[XCSSA] Any Feb Presentations?

xcssa@xcssa.org xcssa@xcssa.org
Mon, 12 Feb 2007 01:01:44 -0600


On Feb 11, 2007, at 7:43 PM, xcssa-admin@xcssa.org wrote:

> I think Thomas King was still planning on the Zimbra  
> presentation... was
> anyone else going to want to present on anything or not (one week and
> counting).

Well, it looks like I'll be down there on Monday anyhow, so I'll  
bring my Sega Genesis development gear.

Yes, I finally got tired of working around the limitations of the  
Atari 7800 and ColecoVision.  So I tried to figure out what the next  
viable platform was.  The Sega Master System was kind of nice, but  
consoles  were quite hard to find, and therefore it would have a  
limited audience.  The NES had cartridges with two buses, weird  
(metric) pin spacing on the edge connector, any but the simplest  
cartridges needed a "mapper" chip, and then there were the general  
reliability problems.  Oh yeah, and I never really liked writing 6502  
code.

The next thing in line was the Sega Genesis.  It had the good parts  
of the ColecoVision (video has its own bus to avoid hogging the CPU  
bus) and the SMS (like Coleco with 16-color graphic tiles and more  
sprites), and none of the bad parts of the NES (single cartridge bus,  
lots of VRAM, video chip ports that don't change function during the  
display cycle).  Plus it had a 68000, which is simply a kick-ass chip  
to begin with, no bank switching (at least not until you hit 4  
megabytes), and there's still a Z80 in there as a co-processor.  The  
only negative is that cartridges need to be 16-bit, which means you  
need either two regular EPROMs or one big 40-pin monster EPROM.   
After about two months of working with it, I think the Sega Genesis  
is probably the easiest to program console there has ever been.

It also helped that I got a floppy disc game copier for the Genesis  
years ago.

What I normally do is test code on an emulator, then when I want to  
run on the real hardware, I write the code image to a floppy disk,  
stick it in the game copier and load it.  I also have an RS-232  
converter board which can plug into any of the controller ports.  All  
three ports (the early models had a rear port which was functionally  
identical to the two front ports except for an opposite gender  
connector) can be set for 4800 baud serial communication.  So I  
hacked up an old version of Macsbug which I can assemble into the  
code when I need to do some serious debugging.  I also had a couple  
of cartridge boards made, so I can burn a pair of EPROMs and play on  
an actual console.

Actually, I didn't have one with a rear port handy, but I had the  
next best thing.  I had a later version of the first model which had  
that part of the board unpopulated, so I cleaned out the holes, got  
the appropriate parts from junk boards, and installed them.  I also  
installed a male DB-9 port instead of the standard female, so I could  
use the RS-232 board with no changes and no need for a gender changer.

Right now I'm still trying to catch up with what I had working  
before.  My ColecoVision RPG is ported over to the same state as it  
was before, except that I still need to port over the Coleco sound  
routines.  I updated my assembler so that I could mix code from  
different CPUs, so now I can mix the Z80 and 68000 code, and use the  
Coleco sound routines with minimal changes.  And I've started work on  
porting my Atari 7800 game "Tubes" to the point where I can make the  
tiles fall and stack up.

  - Bruce -