3ds open bor base build

[Goldfire345]

Hello I'm Goldfire345 i didn't know how else to directly contact you (well git hub doesn't exactly have dms as far as i know)

While testing out modules began to grow curious about and that is... which main build of open bor is the 3ds port based on?

Some modules ask for specific numbers builds like 3.0 6913 (just an example naming scheme) so knowing which build its based on would help a lot with testing.

[MrHuu]

This is based on OpenBOR v3.0 build 6391.

But in hindsight i noticed i've rebased on the wrong commit, so this is ahead of build 6391 by about 20 commits.

[Goldfire345]

Thanks for the reply and thanks for the comment though i have a new question Do you plan on releasing a build that eventually incorporates the stereoscopic 3d functions? Has anyone in the community asked about that being incorporated?

Someone asked me that on Chrono crash when i made a thread about modules for this build. Before that they didn't really know about it.

[MrHuu]

Thanks for the reply and thanks for the comment though i have a new question Do you plan on releasing a build that eventually incorporates the stereoscopic 3d functions? Has anyone in the community asked about that being incorporated?

Short answer would be; no.

This 3DS port uses SDL 1.2 for 3ds which does not take any 3d or wide screen (800x240) in account. To be able to render in 3d, rendering has to be rewritten. Either updating SDL, or using libctru / citro3d directly would be required.

If such renderer is available, there would still be the issue of the content (.pak) not taking 3d in account. Which requires either the .pak to be modified to display the same sprite twice with a offset. Or let openbor handle this, but the required depth would be rather specific to each .pak and it's sprites. This probably doesn't get the expected results with various .paks.

While i do find this interesting, i don't plan on working on it anytime soon.

[Oldyzv2]

Greetings MrHuu - i am an openbor game developer - i left you a private message at the Gbatemp.net forums - we really need to discuss issues i noticed & possible plans for 3DS focused games, i hope you can help make things possible.

here is simple 3D openbor demo i made : https://www.youtube.com/watch?v=bJyc6LffR60

[MrHuu]

I assume, to achieve this, you've made changes to both the engine and module? Since it requires rendering to multiple framebuffers. Can you elaborate on this?

Looks great btw.

[Oldyzv2]

No engine modications & nothing fancy to the module, just edited the game's resolution to double the height & edited the level to feature double layers, double panel and manually made then offsets for the effect - character sprites would need a lot of scripting to make them work, & a critical modification or feature needs to be made to the engine in order to fix a big problem.

i have provided videos and details at chronocrash.

[Oldyzv2]

here is more info from my old thread, https://chronocrash.com/forum/threads/3d-conversion-of-modules-stereoscopic-wip.5298/ also i failed to mention - its possible to program levels for stereo projection as side by side, but sprite contamination is way more likely if there is no way to "mask" or filtering them.

over under is the easiest method & using a "screen" programmers can project the bottom screen to the right or left instead... also i have a weird idea... Its seems that the Nintendo 3ds had the ability to feature cartridges with enhancements, kind of like the FX chip on the SNES - do you think that an "special" Openbor cartridge OR coding the engine as an enhanced cartridge would expand/enhance things?

just as a hyphotesis - take a game that is known to have had special things, emulate it & add the engine code to it (customized to take advantage) ...

[MrHuu]

Thank you for elaborating on the subject. In most cases it does seem to require quite a few modifications to both engine and module.

I don't think SDL would be capable to achieve these results, but with when rendering with citro3d or picaGL rendering to multiple surfaces should definitely be a viable option.

Unfortunately, my day-job prevents me from spending much time learning how to achieve this for now.