| Home | Forums | What's new | Resources | |
| making a 3D racing game for SEGA Saturn |
| RockinB - Feb 23, 2005 |
| Prev | 1 | 2 | 3 |
 |
RockinB | Apr 17, 2005 | ||
Nope. I've been working on a re-implementation of a very popular and very funny simple little flash game for SEGA Saturn. (supposed to be a short term project) Last week I got all sounds, sprites and backgrounds working on Saturn. To ease game-logic debugging, I decided to add the option to use SDL and applied a whole restructuring to ease programming of the game(with other flash games in mind). Well, the SDL version on my iBook is almost at the point where the Saturn version is. I like the idea to have it running on Saturn, GP32, Dreamcast, Windows, MacOS X. But as I ripped all graphics and sounds from the original game(what a work!), I will definitely get problems when I release it without permission. Hopefully I get permission by the author....I'll ask him later when it's almost done. So maybe I'll contribute a GP32 port to GBAX 2005... and a Dreamcast port to Dream Coding Grand Prix 2005... and maybe this will focus some people to the original SEGA Saturn game. :cheers | ||||
|
RockinB | Apr 22, 2005 | ||
Will have to wait untill I get permission by the author. Can't say if I continue with other flash games. I made some cool Gimp scripts such that I could easily create GFX sets for 3 different screen resolutions, each with and without alpha blending. All 6 variants are working (SDL). This has been done to adapt GFX quallity for non Saturn platforms. Hopefully I can finish the MacOS X SDL version in some days. Then I will do the Windows->Saturn->GP32->Dreamcast ports. | ||||
|
RockinB | Apr 23, 2005 | |
| I asked for release permission today. Wish me luck, I'm not very optimistic. This isn't the first time I asked someone on such a topic...without success. | |||
|
RockinB | Apr 27, 2005 | |
| The flash game remake is going on very fine. I do now have SDL versions for Windows and MacOS X. Maybe some trustworthy people want to betatest the stuff? PM me your email-address and I send you a copy. But for now you have to agree not to give the game or any information about it away to anyone. As far as I know the original is still used commercially... | |||
 |
vbt | Jul 31, 2005 | |
| I moved out of the track :| | |||
|
RockinB | Jul 31, 2005 | ||
Of course, collision is disabled. When I paused the work, I was going to verify and test collision detection. (If collision is activated, you can't drive across some some of the track's faces and I didn't know why.) The track is too small and tight, too. What might look like a collision is just the downforce, because the track border are faces of high slope. | ||||
|
vbt | Jul 31, 2005 | ||
Anyway, I found it faster than the original sample, I've added it to the Saturn program page  I have the full week to try to compile the sources and test it  | ||||
|
RockinB | Jul 31, 2005 | ||
This is just unfinished WIP code. <!--QuoteBegin-v bt@Sun, 2005-07-31 @ 10:16 PM I have the full week to try to compile the sources and test it [post=137658]Quoted post[/post] [/quote] Here is a SaturnOrbit version that I made for Timothy: [attachmentid=1397] Note that RB_SaturnMAP (C:/SaturnOrbit/RB/RB_SaturnMAP/) is used for track creation. It's feed with 3D data (.MDL C source files) made with 3DEditor. Last thing added here was a displacement/correction of the track borders, for the case that the track border faces are smaller than their neighboring track faces (I don't expect anyone to understand what I mean, hehe). Note: The track must be oriented correctly according to Saturn coordinate system (-y axis is height) when feed into RB_SaturnMAP. That involved some try and error for me. RB_SaturnMAP can output the track dimensions, so you see if somethings wrong there. things to do: - verify and debug collision detection - find out an appropriate size of the track (too small currently) - the ground (rotation background) is not 100% in sync with polygons, when turning around - car display - camera views - fix 2 player background display - view frustrum culling (should be easy, partially implemented) - a much better example track AFAIR only the collision response is commented out. The collision detection takes place and result is displayed as text. I think to debug collision detection, we should have a talk about the principle used later. I'm very thankful for any progress on this project | ||||
|
RockinB | Aug 1, 2005 | |
| Some of the runtime code (track polygon drawing) is located in the RB library, file is called MAP_run.c or similar. | |||
|
RockinB | Feb 3, 2007 | |
| I took the racing game project as a base for my entry to the SEGA Saturn Coding Contest (C4 - 2006). I've applied a lot of small fixes, improvements and quick hacks. To both, the racing game engine as well as the polygon space partitioning tool RB_SaturnMAP, included in RB library. The 3d landscape is actually just a new and much better track. It is a flying (and not a racing) demo, because I disabled ground following and collision detection, these parts are still too buggy. My blender SGL export plugin was used to export the track. Here are some messy screenshots taken with yabause. It looks a bit better on real Saturn: | |||
|
RockinB | Feb 20, 2007 | |
| A lot of textures have been fixed, clipping is reduced a lot, some drawing bugs have been removed: | |||
 |
Runik | Feb 20, 2007 | |
| Are you sure about the linked file ? It points to "yabause_some_rock_polys_not_displayed.zip" ... | |||
|
RockinB | Feb 20, 2007 | ||
It should be a newer, bugfixed and improved version of my C4-2006 entry, compiled on sunday. I took it as a bug report for yabause, so developers can reproduce the problem. It works both in yabause and on real Saturn, in yabause there are only 2-3 poylgons of the rock not drawn, of about 5500 polygons total. | ||||
|
vbt | Feb 21, 2007 | |
| 2 shots from girigiri. | |||
|
Runik | Feb 22, 2007 | |
| Nice ! Definitely better  | |||
|
RockinB | Oct 21, 2007 | |
| Today when I woke up in bed, I suddenly had a the solution in mind, the reason why the ground following/collision detection is buggy. It's also the reason for the polygon clipping noticed in my C4-2006 entry. So I make a note about that here, because I use to forget such things: The polygon space partitioning algorithm currently assigns a polygon to a rectangular space area, if it's mean or middle point (can be selected) lies inside that reactangular area. This way, a polygon is assigned to one and only one rectangular area. This means, that it cannot be guaranteed, that the area got all polygons assigned that are physically present there. So there are holes in space both, in drawing and collision detection. In game, there is always a neighborhood of areas processed, which lowers the problem, but the problem still remains. Solution: assign a polygon to all rectangular areas, that cover a part of the polygon, even if it's only a small part. Note #2: mirror track faces at the border | |||
|
vbt | Oct 22, 2007 | ||
Can't wait to see the results | ||||
|
RockinB | Oct 23, 2007 | |
| I started making a todo remember list for all of my projects. it happens from time to time, that I got a clue or solution for a previously unresolved issue. When I write it down, the knowledge is not lost and I got it out of my head, for a while. The list is already quite long.. | |||
|
vbt | Oct 27, 2007 | ||
Sure, it's hard to make/finish projects with lack of time  . I wanted to port a simple game this year and do a new SMS Plus release but it's already time for the C4 contest, so I'll have to do some choices (delay Raze port). | ||||
| Prev | 1 | 2 | 3 |