Castle Wolfenstein 3-D for IIGS

[spector@vax2.concordia.ca (Mitchell Spector)]

	Those who've read the May 1994 GEnie Lamp, probably noticed a
call for those interested in seeing "Castle Wolfenstein 3D" ported to the 
Apple IIgs, to send letters of encouragement to the staff at ID Software.
Well, in light of ID Software having already written and completed a
65C816 version for the SuperNintendo using an Apple IIgs, I figured a
GS-specific port wouldn't be too difficult (afterall, remember Bill
Heineman originally wrote "Out of This World" for the SuperNintendo
using an Apple IIgs, and was then able to port it to the IIgs -- the
realtime 3D polygon animation in OOTW was nothing to sneeze at either!).

	So, to get to my point, I went ahead and wrote ID Software and
received a reply (from a programmers point of view) of why an Apple IIgs
port would be non-feasible. Since I'm not a programmer (don't consider
BASIC a real language ;-)) I thought maybe some of you more experienced
could look over this list of points and see if there's any real truth in
them. And then, even if they are, I'm quite sure you most of you could
make what's technically impossible to program...possible! :-)

(clipping from mail below)
==========================================================================
I've received a fair amount of mail about this in the past couple of
weeks.  The bottom line is that we have no plans for a IIGS port.
There are two major reasons (feel free to pass this around):

1: (The technical)  The SNES implementation of wolf is the absolute
borderline of acceptable gameplay.  The GS is simmilar, but not quite
as capable as the SNES.

The most time consuming parts of the program are the clearing of the
screen, the scaling of the walls, and the copying to the video
memory.  The SNES can do all of these thing faster than the GS.

The SNES can do the clearing and copying with DMA, which is about
four times faster than the tightest possible pea/pei stack tricked
code.

The SNES rom clocks at 3.5mhz and the ram at 2.5.  I am also using
one of the SNES dma registers to fake an autoincrement addressing
mode.  This lets the scaling be about 60% faster than the best
possible on the GS.

You couldn't reasonably use more than 16 colors in a scaled GS game,
which would require ALL THE ARTWORK TO BE REDRAWN.

Bottom line: While a GS game would be possible at a lower frame rate
and resolution (or if it required an accelerator), it would not do
justice to the game.

2: (the market)  A runaway hit in the GS community might sell a few
thousand copies.  Possibly.  Our time right now is split between
jaguar ports, retail DOOM, and preliminary work on QUAKE, all big
ticket items.  We are a small company, with only six people in the
development effort.  We just don't have the time for niche products.

I feel for you (a GS was my first real computer), but we are making
our mark by looking forward, not back.

Sorry.

John Carmack
=========================================================================

	Private follow-up can be sent to John (johnc@idsoftware.com), but
I suggest you only do so if you can suggest alternative methods of
programmering where a game port would be possible. Writing more along the
lines of "Hi, I have a IIgs and would buy your game..." show them there's
a potential customer base out there, but won't really do much to change 
their minds. (Um, off on a bit of a tangent, but didn't BurgerBill start
work on a IIgs port of CastleWolfenstein 3D, or was it just a rumour...?
Another idea is to have someone else do the port with ID's permission, if 
they themselves refuse).

Mitchell Spector
sb_spec@pavo.concordia.ca / spector@vax2.concordia.ca