ABS != PCB. I was drilling an ABS battery case from Radio Shack to put a Ps/2 adapter in for one of my TICI 2.0 hubs.

Ah, okay I got it:

Kerm==ABS Battery Case
Jpez==PCB

How do we plan on raising money for this endeavor?

about the memory, please have a USB2.0 port for thumbrives to use for storage.

yeah, that would be nice, or at least some kind of full size USB function.

It's a pain in the next to write a usb root host function, but if I remember correctly there is a discrete IC that will handle it....

Well, there's a virtual usb peripheral source code out there for the SX...it could be tweaked to be made into a USB host (or we could always write one from scratch ; if that's the case, then maybe we should program in something other than 100% assembly...)

This will be invaluable to our project: http://www.beyondlogic.org/

For USB: http://www.beyondlogic.org/usbnutshell/usb1.htm

Yup, seen that site. They have some excellent guides on the USB protocol; I browsed through it at one point, and they have crammed an amazingly large amount of info into that tutorial.

I'm a little late coming to this topic, and I'm not 100% sure whats going on, but if you can get away with not using 100% ASM for whatever chip you are looking at, as in there is actually a decent compiler for it (or a gcc port), go for it. Programming an OS in C will lower the time it takes to code it, make debugging much easier (especially with debuggers telling you what function ain't working), and can make fixing/adding features take much less time. Instead, just inline ASM into C functions where speed is absolutely critical.

Ultimate Dev'r and I have been theorizing for years about developing two different new education handhelds called respectively the Blue Edition and the Red Edition. Each has a different form factor and intent, but the overall hardware is identical, and the OS used will be the same. I believe I have some drawings, renderings, and videos of the Red Edition somewhere on this forum... Anyway, writing the OS in C is probably an excellent plan.

Did you consider just using Linux for the kernel? You can, of course, run your own shell ontop of it (instead of, say, BASH), but that would give you an already stable kernel, along with things like threading, schedualing, etc... that programmers would expect for a modern system

true, Linux can run on anything. I second the Linux idea.

I was leaning more towards a brand new OS instead of a Linux port; but it all depends on what hardware we decide to use. Speaking of which, I'll start a topic about that; Kerm, you make us a new subforum.

That's what I was thinking as well



It just occured to me; the Red and Blue Editions of what? We need a broader project name.

that is very true, maybe of Cemetech Handhelds or something.

That won't work; Kerm and I are developing the Red and Blue Editions separately (to an extent). So it has to be somewhat of a neutral name.

I think just RE and BE are fine names on their own. They apply relative to one another, so it works without a further referential.

true, but when you say Red Edition to someone else, they would say, the Red Edition of what? I think we need something.

What Kerm's trying to say is that it's the Red Edition of the Blue Edition, which is the Blue Edition of the Red Edition

Oooh, recursive naming. But not nearly as cool as, say, GNU == GNU is Not Unix

Exactly. Nothing is cooler than infinitely recursive self-referential naming. The chix totally dig it.

uh-huh. Well, I guess that would work.