GP2X Networking Toolkit: Waste Of Time?


GeminiDomino

Member
Joined
Dec 17, 2005
Messages
374
With the summer coming up and the tourists going home, I've got lots of slow time at work in my future and was thinking of doing some 2X developing (assuming of course I can ever get codeblocks to @#$!@! compile... if this run doesn't take, I'm going back to Komodo).

Since my biggest strength is in networking (the TCP/IP kind, not the BS sales-speak word for "kissing ass") I was thinking about making a toolkit for a simple networking framework for network enabled games.

The question is: would it be worth the effort? You guys are the devs, so you'd probably know better than anyone else if netplay will ever take off (the users would all be for the idea, of course, even if they'd never end up using netplay) or if it's a pipe dream.
 
Since my biggest strength is in networking (the TCP/IP kind, not the BS sales-speak word for "kissing ass") I was thinking about making a toolkit for a simple networking framework for network enabled games.
The question is: would it be worth the effort?

There is no need for a 'toolkit approach', in my opinion .. as soon as the GP2X cable scene gets sorted out (i.e. people can get EXT->USB cables for their GP2X's), then things like SDL_Net et al. become available .. and this is about as good as it gets for multiplayer gaming/networking ports/gaming ..

To highlight this I have 3 GP2X's right now connected to my laptop (linux), and they are all on the 'net, just like any other device in the known universe. Its a piece of cake, also, to rig up a cable with EXT on one end, and Mini-USB on the other end: with such a thing, you have a network rig quite usable for multiplayer gaming using the 'normal Linux TCP/IP' libs ..

But because I believe there is no need for the toolkit/libs/port approach, I also see that there is a need for another set of tools: basic, common, admin-task related things .. like, for example .. a decent 'driver manager' for the GP2X interface which allows you to move in and out new modules as drivers, freely at will. Right now the 'Advanced Settings' approach of the GP2X firmware uses custom scripts and 'accepted onboard driver modules' to manage everything in a kind of 'kludgy' manner, but it does, in fact, work out okay. What would be great is to see more .GPU (GP2X Utilities) that allow you to set up the basic, common, system stuff that you'll find in /etc and /usr/bin for whats needed.

Like, for example, an SD-card formatter GUI. A Module loader/unloader GUI, etc.

So I'd say put your skillz to use on the .GPU side of things, with a focus on managing whats on board in a way that makes it easier for people to pack a module with their apps, manage it in the local linux context, etc...
 
Last edited by a moderator:
Like, for example, an SD-card formatter GUI. A Module loader/unloader GUI, etc.

So I'd say put your skillz to use on the .GPU side of things, with a focus on managing whats on board in a way that makes it easier for people to pack a module with their apps, manage it in the local linux context, etc...

Good idea. I've got most of the code logic down but, as usual, I'l choking on the GUI. My first gut instinct was a simple ncurses-type menu but this runs me into two problems: 1) AIUI, the GP2X screen generally uses the framebuffer, so I'm not sure yet if it will act as a terminal 2) ncurses seems to depend on having a terminfo database, which doesn't exist and is big enough that I'd like to avoid putting it on the nand.

A workaround for 2 is useless without solving 1, so it's on the back burner now. Do you think a graphical UI would be better considering the small size of the screen, though?
 
Last edited by a moderator:
I compiled some stuff that controls the terminal with *curses (screen, gdb) for the gp2x. I just included a few of the terminfo files with it to save space. It does work, but NOT using the gp2x's screen. Those things were meant to be used in combination with an external terminal.
You could run them inside of sterm, although I think a terminfo file for it would need to be made...

P.
 
With the summer coming up and the tourists going home, I've got lots of slow time at work in my future and was thinking of doing some 2X developing (assuming of course I can ever get codeblocks to @#$!@! compile... if this run doesn't take, I'm going back to Komodo).

Since my biggest strength is in networking (the TCP/IP kind, not the BS sales-speak word for "kissing ass") I was thinking about making a toolkit for a simple networking framework for network enabled games.

The question is: would it be worth the effort? You guys are the devs, so you'd probably know better than anyone else if netplay will ever take off (the users would all be for the idea, of course, even if they'd never end up using netplay) or if it's a pipe dream.

I'll answer with a question, would you enjoy coding it? If so do it, its as simple as that. I'm doing a 3D engine and i'm sure its not needed that badly as there are other 3D engines in development. I'm doing it because I enjoy coding it.

For someone like my self with no network knowledge it may help although as the gp2x has no real built in network HW I doubt I would support it.
 
Last edited by a moderator:
Back
Top