac software faster than the Mac > SE !! Dave Small, the mastermind of the Spectre and Magic Sac is > going to try and make a similar item for the Amiga! 2+2=5. > So, what was that you were saying about emulators?? ;-) Dave Small has on several occasions said he is *NOT* going to port the Magic Sac or the Spectre 128 to the Amiga. Dave did mention the possibility of porting his work to the Amiga but there was very little interest expressed in the product by Amiga owners. This convinced him not to pursue it any further. There is a good possibility that a Macintosh emulator will be released for the Amiga. The developer (who lives outside the US) has been talking to a couple of different publishers and last I heard (about 6 weeks ago) had made a deal with one of them. It should be available sometime between now and the end of the 1st quarter '89, assuming they are serious about releasing it (ie. they spend the time & money to get it ready to ship). Dan Moore AT&T Bell Labs Denver dlm@druhi.ATT.COM
daveh@cbmvax.UUCP (Dave Haynie) (10/07/88)
in article <389@ga.ecn.purdue.edu>, rodmar@ga.ecn.purdue.edu (Rodney L Martin) says: > > It may be true that there is a mac emulator COMING out for the amiga, and that > it is faster than an SE on the Atari ST. The difference being that the SE runs a slightly slower clock rate than the ST, and also, in typical Apple style, has more things that get in the way of the 68000 (though fewer than on a Mac Plus, I hear). You're really not EMULATING much of anything here; Macs run 68000 instructions, STs run 68000 instructions, Macs use a single bitmapped display, and that's one of the ST video modes, etc. You may have to emulate a Mac I/O chip or something, but since more if not all of the Mac applications go through the Mac's toolbox ROM, these "emulators" may just replace the proper routines with ST-native versions of the same. In any case, it's really just porting the Mac OS over to the ST. If Apple didn't watch their code like hawks, it wouldn't be much more difficult than porting MS-DOS to yet another clone machine. > My question was why the IBM emulator on the ST is so much faster than on the > Amiga. Isn't the Amiga a more powerful machine? Same reason the BASIC that comes with the Amiga is so much faster than the BASIC that comes with the ST. Or any number of other similar comparisons. The PC emulator that runs on the Amiga was a POORLY WRITTEN piece of software. Apparently the PC emulator that runs on the ST was better written. These things do happen. If this unnamed PC emulator for the ST were ported to the Amiga, it would give similar performance. -- Dave Haynie "The 32 Bit Guy" Commodore-Amiga "The Crew That Never Rests" {ihnp4|uunet|rutgers}!cbmvax!daveh PLINK: D-DAVE H BIX: hazy "I can't relax, 'cause I'm a Boinger!"
dca@kesmai.COM (David C. Albrecht) (10/08/88)
In article <gXGgcst88k-00=SJ5p@andrew.cmu.edu> rg20+@andrew.cmu.edu (Rick Francis Golembiewski) writes: >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
garrett@oscar.ccm.udel.edu (10/08/88)
I've been away from this group for a good while, and I'm sure this has come up in the past couple of weeks, but I'm interested in hearing what others think of the plausibility of this (?) rumour... There's an article in MacWeek, 10/4/88, page 1 mentioning a Mac emulator for the Amy by Small, Inc. Evidently this guy Small was the one that did did Magic Sac for the ST or at least had a hand in it. They say that the Amy version is "in the works" and I was wondering if anyone else had heard anything else about this product (or product-to-be as the case may be). I've got a lot of PC and Mac software that I have been working on/with for the last couple of years, but the Amy has always been an intriguing machine because of the NTSC graphics and sound and multitasking, etc. There's no way I could afford all three machines, but a 2000 (or the next one, if it ever comes into existance) with a 80x8x bridge board, and Spectre 128 would probably be just what I'm looking for -- or is it? Am I expecting too much here? What are the "real" costs in such a system? Does the extra RAM/Disk space needed to store each of the "3" machines' code/date negate the savings of using the same keyboard/peripherals/etc for all "3" machines? Thanks in advance, Joel +-------------------------------------+--------------------------------------+ | Joel J. Garrett, Research Associate | Phone: (302)-451-2332 | | Center for Composite Materials | inet: garrett@oscar.ccm.udel.edu | | University of Delaware +--------------------------------------+ | Newark, Delaware | Elvis needs boats! -- Mojo Nixon | +-------------------------------------+--------------------------------------+
pnelson@antares.UUCP (Phil Nelson) (10/08/88)
In article <389@ga.ecn.purdue.edu> rodmar@ga.ecn.purdue.edu.UUCP (Rodney L Martin) writes: >It may be true that there is a mac emulator COMING out for the amiga, and that >it is faster than an SE on the Atari ST. My question was why the IBM emulator >on the ST is so much faster than on the Amiga. Isn't the Amiga a more powerful >machine? Assuming you are talking about the Transformer, if it's slower, it's probably because it is not quite as efficiently written. Anyway, turning an Amiga into a PC clone (even temporarily) is not very useful, in my opinion. The Amiga is a multitasking machine when running under AmigaDOS, but not when you run the Transformer. I tried it, and I didn't like it. Probably if I had an Atari, I would not have noticed, since I would not be accustomed to running more than one program at once. Running *BM PC programs in a window while still running Amiga programs is useful, You can do this with the 2000 and a BridgeBoard. I have both, and am happy to say that I use the BridgeBoard often. With the BridgeBoard, I have compatibility without losing multi-tasking. For me, the BridgeBoard works where the Transformer did not. -- {ames|pyramid}oliveb!tymix!antares!pnelson | Parallel IQ (the IQ of a group) OnTyme: NSC.P/Nelson POTS: (408)922-7508 | may be easily calculated given Disclaimer: Not officially representing | the IQ of each member - use the McDonnell Douglas Corporation policy. | formula for parallel resistance.
ltf@killer.DALLAS.TX.US (Lance Franklin) (10/08/88)
In article <389@ga.ecn.purdue.edu> rodmar@ga.ecn.purdue.edu.UUCP (Rodney L Martin) writes: >It may be true that there is a mac emulator COMING out for the amiga, and that >it is faster than an SE on the Atari ST. My question was why the IBM emulator >on the ST is so much faster than on the Amiga. Isn't the Amiga a more powerful >machine? It's my understanding that the IBM emulator for the ST was, in fact, written by the same guy who did the Transformer for the Amiga...it was, however, done AFTER the Transformer, and hence, benefits from whatever the guy learned on the Amiga. The Transformer, unfortunately, has not been improved (except for a few patches to let it run under 1.2 and on the 2000) since it's introduction, whereas the IBM emulator on the ST has gone through several revisions since its initial introduction, improving in the process. In other words, it's the software, not the hardware. :-) Lance -- +-------------------------+ +-----------------------------------------------+ | Lance T Franklin | | I never said that! It must be some kind of a | | ltf@killer.DALLAS.TX.US | | forgery...I gotta change that password again. | +-------------------------+ +-----------------------------------------------+
mlelstv@faui44.informatik.uni-erlangen.de (Michael van Elst ) (10/10/88)
In article <389@ga.ecn.purdue.edu> rodmar@ga.ecn.purdue.edu.UUCP (Rodney L Martin) writes: >It may be true that there is a mac emulator COMING out for the amiga, and that >it is faster than an SE on the Atari ST. My question was why the IBM emulator >on the ST is so much faster than on the Amiga. Isn't the Amiga a more powerful >machine? Hmm, maybe it is possible to emulate a mac with an AMIGA. But unless you rewrite a big part of the mac kernel, you won't be faster. (Except for a few percent like the ST because of the slightly faster CPU). And even then only graphics will get faster because of the AMIGA blitter. The IBM emulator has its own problems. Because of the computing-intensive nature of emulating another CPU, the major point is the speed of the host CPU. You cannot make much use of a drawing processor or DMA channel to emulate foreign instructions. As the ST is for about 10 percent faster with its CPU, it would be 10 percent faster when using the same emulator code. Nevertheless, the ST MS-DOS-emulator has some better code running up to 50 percent faster under average conditions. The Amiga MS-DOS-emulator I think is faster in text output but (sigh!) has NO graphics emulation code. Michael van Elst E-mail: UUCP: ...uunet!unido!fauern!faui44!mlelstv
cmcmanis%pepper@Sun.COM (Chuck McManis) (10/11/88)
All "good" emulators compile the object code into native code before
running. If everyone wrote completely vanilla programs this would make
for very fast emulators, unfortunately not everyone does. Instead they
do things like self modifying code, and pushing data on the stack and
then doing a "return" (sort of a poor man's computed branch). Consequently,
emulators have to constantly test for these cases and work around them.
The other major problem with emulating different architectures are the
various ways in which internal pieces of the architecture are handled.
In the case of the Intel and x80 architectures one of the big problems
is the flags, especially the half carry flag. Because there is no real
equivalent of half carry on the 68K, to emulate it requires several
computations on the result register. (generally two masks and an XOR
operation). Many compiler techniques can help, in particular things
like register/flag following. Register/flag following is a technique
that "knows" all routines that affect the flags, when a "flag instruction"
is encountered you can look ahead to see if another instruction is
going to change that flag before it is used and if so, *not* compute
the flag value. Another problem which is less severe is people who use
undocumented opcodes. These will cause you compatibility nightmares too.
And of course last but not least, when you say "emulate and 8086" do
you really mean "emulate an IBM-PC?" because in that case you have to
emulate such things as floppy controllers, DMA channels, and interrupt
controllers. None of which can be "compiled" per se.
Finally, if you do this, and it is reccomended because it can really
broaden your "systems" view of the world. You will find you have a
tool that is not only capable of running programs written for a
different machine, but also a fantastic simulator/debugger for
programs and hardware for that machine.
--Chuck McManis
uucp: {anywhere}!sun!cmcmanis BIX: cmcmanis ARPAnet: cmcmanis@sun.com
These opinions are my own and no one elses, but you knew that didn't you.wtm@neoucom.UUCP (Bill Mayhew) (10/13/88)
I believe (one of) the IBM emulator(s) for the Atari ST is called "PC Transporter". I think there might be more than one IBM emulator packages. The guy that lives across the street from me has an ST, and I've seen him using PC transporter. It really is much better than Transformer for the Amiga, the which I use. I had trouble running the PC Magazine PC Labs Benchmark disk test under Transformer, while PC Transporter breezed through it. One thing that makes life a bit simpler on the Atari side of life is that the disks are 720K just like IBM, even when the Atari is running TOS. In fact many TOS system calls are exact analogs of MSDOS INT21 calls. Only minimal work is needed to convert from DOS to TOS. Trapping the direct writes to IBM-specific chips that aren't in the Atari is the hard part. Actually, the magic sac Macintosh emulator would have been tougher to do, since the Atari disks do not have electronics to write in GCR format. I haven't seen magic sac, so I don't know if it can read mac disks .. it seems like it would be fairly hard to do, given the hardware on the ST. In some ways the Mac emulator would be easier, as the instructions are native 68K and don't have to be run as p-code. Also the mac display isn't all that different from the ST's monochrome bitmapped display. I do believe that if somebody really wanted to get serious about writing a PC emualtor for the Amiga that a pretty decent job could be done. Transformer doens't really seem like it ever got real far beyond a beta release. It would be neat to be able to have a way to multitask Transformer under Amigados. It seems like it is plausible that one could provie multiple 640K MSDOS windows. Even Windows 386 can't do that, as it has a 390K root segment that is always resident in the lower 640K of memory; that sort of crimps the style with things like 1-2-3. --Bill
finkel@TAURUS.BITNET (10/13/88)
Newsgroups: comp.sys.amiga Subject: Re: Amiga Emulators Summary: Expires: References: <gXGgcst88k-00=SJ5p@andrew.cmu.edu> <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 <gXGgcst88k-00=SJ5p@andrew.cmu.edu> 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 -------------------------------------------------------------------------------
ifarqhar@mqccsunc.mqcc.mq.OZ (Ian Farquhar 7420) (11/25/89)
In article <4666@lab.udel.EDU> CCUCARD%indsvax1.bitnet@uicvm.uic.edu (Time Lord) writes: > >Isn't it funny how people say," Amiga is just a toy." Then I ask, >if Amiga is just a toy, then why can't IBM multitask using 1meg with several >processes running? Why can't they emulate the Amiga??? Amiga can >emulate :IBM,MAC,UNIX,and C64. I think the real toy makers are: and so on.... The Amiga emulates the Mac, PC and the CBM64. It runs UNIX. It also runs Minix (real soon now), and I believe there is an OS/9 implimentation, and a BBC and Apple II emulator somewhere out there. As it can run PC-DOS software, it will also run Xenix and (given a hell of a lot of money and hardware), OS/2. Actually, the idea of a PC emulating an Amiga is quite an interesting idea, but not as interesting as the idea of a Mac or a ST emulating one. When I worked in retailing (no flames please, I have repented) we used to have a suppiler who was a real hacker, and just loved the ST. Before they put him away (:-), he mentioned that someone had developed an Amiga emulator for the ST. This is not impossible, assuming: 1. That all hardware dependent libraries (eg. the graphics.library, and so on) are rewritten or heavily patched (not impossible with the so-so-convenient and oh-so-dangerous SetFunction), and that all devices are rewritten. This is a difficult task, but certainly not impossible, particularly as you are reverse engineering an existing product rather than designing one from scratch (in my experience, design is about 40% of the work). 2. You were willing to put up with incredibly slow graphics performance (particularly when writing to partially obscured windows). An emulator would also need hardware (especially a disk drive), unless the current drive could be coaxed into reading Amiga format disks. 3. The final assumption is that CBM would probably sue you out of existance. However, it is not TECHNICALLY impossible to produce an Amiga emulator, the difficulty is one of motive, and of legal considerations. D
cmcmanis%pepper@Sun.COM (Chuck McManis) (11/28/89)
In article <21@macuni.mqcc.mq.oz> (Ian Farquhar 7420) writes: >However, it is not TECHNICALLY impossible to produce an Amiga emulator, >the difficulty is one of motive, and of legal considerations. It is technically impossible to emulate the Amiga on an Atari ST. The Amiga has a larger color map, stereo digital sound (4 channels), and a very powerful Copper that runs in parallel to the CPU. None of these exist on the current ST (although some of them are on the ST-E). Yes, I've seen Spectrum512 and it's variants but 512 != 4096 no matter how much you squint. Sorry, end of discussion. The Amiga can't emulate a VGA card either, so what? --Chuck McManis uucp: {anywhere}!sun!cmcmanis BIX: cmcmanis ARPAnet: cmcmanis@Eng.Sun.COM These opinions are my own and no one elses, but you knew that didn't you. "If it didn't have bones in it, it wouldn't be crunchy now would it?!"
jtreworgy@eagle.wesleyan.edu (11/28/89)
In article <128445@sun.Eng.Sun.COM>, cmcmanis%pepper@Sun.COM (Chuck McManis) writes: > In article <21@macuni.mqcc.mq.oz> (Ian Farquhar 7420) writes: >>However, it is not TECHNICALLY impossible to produce an Amiga emulator, >>the difficulty is one of motive, and of legal considerations. > > It is technically impossible to emulate the Amiga on an Atari ST. The Amiga > has a larger color map, stereo digital sound (4 channels), and a very powerful > Copper that runs in parallel to the CPU. None of these exist on the current > ST (although some of them are on the ST-E). Yes, I've seen Spectrum512 and > it's variants but 512 != 4096 no matter how much you squint. Sorry, end of > discussion. The Amiga can't emulate a VGA card either, so what? > --Chuck McManis That's fine... but the Transformer which we claim emulates an IBM doesn't have any graphics or sound emulation whatsoever. You wouldn't NEED to emulate HAM mode to have a reasonably complete Amiga (the only things that use it are digitising & respective paint programs and for looking at pretty pictures). Still, though, any Atari ST trying to emulate Amiga graphics (copper etc) would be extremely slow... On the other side of the coin, though, I'm still surprised no-one has released an Atari ST emulator commercially. It seems to me like that would be a lot easier to copy than a Macintosh, and you wouldn't need any hardware with it either. Maybe no one thinks there would be any interest, but from what I have heard the music software for the ST is far superior to likewise existing on the Amiga (of course you'd need to wire up a MIDI interface). -- James A. Treworgy -- No quote here for insurance reasons -- jtreworgy@eagle.wesleyan.edu jtreworgy%eagle@WESLEYAN.BITNET
king@cell.mot.COM (Steven King) (11/29/89)
In article <21@macuni.mqcc.mq.oz> ifarqhar@mqccsunc.mq.oz (Ian Farquhar 7420) writes: >...and I believe there is [...] [an] Apple II emulator somewhere out there. Does anyone know if this does indeed exist, and where I can get my hands on it? I've got an Apple //e sitting right next to the Amiga, and I think it would be fun to play around for a bit with the emulator... :-) -- ---------------------------------------------------+--------------------------- It doesn't matter how good a computer it is, it | Steve King (708) 991-8056 can't multitask running anything and being turned | ...uunet!motcid!king off. | ...ddsw1!palnet!stevek
cmcmanis%pepper@Sun.COM (Chuck McManis) (11/29/89)
In article <4120@eagle.wesleyan.edu> jtreworgy@eagle.wesleyan.edu writes: >That's fine... but the Transformer which we claim emulates an IBM doesn't have >any graphics or sound emulation whatsoever. You wouldn't NEED to emulate HAM >mode to have a reasonably complete Amiga (the only things that use it are >digitising & respective paint programs and for looking at pretty pictures). >Still, though, any Atari ST trying to emulate Amiga graphics (copper etc) would >be extremely slow... James you are both correct and incorrect at the same time. The reason the Transformer (and PC ditto for that matter) can get away with just emulating a CGA or MDA card is that a lot of IBM PCs went out the door in that configuration and thus most of the PC programs support those graphics adapters. There is however no Amiga without HAM, nor is there one without stereo sound. Thus a lot of programs can and do depend on these being available. The sound system is used by nearly every game and a few mainstream applications as well. The ability to emulate is directly tied to the ability to emulate the base configuration. The ST can't do that. Sorry. >On the other side of the coin, though, I'm still surprised no-one has released >an Atari ST emulator commercially. It seems to me like that would be a lot >easier to copy than a Macintosh, and you wouldn't need any hardware with it >either. Wrong again, because it is illegal to make a copy of TOS without Atari's permission. You might be able to get an old buggy version of Disk TOS but it would hardly be worth it. > Maybe no one thinks there would be any interest, but from what I have > heard the music software for the ST is far superior to likewise existing > on the Amiga (of course you'd need to wire up a MIDI interface). This is also incorrect, to the extent that Notator(sp?) exists for the ST and there is no comparable system for the Amiga, the availability of sequencers is nearly identical (with the exception that you can't get Music-X for an ST). That is because it is generally only worth writing something to sell into _both_ markets, at least that way it is almost the size of the Mac market. --Chuck McManis uucp: {anywhere}!sun!cmcmanis BIX: cmcmanis ARPAnet: cmcmanis@Eng.Sun.COM These opinions are my own and no one elses, but you knew that didn't you. "If it didn't have bones in it, it wouldn't be crunchy now would it?!"
ltf@attctc.Dallas.TX.US (Lance Franklin) (11/29/89)
In article <128510@sun.Eng.Sun.COM> cmcmanis@sun.UUCP (Chuck McManis) writes: >Wrong again, because it is illegal to make a copy of TOS without Atari's >permission. You might be able to get an old buggy version of Disk TOS but >it would hardly be worth it. Well, you could always go the way the Macintosh Emulator has gone...Atari has recently upgraded their TOS to V1.4, and has been releasing ROMs (well, EPROMS, actually) with their latest TOS. put them in a box similar to the AMAX box and Voila! Then it's just a matter of reading them into memory, patching what needs to be patched and popping an ST disk into the drive. Who knows...you might even be able to make it multitask. :-) Lance -- +-------------------------+ +------------------------------------------+ | Lance T Franklin | | "And all who heard should see them there, | ltf@attctc.DALLAS.TX.US | | And all should cry, Beware! Beware! +-------------------------+ + His flashing eyes, his floating hair!"
pfaff@mercury.asd.contel.com (Ray Pfaff - Oakwood 457 934-8162) (11/30/89)
Steve King writes: >Does anyone know if this does indeed exist, and where I can get my hands on it? >I've got an Apple //e sitting right next to the Amiga, and I think it would be >fun to play around for a bit with the emulator... :-) I'm curious as to whether the Mac's apple II emulator works on the amax... :-} Ray Pfaff pfaff@europa.asd.contel.com
king@cell.mot.COM (Steven King) (12/01/89)
In article <5001@nigel.udel.EDU> pfaff@mercury.asd.contel.com (Ray Pfaff - Oakwood 457 934-8162) writes: >I'm curious as to whether the Mac's apple II emulator works on the >amax... :-} >Ray Pfaff What the world needs is an Amiga running a Mac emulator running an Apple emulator running a C64 emulator running a VIC20 emulator running... Or, maybe an Amiga running an ST emulator running an Amiga emulator running Amax? "Oo eck, DM!" -- ---------------------------------------------------+--------------------------- I'm very good at giving directions, especially if | Steve King (708) 991-8056 I'm giving them to myself, 'cause I know what I'm | ...uunet!motcid!king talking about. | ...ddsw1!palnet!stevek
capit@peun11.uucp (Capitain) (12/04/89)
king@cell.mot.COM (Steven King) writes: >What the world needs is an Amiga running a Mac emulator running an Apple >emulator running a C64 emulator running a VIC20 emulator running... >Or, maybe an Amiga running an ST emulator running an Amiga emulator running >Amax? Aladin, a German Mac Emulator for the Atari ST, perfectly runs _][_in_a_Mac_, so the ST is emulating a Mac emulating the Apple ][... +-----------------------+---------------------------------------------------+ ! Pit Capitain, DX-PC ! US: ...uunet!philabs!linus!nixbur!capitain.pad ! ! Nixdorf Computer AG ! not US: ...{mcvax!}unido!nixpbe!capitain.pad ! +-----------------------+---------------------------------------------------+
a218@mindlink.UUCP (Charlie Gibbs) (12/06/89)
In article <681@nixpbe.UUCP> capit@peun11.uucp (Capitain) writes: >Aladin, a German Mac Emulator for the Atari ST, perfectly runs _][_in_a_Mac_, >so the ST is emulating a Mac emulating the Apple ][... Just for the hell of it, one evening I fired up the Transformer on my 1000. Under this fake MS-DOS environment I ran Z80MU, a Z-80 CP/M simulator. As if two levels of software emulation (68000 -> 8088 -> Z-80) weren't bad enough, I then ran BASIC-80, the CP/M BASIC interpreter. When I typed PRINT SIN(.707) it managed to come back with the correct answer - eight seconds later. Afterwards, of course, I apologized to my 68000 for abusing it so badly. :-) Charlie_Gibbs@mindlink.UUCP My Amiga IS a PC! It says "Personal Computer" right on the box!
pfaff%mercury.asd.contel.com@cunyvm.cuny.edu (12/06/89)
Steve King writes: >Does anyone know if this does indeed exist, and where I can get my hands on it? >I've got an Apple //e sitting right next to the Amiga, and I think it would be >fun to play around for a bit with the emulator... :-) I'm curious as to whether the Mac's apple II emulator works on the amax... :-} Ray Pfaff pfaff@europa.asd.contel.com
a218@mindlink.UUCP (Charlie Gibbs) (12/07/89)
In article <1989Dec6.213459.24177@i-core.UUCP> pete@i-core.UUCP (Pete Ashdown) writes: >Where's that rumored Apple II emulator? Don't forget CP/M emulation :-). I don't know about an Apple II emulator, but there's always my CP/M emulator program, SimCPM. It's entirely in software so it's not that fast, but it does a pretty good job of emulating a Z-80 CP/M system. The latest version is on Fish disk 186 (along with an older version of my assembler, A68k). If you can't find it anywhere, send me a blank disk at: Charlie Gibbs 2121 Rindall Avenue Port Coquitlam, B.C. Canada V3C 1T9 and I'll send you the latest versions of my goodies. Charlie_Gibbs@mindlink.UUCP For every vision there is an equal and opposite revision.
pete@i-core.UUCP (Pete Ashdown) (12/07/89)
Where's that rumored Apple II emulator? Don't forget CP/M emulation :-).
wendell@medsys.uucp (Wendell Dingus) (12/08/89)
capit@peun11.uucp (Capitain) writes: >king@cell.mot.COM (Steven King) writes: >>What the world needs is an Amiga running a Mac emulator running an Apple >>emulator running a C64 emulator running a VIC20 emulator running... >>Or, maybe an Amiga running an ST emulator running an Amiga emulator running >>Amax? >Aladin, a German Mac Emulator for the Atari ST, perfectly runs _][_in_a_Mac_, >so the ST is emulating a Mac emulating the Apple ][... AMAX, an Amiga-based Mac Emulator perfectly runs ][ in a Mac also. I recently saw it at a user's group meeting, and it was great! Wendell Dingus FAX: (615)/282-1517 FONE: (615)/246-6015 USENET: ...uunet!medsys!wendell
pete@i-core.UUCP (Pete Ashdown) (12/13/89)
In article <5001@nigel.udel.EDU>, pfaff@mercury.asd.contel.com (Ray Pfaff - Oakw >Steve King writes: >>Does anyone know if this does indeed exist, and where I can get my hands on it >>I've got an Apple //e sitting right next to the Amiga, and I think it would be >>fun to play around for a bit with the emulator... :-) > >I'm curious as to whether the Mac's apple II emulator works on the >amax... :-} > >Ray Pfaff >pfaff@europa.asd.contel.com It does in fact, work. However, it is VERY slow. I really think the Amiga could do a lot better job of emulating the Apple IIe than the Mac does. It did an 'ok' job of doing the C64 with all its custom hardware, the Apple should be a snap because it has NO custom hardware. I can testify to the ability of the A-Max to run "II in a Mac" though. I played a game of Sierra's Sabotage in it. I did however get too bored to finsh it :-).