Skip to main content


A couple games are getting so close to playable!

I’m going to try running
some FPGA simulations in GHDL for some basic test ROMs, and compare output logs to the Uxn software emulator. The simulator runs many thousands of times slower than the real thing, and is a pain to set up, but I think I can no longer guess my way through it. Too often I’m trying random ideas, waiting 10 mins for a build to finish, and then seeing if it worked on the Pocket 😅

#AnaloguePocket #uxn #varvara

in reply to tsalvo

Wow, that's pretty damn close. How does it fare with the screen device examples?
in reply to Devine Lu Linvega

@neauoire I’ve been focusing on the screen.blending ROM logs, and I think I’m in a state where the logs are matching up with what I’m writing to VRAM, at least for 1bpp, but I have some visibly missing regions onscreen on some ROMs. I think I have some issues with the double buffering (maybe not swapping buffers at the right time), or the drawing command queue that I’ve set up for pixel and fill operations.

screen.rom looks perfect except for 2bpp though!

in reply to tsalvo

wow, okay yeah, that looks perfect for 1bpp. Tetris uses some auto values and also sprite flipping, to test those I think screen.auto.tal might be a good test.
in reply to Devine Lu Linvega

@neauoire screen.auto looks like I’ve got the right idea, but it’s not colored the same way as it is in the software emulator (not sure - maybe a 2bpp issue on my end), but I think I’ll try comparing logs for that rom next too.
in reply to tsalvo

Oh wow, that means that the auto byte is working properly. That's usually what trips people when implementing the screen. The color is issue just a 2bpp thing, otherwise it's perfect.
This entry was edited (8 months ago)

Lo, thar be cookies on this site to keep track of your login. By clicking 'okay', you are CONSENTING to this.