XGameStation Designer Talks Specifics
Thanks to GameZone for their interview with Andre LaMothe about the XGameStation, the DIY, programmable game console theoretically due this December, but likely somewhat delayed. Although details of the XGameStation are still being finalized, LaMothe describes the specific technical details: "I think the ARM7 is going to be my choice as the final main CPU at 33-66 MIPS, and an FPGA GPU that does basic sprite, character, and bitmap graphics in 4-256 colors, with 1-4 Megs of RAM", and goes on to evangelize the software: "We will surely encourage people to port as many games and emulators as possible to the XGS. I am mainly concerned with getting MAME, Intellivision, Atari 2600, etc. ported ASAP."
He's going to have some trouble getting a MAME port with only 1-4MB of RAM to work with. Even games with relatively simple hardware go over that easily: pacman needs about 6MB just for the emulator core.
On the other hand, if the graphics chip is thoroughly customizable, we might see some dedicated single-system emulators that use the built-in graphics and are designed with low-memory situation in mind. Could be pretty cool.
I'm sorry, but I think I'll see bitboys videocards before that "console" makes it to the market.
Mother is the best bet and don't let Satan draw you too fast.
Yeah right. Why would anyone want to buy stuff to cook his own meal, if you can go to McDonald's?
The fun't in creating something, not using it.
--
Finagle's First Law: If an experiment works, something has gone wrong
One of the Computer Architecture projects I did for college consisted of using a programmable Xylinx board (~$1000), hooking it up to a monitor and then hooking two orginal NES controllers and playing pong on it. It was one of the most incredible learning experiences I ever encountered. We wrote our own processor, uploaded it to the board, wrote our own compiler and assembler and then wrote our own version of pong. We also had to program a chip for our video controller and NES controllers. Needless to say, it was a very expensive game of pong.
If the XGameStation is coming out for somewhere around $200 and allows users to manipulate hardware as well as software, I say that's a great deal. Granted, I'm sure that it won't be as robust as a Xylinx board but the educational benefits are still excellent, even if you're not going into games.
That's only if you have all the MAME drivers loaded into memory at once, which is something you'll never do.
The main executable can be as small as a few hundred kilobytes, and then load the proper game driver from a datfile full of drivers (as is the case with some current distributions of MAME, like MAMEplus). There's no reason for Pac Man to require 6 MB, unless you were using some incredibly inefficent form of dynamic recompilation.
Of course, with 4 MB of ram, you can never run anything more complex than Capcom CPS1 games.
Happiness is relative, Based upon the way we live.
A cheap console, that you can burn any game for, run emulators on, control the operating system, and totally hack apart..... Console, thy name is DreamCast.
"If you think you have things under control, you're not going fast enough." --Mario Andretti
the project is intended _mainly_ to teach hardware design and tinkering for consoles. if you start too complex only hardcore elctronic engineers will be able to do anything with it. it's more of an introduction to console design rather than another platform to code for but i am sure many will do just that, which is fine but not its primary purpose.
remove NOT from email.
Personally, I'm probably going to pick one up if only for the enjoyment of self-challenge - with the limited power of it (cpu/ram) what exactly can I pull off? As a programmer, what tricks will I need to use and learn in order to push this machine to its limit? And in a more commercial environment with more powerful hardware, how can I use what I've learned from developing on the XGS to improve my applications?