Executor 2.0/executorv2.0.iso/pc/linux/extra/docs/maillist/text/archive.96/text2227.txt

Text File | 1996-07-25 | 3.6 KB | 66 lines

In article <[e-mail]>, Steve D. <[e-mail]> wrote:
>I think ARDI has the right idea just be making a DOS version. It runs
>fine in Windows95, even in Dos. So why are the OS/2 users complaining?
>They have DOS capability. [bla bla bla]). If they do [a 32 bit port],
>should make it to the Win-32s/Windows95 platform. This is the most
>widely used platform. These people [allegedly] account for the majority
>of the EXECUTOR users. [bla bla bla].  I beleive ARDI has already done
>great. I don't think we should complain, we purchased the product (at a
>lower cost than the shipping cost), [etc...]

What?  Is this another troll?  The only things said here that I agree
with are that ARDI has done a great job on Executor and that we shouldn't
complain.  The more so because ARDI has already indicated it plans to
suuport OS/2, nothwithstanding the conventional wisdom that only the
Windows platforms are of any consequence and that all others not only
can be, but should be, ignored.  Windows users need not worry, ARDI
indicates that it plans to support you, too -- and it will not hurt
you one bit for ARDI to do something you don't agree with.  I care
nothing for the OS wars, preferring a diverse and competitive market-
place because I believe it to be both more interesting and better for
the industry.

ARDI deserves support for many reasons, not least of which is their
choice to support a diverse marketplace.  Still, there's plenty of
reason to support them even if you don't agree with my diverse market-
place philosophy.  For one, the product is outstanding, with better to
come.  Another is their conduct, or should I say philosophy: they have
solicited and used the support of their target user community and
haven't even demanded that we fork over big bucks to get a peak at
their product.  Recently, Cliff allowed as how some $2M has been spent
with (I gather) little compensating revenue to date, but despite their
obvious need for cash, they allowed me the educational discount on the
modest strength of my part-time instructorship at Columbus State
Community College, a large institution of its type, but no powerhouse
of computer science, to be sure.

I anticipate that I will be send them more money for some sort of
cross-platform license for future versions as I actually boot and use
Win95, OS/2 Warp, and Linux on my home machine.

>ON A SIDE NOTE, WILL EXECUTOR RUN ON AN 80286?? WE STILL USE THEM AT
>SCHOOL, AND THIS MIGHT BE INTERESTING TO SEE...

Well, I think we can safely say that the answer is no.  Not only would
emulation of the 32 bit flat address space Moto 68K family be difficult,
if not impossible, to emulate successfully using 64k segments on the
286, the emulation of the Moto 32 bit instructions on the 286' 16-bit
architecture would impose an additional heavy overhead, as well.  (O.K.,
O.K., I know that 68000 had a 24 bit address space not 32, but
Executor's emulation target is a 68040, anyway, right?).

I may not remember perfectly but 286's topped out a 20MHz or so.  Now
I've successfully run Executor on a 25MHz 386sx, but it was pretty
leisurely (q12 might be faster, but I no longer have the machine).  If
the address space issues were solved through some scheme for mapping
the Moto address space into a sheaf of 64k segments and 32 bit
operations successfully emulated, the result would likely amble along
at a pace rather slower than "leisurely."

Larry
--
Larry A. Shurr ([e-mail] or [e-mail])
A Keane consultant on assignment at Lucent Technologies/Bell Laboratories
Innovations, but not officially representing the views of either company.