ducalex

> Oh, and it looks like there's an issue with the audio sampling rate. The audio task's queuing limits the main thread to 50 FPS (`xQueueSend(task->queue, msg, portMAX_DELAY) == pdTRUE;`)...

> I think the issue might be: `#define AUDIO_BUFFER_LENGTH (AUDIO_SAMPLE_RATE / 50 + 1)` (`shared.h`) vs `.frameTime = 1000000 / 60` (`rg_system.c`) Okay in https://github.com/ducalex/retro-go/commit/a67f5eb9d15c417af5104817615d5a5c71e8ad2d I've now made it so...

I've looked at your repo and doh I just removed double audio buffering in SNES, I guess I'll have to add it back (or wait for your code)! > Maybe...

> Are you trying to adapt the generated samples depending on how long it took to render the frame? I don't understand what you are doing with the following: >...

> BTW: How fast is Snes9x on the ESP32? On the ESP32-S3 I get 45 fps (that is 22 rendered frames per second) for Super Mario World's intro. As of...

> Because I have experienced issues with only portions of the screen being sent (but this might have been me not yet fully understanding all paths in rg_display.c). I think...

> That being said, Snes9x with Retro-Go runs great on the P4, I uploaded a short video [here](https://www.youtube.com/watch?v=m7ioy6VdjsU). (I hope I pronounced your name correctly, sorry if I did not)....

I don't think generating releases for each PR makes a lot of sense. I don't use a PR workflow myself when I work on retro-go, so it would only be...

The fact that the display is sometimes corrupted might suggest a bus conflict maybe? It's hard to say what why the sd card init fails with 0x106 otherwise. You can...

I'm not anything obviously wrong, you're not using reserved pins and there doesn't seem to be any conflict. Is the image still distorted when you disconnect the SD Card board...