What is the FPS?

Between 3 and 6 fps.
At least for the screenshot on the first post in this topic.

thanks!

Keeping in mind that that particular scene consists of around 1500 triangles, it is incredibly impressive.

Less complex scenes would be faster - mind you full screen rendering takes a fair toll performance wise.

Where can I find the .obj converter?

I don't see anything in any of the posts that suggests that there is such a converter; did I misread?

Oops my bad.

I must be pretty tired.

There is a .obj converter to asm file readable by the lib in the tools/ folder (check mdlconv). Example on how to use it are mainly in the build_model.bat file. There is also a texture converter (build_texture.bat) and an animated model converter which is such a pain to make it works that you should probably forget it.

Thank you.

Is this still being worked on TheMachine02?

Do you have any interest in turning this into a C library of some kind?

I blame awesome tr1p1ea for getting me out of my books and reviving a small spark to get it run.

Alright, seems like this is better than the PS1 Classic. Already has more time put into it xD

Looks great
At this framerate, a FPS game could be playable, but adding other scaled, moving 3D objects, and visual effects (bullets, explosions) would take a toll on the framerate.

Running at a bit over 5fps?! That is awesome man!!! The best thing ever on a calculator.

Glad I helped with the inspiration lol.

My dream since seeing this project has been to make a graphically primitive version of Temple Run. Does the FPS correlate to the poly count, or does it only make a small difference? If I was only refreshing a single poly, what do you think the framerate would be?

What about SM64? You could test it with the front of the castle. A lot less polygons to render, just make sure you remove the trees for the first test. I'd love to see the front of the castle as a demo!

The framerate is strictly dependent from the polygon count and the vertex count, since it is rasterization. The more works you send down the pipeline, the slower it is. Texturing take away quite a bit of speed too.

How do I compile this and run it?

Speaking from someone with almost no knowledge in the field, I often wonder if this level of optimization is used when writing graphics engines for current-gen computers with nearly millions of times as much power as this calculator. On a calculator, you have absolutely no choice but to make your code as small and as fast as possible if you want something usable, as I can see from this program. Poking through the code and reading this thread, it's very clear to me that a great deal of time was devoted to simply optimizing. Does this level of clock cycle milking happen when people develop, say, Unreal engine? I feel like the sheer magnitude of power available to us today would inadvertently lead us to search for optimizations less often, as the bar that must be cleared (something usable) can be written by a fifth grader when you're running on a dedicated GPU at 50+ TFLOPS.