Executor 2.0/executorv2.0.iso/pc/dos/extra/docs/maillist/text/archive.96/text7258.txt

Text File | 1996-07-25 | 10.9 KB | 198 lines

    id m0uVR0w-0007twa; Sun, 16 Jun 96 17:13 MDT
Sender: owner-executor
Received: from ardi.com by ftp.ardi.com
    (Smail3.1.29.1 #3) id m0uVR0M-0007tvn; Sun, 16 Jun 96 17:13 MDT
Path: sloth.swcp.com!news.ironhorse.com!tomato.dussco.com!nwfocus1.wa.com!nwnews.wa.com!uw-coco!uw-beaver!uhog.mit.edu!news.mathworks.com!newsfeed.internetmci.com!newshub.csu.net!bobcat.csusm.edu!erbec001
From: [e-mail] (Patrick McKinnion)
Newsgroups: comp.sys.mac.advocacy,comp.os.msdos.misc,comp.os.linux.advocacy,alt.folklore.computers,alt.binaries.warez.mac,comp.emulators.mac.executor
Subject: Re: Lets talk about somethine that DOES emulate Macs.
Date: 16 Jun 1996 10:47:32 GMT
Organization: Rhiannon Productions
Lines: 175
Message-ID: <[e-mail]>
References: <4idejj$[e-mail]> <[e-mail]> <lexf-10049 <4mi1t7$[e-mail]> <4mk1tt$[e-mail]> <4mu7qr$[e-mail]> <4ng6tn$[e-mail]> <01bb4398.1ed6bc60$[e-mail]> <jargon-2 [e-mail]> <4o03g7$[e-mail]> <4o6t5d$[e-mail]> <4obk1g$[e-mail]> <[e-mail]> <[e-mail]> <[e-mail] ix.net> <[e-mail]> <[e-mail]> <[e-mail]> <[e-mail]> <[e-mail]> <[e-mail]> <[e-mail]>
NNTP-Posting-Host: ciscots2_5.csusm.edu
X-Newsreader: Yet Another NewsWatcher 2.2.0b7
Xref: sloth.swcp.com comp.sys.mac.advocacy:116617 comp.os.msdos.misc:41577 comp.os.linux.advocacy:63470 alt.folklore.computers:77174 alt.binaries.warez.mac:16044 comp.emulators.mac.executor:3514
To: [e-mail]
X-MailNews-Gateway: From newsgroup comp.emulators.mac.executor
Sender: [e-mail]
Precedence: bulk

In article <[e-mail]>, Clifford T. Matthews
<[e-mail]> wrote:

> I suspected Executor was mentioned, but I didn't know for sure,
> especially since the 1/100th speed strongly suggests that Executor is
> *not* what's being discussed.

   And it was somewhat obvious that the person in question didn't know
alot about Executor.

>     Patrick>    I've played with Executor on a '486, and while it was
>     Patrick> a nice application enviroment, doesn't quite equalify as
>     Patrick> a emulator, in the sence of Emplant on the Amiga, or
>     Patrick> SoftWindows on the Mac, i.e. - replicating every function
>     Patrick> of the Operating System, rather than acting as a
>     Patrick> application "envelope".
>
> I'd prefer a different term than emulator, too, since emulators are
> usually very slow and require the firmware and/or OS of the machine
> they are emulating.  Executor is very fast, doesn't require an Apple
> ROM or System file, but consequently doesn't have the degree of
> compatibility a "traditional" emulator would.

   True, which is why I tend to call Executor a "Application Enviroment",
since that, (IMHO), is what is does, it gives a MacOS application a
enviromental "envelope" to run on a x86-based system.   The problem is
that many PC advocates *think* that Executor is a full-fledged emulation
system.   Granted that's a perception problem that's not ADRI's fault, but
it's still a problem.

>     Patrick> In terms of speed, well, I would be far more impressed if
>     Patrick> it could emulate the entire MacOS, including system
>     Patrick> calls, networking, printing, System 7 support, etc., at
>     Patrick> the speed, (or faster), of a 25Mhz 68040.
>
> Some people are easily impressed, some are less so.  We're the only
> company that has *any* Macintosh binaries running on an x86, with or
> without using Apple's code, in addition we're the only company to have
> rewritten enough of Apple's OS and toolbox code to run as many
> applications as we do.

   True.   It's like the singing pig.  It's not how well the pig sings,
it's that the pig sings at all.

>     Patrick> As it is, I feel ARDI is being somewhat dishonest or
>     Patrick> misleading in making the claim that Executor is faster
>     Patrick> than a 25Mhz 68040.  Yes it is, but again, it doesn't
>     Patrick> emulate the full MacOS, doesn't support all the features,
>     Patrick> and acts as a limited application enviroment.
>
> I made the claim because it's true and it's relevant.  Yes, Executor
> is not as compatible as a real Mac, but if the Mac app that you want
> to run on a PC *does* run under Executor, then it will probably run
> significantly faster on an entry level pentium than a 25 MHz 68040
> based Mac.

   It's true and relevant in a sense, however, it's the claim that I think
has been causing the above mentioned PC advocates to think that Executor
is a full emulation system.

   Yes, Executor runs a limited selection of programmes on a '586, at
speeds matching or surpassing a 25Mhz 68040.    This claim is used by some
PC advocates I've encountered to claim that the PC "does Mac emulation
better than the Mac does PC emulation."     Quite frankly, if I was to
write a clean-room application enviroment that ran a limited selection of
x86 apps on the Mac, I might be able to get speeds that are better than
Insignia's SoftPC/SoftWindows.   However, it wouldn't have the same
functionality.    By the same token, if ARDI was to get permission from
Apple to use MacOS routines in Executor, (in much the same way Insignia
uses Windows routines in SoftWindows), I have no doubt that the
compatibility of Executor would go up, but I also have no doubt that the
overall speed would not remain as fast.   I.e., I doubt that a variant of
Executor with full MacOS emulation wouldn't keep the speed that the
current version has.

   I will say that perhaps I was wrong in stating that ARDI was being
misleading or dishonest.  ARDI certainly has no way to control how people
interpret their information.    However, I wish that ARDI could be a bit
clearing in stating that the speed increase is more due to the fact that
Executor isn't emulating the same amount or level of code that something
like SoftWindows does.

>     Patrick> SoftWindows, on the other hand, does emulate a full x86
>     Patrick> chip set, and, (on a PowerPC-based Mac), emulates '386
>     Patrick> code at '486 or faster speed.  Heck, SoftWindows 1.0, on
>     Patrick> my 33Mhz 68040 Mac, is slightly faster than a '286.  Of
>     Patrick> course, I tend to use SoftPC rather than SoftWindows, but
>     Patrick> it's still not bad.
>
> SoftWindows doesn't run everything, although it does indeed run a
> larger percentage of programs than Executor does.  However, speedwise,
> there is no comparison.  Mac apps that *do* run under Executor, run
> much faster under Executor on an N MHz P5 than an equivalent x86 based
> application would run on an N MHz 601 under SoftPC or SoftWindows.

   This is true.  However, as I mentioned above, Executor isn't emulating
the same functionality or amount of code that SoftWindows is.    In the
amount of code that Executor is running, the speed is pretty dammed
impressive.    However, if Executor was emulating the same functionality
and amount of code, I would still think you would get a speed hit.
Personally, I'd rather use a hardware DOS card than software emulation,
but when you don't have one of those critters.....:-)

> I don't think it's misleading for us to say:
>
>         "Executor on a modern Pentium machine runs much faster than a
>         25 MHz 68040 based Mac.  Executor is not as compatible as a
>         real Mac, but speed problems?  I don't think so."

   Misleading ?    Possibly.   Most people don't seem to read past the
"Executor on a modern Pentium machine runs much faster than a 25 MHz 68040
based Mac." part.    Perhaps a better way to write this might be:

   "Within the compatibility issues mentioned in our FAQ, Executor on a
modern Pentium machine runs much faster than a 25 MHz 68040 based Mac.
No, you will not be able to run all the programmes a real Mac can run, but
speed problems?  I don't think so."

   Granted, it's slightly wordier, but it does qualify your claim in such
a way that doesn't leave it as open to misinterpretation.  Mind you, this
is only my opinion, based upon discussions with PC advocates that haven't
seemed to read your FAQ.....:-)

> Executor's
> CPU emulation (which is basically flawless) *is* a much faster way to
> run 680x0 code than a 25 MHz '040, which is especially important in
> the context of using an Amiga to run Mac code, because the Amiga is
> going to run CPU intensive code approximately at the same speed that
> a comparable 680x0 Mac would run it.

   Well, the Amiga advocates tend to claim that the specialized chip sets
in the Amiga help to speed up the Mac emulation, making a 25Mhz
68040-based Amiga faster than a 25Mhz 68040-based Mac.   However, I found,
(with my old Amigas), that it depends on the MacROMs used.   If you use
the ROMs from a 68040-based Mac, you get a speed boost.   If you use the
ROMs from a older machine, the speeds are roughly the same, or even
slightly slower.

> That's what I was trying to point out.  I didn't go into
> Executor details at length because I didn't even know if Executor was
> what was being discussed.

   Executor was being mentioned in passing, but the gist was, at that
time, Mac emulation on the Amiga.

>  I did mention that it's not as compatible
> as a Mac and I gave a URL where a demo copy and a FAQ can be read to
> get much more detail than I could post.

   Which was a help - at least for me, since my copy of your FAQ was a bit
out of date.....:-)

> Basically I agree with everything you've said with the exception of my
> claim being misleading or dishonest.  I see speed and compatibility as
> orthogonal issues.  As I mentioned before http://www.ardi.com/
> contains enough information for people to see both what Executor is,
> but also the claims we make for it.

   Hopefully I qualified my opinions on this a bit better this time.
While I don't say that Executor gives as much functionality as a PC
emulator on the Mac, it's still a dammed impressive bit of software.    I
don't think I'd object to seeing if ARDI could do the same with other
platforms, (say, a x86 emulator or application enviroment for
PowerPC-based machines, or if they could create a better 680x0 emulator
for the PowerPC - hint, hint.....:-)

   - Patrick McKinnion

--
<*>   SP2    Web Page at: http://www.csusm.edu/public/PWMcK/pmck.html
-- Brought to you by "ouchies".  The sharp, prickly toy you bathe with...
-----BEGIN GEEK CODE BLOCK-----
Version: 3.1
GSS/CM d- s: a C++ UL+A W++> N+++ K w-- O+ M++ PS+ PE+ Y+ PGP++ t++@ 5++> X+ R b+++ DI++++ D+ G e+> h--> r++ z+>*
------END GEEK CODE BLOCK------