Closed daonguyen207 closed 18 hours ago
Hello splitting in 3 the 1.25us signal is the common approach used in numerous drivers. I drive my Panels with the same timing and never any issues. Artifacrs usually come from the setup for instance it's better to power the board independently from the esp32. Can you tell me more about your setup ?
thanks for your library, however after some testing i determined the cause of the noise comes from incorrect bit timing. can you provide a version with 0.25 anhd 1 approach?
I am sorry for that but you are the first one with the timing being an issue. Can i ask you which leds you are using ? because in the ws2812 i know the timing is not .25 and 1 can you share the data sheet of those ?
This is the type of led I use. I have checked and measured the control signal from the available boards. They are 0.3+0.3 for bit0 and 0.5+0.3 for bit1, strangely 0.25+1 still gives good results, but 0.4+0.8 like the ws2812 strip is noisy. Can you provide some instructions to be able to customize the DMA buffer for different types of leds?
for the virtual driver is not that easy for get close to .25 1 and keep the 8:1 ratio i would need to push the driver at 25.6 Mhz which willnot work on a regular esp32 because above 20mhz it will skip bits. to keep the master clock speed under 20Mhz I would have to reduce the number of outputs per 74h595 to 6 instead of 8 so you would be able to control 15*6=90 strips in parallel
So are 0.3+0.3 and 0.5+0.3 (or 0.6 + 0.3) achievable?
it's feasible but will require to rewrite part of the driver
My led strip sometimes has noise, I checked the control signal and found that with bit0, logic 1 is 0.43us and logic0 is 0.8us. While according to the documentation it is 0.25us and 1us. Is it possible to make the signal more accurate?