Build Your Own Arcade Controls Forum
Main => Software Forum => Topic started by: Lilwolf on June 13, 2002, 03:11:23 pm
-
Well, for the last few months... But I've started to move my frontend to be 3d. Not anything link >p< really... but just 3d.
But I started writing my own 3d engine using openGL in java... then I started getting into the openGL and ended up changing it so that it was a full skeletal system with weighted vertexes and a few other fun things. Well, it was WAY overkill for anything I could use for a frontend....
Now, I'm ripping apart the engine, and making a very simple version that I would use for a frontend. I'm hoping to have somethign done soon... but anyone who reads the software boards know that could mean a year.
Now the questionable timedump is that I've started writing a 3d modeler program for myself. Something small, but can make it so you can build your own objects and combine them. But then I start adding a skeletal system so I can add animation... well... even my simple 3d modeler application seems to be overkill.
I think I have some neat ideas for the look and feel. Hopefully I can get a picture.
-
Well, for the last few months... But I've started to move my frontend to be 3d.
-
Well, not ready for the pictures. I've spent time creating a 3d engine from scratch... I have a few pages of notes on replacing my graphics from my frontend with the 3d stuff. Playing around a lot with animation.
The modeler is a replacement for StudioMax or Maya. I have the personal version of Maya, but the file format kinda sucks. And I've built about 5 graphic editors in my day professionally (from simple graphic editors, to full ide's), so I might just write a simple one. The trouble is skinning the textures. If I write it right, might be able to sell it. (already have the engine running with our software at work reading realtime data from real plc's and manufacturing equipment).
But like always... when you get only a few hours a week, and you have plenty of time to come up with ideas... you hardly ever get anything done.
-
That sounds great man!
I'm not that good at 3d yet so maybe I could give you an idea in leu of a full-fledged 3d modeler.
How about a "Build your own cabinet wizard"
Here's what I'm thinking.
You start a new file and it asks for the cabs basic dimensions. (Height,Width,Depth) Now it asks what width of wood is used in the construction (3/4, 5/8, ect) Next an outline box pops up where you can draw the outline of the cabs profile. Then it shows the frontal view where you can adjust the bezels positon /angle as well as the type of coin door, ect. Then you can go in control panel mode where you can make the cp's overhead view and place pre-made controls on it. Finally select textures and it's done!
I know that would be hard to pull off, but think how fast >p<'s 3d cab models would get built!
-
I'm considering it already (the modeler). But mainly because I don['t want to parse maya's files. I would rather create my own simple modeler.
But I wouldn't use measurements like that I don't think. Mainly because I need different information. But make it so you can modify the cabinet by stretching the sides themself.
It will work in two ways.
If I write a good skinner algorithm (to take a single texture over multiple triangles in a polygon)... then I will have a create polygon (or face) and each point will be able to move around in all 3 dimensions.
Or
I will force each polygon to be in a single plane. Then you will work on a 2d plane, and then connect the pieces together (but only able to modify a face on a 2d plane, and only rotate at the seams).
btw, unlinke >p<'s, there will only be one cabinet per screen. So I'm hoping to get allow more detail at a good refresh rate.
but the main part of the FE will not really be the 3d itself... but other fun parts...
again, you know the release rate I had with jfront before I started on the 3d stuff....