#103249 - heyyeah - Wed Sep 20, 2006 8:21 am
Hey, I was wondering if some of the more experienced developers here could give any tips on estimating development time for homebrew apps.
We are doing a session for faculty members at uni who might be interested in building games for use in education. Not so much the learning curve to getting used to the libraries and tools but more general things
for example, an estimate to code:
a scene with 10 interactive objects
and create the following assets:
textures for objects
background artwork
building and animating 2 3d objects
this would be probably split between a couple of programmers and a 3d savvy graphic artist
#103319 - TJ - Wed Sep 20, 2006 10:55 pm
I don't know much about the 3D hardware of the machine, but I do know that it is very similar to OpenGL, so if they have experience with that, the learning curve would be minimal.
Taking a look at an example program or two from libnds would give them enough of an idea about initializing the hardware, input, etc.
#103351 - heyyeah - Thu Sep 21, 2006 7:09 am
thanks. there is a couple of us here with opengl and the initialising looks ok.
Can anyone here post info on their estimated + actual development time experience with homebrew?
thanks again
#103364 - PypeBros - Thu Sep 21, 2006 9:48 am
i'd say give them 1 weeks to get their hand with the tutos, find out how to compile/modify/run demo code on an emulator and see how the things work together ...
if they don't have any experience with 3D programming, i'd start making them a openGL animation on PC before they run it in the emulator .. I guess if they have a fair programming background, that should take a max of 2 weeks or so ... then two extra weeks to get the animation running in the DS emulator ... running it on real hardware should be cherry-atop-of-the-cake, not something required, imho.
_________________
SEDS: Sprite Edition on DS :: modplayer
#103399 - josath - Thu Sep 21, 2006 4:21 pm
| PypeBros wrote: |
| then two extra weeks to get the animation running in the DS emulator ... running it on real hardware should be cherry-atop-of-the-cake, not something required, imho. |
emulators tend to be more forgiving than hardware, and at the same time not supporting some functions. if you get it to work on an emulator without testing on a real ds, don't assume it will work at all. ideally you'd do most of your development on a real ds, occasionally using an emulator once you've determined that it supports your app well enough.
#103493 - PypeBros - Fri Sep 22, 2006 9:56 am
| Quote: |
| emulators tend to be more forgiving than hardware. |
true. and that's precisely why i suggest to suggest them to focus on the emulator, still letting them know that making it run on RealHardware(tm) might be trickier ...
That's still a student project, right? so unless you can give each of them a flash cart, a DS, etc. and can provide them with a development cycle of less than 1 min per test, you're going to kill the project if you're not happy with a program that runs in the emulator.
At least, that's the kind of policy i apply when correcting students programs. do they work fine with netscape/xmms ? they can hope to get 17/20 ... do they also work with <any browser>/<any player>? then they can hope 20/20. Otherwise, they spend too much time trying to make it work with anything and it eventually don't run on the system you use to test the program :P .. or they end up with something that isn't finished.
_________________
SEDS: Sprite Edition on DS :: modplayer
#103502 - gmiller - Fri Sep 22, 2006 1:51 pm
I have students work with the emulators first and then try it on the hardware for my GBA class. I am still trying to find a DS emulator that I can put source code debugging into since I have not seen much in any of the DS emulators I have looked at. For the GBA the SDL verson of VBA was the easisest to fix to get all of the source code debugging stuff working. I am still hoping to find something like this for the DS but if it is there it is mainly hidden. I don't really consider the build in debuggers thast useful in a teaching environment because most are targeted at assembly aware programmers and my students don't get enough of that for it to be very contructive in my class.