Contents | < Browse | Browse >
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
%% Emulation Rambler                                  by  Jason Compton  %%
%% Bringing a little sense of closure to things.   jcompton@bbs.xnet.com %%
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

This Emplant/A-Max IV string is, perhaps, running a bit long.  I plan
to offer my Amiga-side benchmarks this week, then sum it all up
in the next week or two and offer my final evaluation.  Right now,
I'd like to thank Jim Drew of Utilities Unlimited and June Brown of
ReadySoft for providing the boards in the first place.

Qualitative...
--------------
A friend owns a Mac LCIII.  He's the one who provides me with a large
amount of the Mac software I run in these reviews, and he told me to
run Mac BZone (a BattleZone, the old arcade game, clone.)  I showed him
its performance on my system and he was thoroughly impressed.
Mind you, he had to run FPU emulators to get it to work, but nonetheless,
he was surprised at the results on both emulators.  From my experience,
it winds up how the benchmarks suggest it should:  faster on the A-Max IV
in black and white and faster on the Emplant in 16 color.  I will say 
this: The Emplant tends to do gameplay slightly better than the A-Max IV.
Largely, it is due to the faster 16-color performance on my ECS chipset,
and it may have something to do with mouse and "unconventional"
routine support, such as with the game Star Wars: clearly not a system-
friendly application, but Emplant is playable while A-Max IV stutters 
quite a bit with the mouse translation.  On a broad scale, however, 
(general application work and the like) I make out no real speed 
differences between the two...in fact, surprisingly, MacMan Classic (a 
near-perfect PacMan clone) runs almost identically on the two systems, and
it uses color.  Strange, but then, so is multitasking Macintosh emulation.

Quantitative...
----------------
Ok, so which machine lets you do what you want on the Amiga side while
it runs?  Most importantly, can you just leave it running and do something
else while the Mac sits idle?  Emplant's method of handling it is a 
selectable task-priority for an unselected emulation.  ReadySoft, through 
checkbox options, lets you decide how system time will be allocated, the 
optimum situation for the Amiga being that all unused time goes back to 
the Amiga, especially while not selected.  I didn't expect the performance
to differ much, actually, in terms of making time available for the Amiga.
I was right.  In looking at the AIBB modules (I will make them available to
anyone who is interested, just ask), on tests run on the following systems:

A-Max IV, priority 10
A-Max IV, priority 0
Emplant, priority 0
Stock A3000/25 (multitasking disabled)

with multitasking enabled for the emulator tests and the Mac Bzone
intro screen running on the Mac emulation, -1 priority set to unselected
for Emplant and the "turn-over-all-unused-time" option set on A-Max.
No other programs are running anywhere else.

The results are pretty close to equal.  There were a few exceptions, but
I wouldn't read too much into them:

Counting ties and treating the A-Max 10 and 0 priority together, A-Max
actually won most of the tests.  The 0 priority A-Max won more than the
priority 10 A-Max did (understandably):
Flops, TGTest, EllipseTest, LineTest, BeachBall, and InstTest

Emplant took WritePixel only, but it is worth mentioning LOUDLY that
the majority of the tests were very, very close.  Only a couple of tests
have strange results:

On TGTest, both A-Max trials had a 1.  Emplant was close behind at .96,
but the clean 3000 took a .9.  Strange.  Similar results occurred with
Ellipse, but the Emplant was a mere .01 off the A-Maxes, and with
LineTest, which has A-Maxes at 1, Emplant at .99, and the 3000 way
down at .83.  This was not the sort of thing I expected.

Emplant was the only test that didn't take a 1 in TranTest: it had a .93.

So what do I make out of all of this?  Taking the tests and the 
performance of the machines into account, there's really not much of a 
difference. A few of the tests come up with noteworthy results, but the 
overall effect is very similar.  When it's time to give up time, both 
emulators do a good job.

A-Max IV serial files?  Apparently, a good thing.
----------------------------------------------
If you remember, I posed the question "What good is directing serial
output to a file?"  Maxwell Daymon (mdaymon@rainbow.sosi.com) knows
more about it than I do, so I'll let him tell you.
[begin quote]
> Serial/Parallel:  Configures A-Max's port emulation.  Since A-Max comes
> with a Mac Printer and Modem/MIDI port, you can choose that the Modem 
> and Printer support come from either the Amiga ports, the A-Max ports, 
> or file, in which case all the port's output is directed to an AmigaDOS 
> file. 

> Can someone explain the practicality of this to me? 

Sure, we use one of two devices: Our $2,000 laser printer (600dpi) or an 
$86,000 Linotronic imagesetter (>2000dpi)

In order to use the Linotronic, we have to output to a file which many 
Mac programs don't like to do. They ASSUME (Quark 3.1.1) that the 
Linotronic is hooked up (via Appletalk) or you're a worthless pig who 
doesn't deserve output. Thank Readysoft for this feature!!!

Also, I don't want to have my laser printer hooked up to both ports, and 
I prefer using the Amiga port - so being able to select an Amiga port is 
also very helpful for programs/drivers that don't *demand* Appletalk.

Anyhow - they are very useful features when you consider service bureau 
output considerations.
[end quote]

There you have it.  Serial files explained.

Bringing it all to a close...
-------------------------

Believe me, this has been a fun run.  I believe I've done a pretty good 
job of showing the strengths of each emulator as the series has gone on.
Now, though, I think it's time to wrap things up and offer some final
analysis and conclusions.  So I'll do it in the next issue. :)

Seriously, coming soon should be reviews of the GoldenGate II board,
the SX-1 (not truly an emulator, but I'll pretend and you'll play along),
and the Vortex 486/50 board.

Thus, within two weeks, the Emplant/A-Max IV saga shall draw to a close.
Oh, expect random comments about them to flow in future issues...you'll
see the occasional benchmark or quote or two, some performance reports,
maybe even a couple of really useful Mac programs on the new Amiga
Report coverdisk.  We'll see.

Until then, keep the emulators warm.