ESPUSB - Chat about the software components

User avatar
By nlhommet
#54435 Hi Charles, I'm Nicolas and new here. Exploring your different projects in the last months eventually convinced me to dig into ESP and electronics, so thanks for that :)
cnlohr wrote:... When the ESP receives a pin-change interrupt, it is cycle-precise all the way down to 160MHz!...

Even if that didn't solve your problem, could you elaborate about this particular aspect ? Does it mean that an interrupt routine can start to execute the cycle immediately following a pin change (or at least after the same fixed number of cycles) ?

I'd like to be able to measure/count the exact number of cycles betwen two pulses (a few microseconds distant, applied on same or different input pin) as precisely as possible. Is there any special interrupts setup to properly achieve this ? (If you, or someone else here, could provide a simple example..)
User avatar
By cnlohr
#54480 It is a fixed length afterwards. It's a bunch of cycles. I figured out a way to whittle it down in my fullspeed branch on github, but, the problem is still that I can't read the GPIO fast enough.

I did have to do a bunch of tricks to get it precise, so, it's annoying, but totally do-able.
User avatar
By RichardS
#54567 @CNLOHR do you have an ESP32 to play with yet?

RichardS
User avatar
By cnlohr
#54574 I do not, however, I am still hoping someone help make this bootloader thing happen since I am totally lost on it... and before I play with the 32, I really really really want to figure out how to get incoming wireless packets to raise an interrupt so I can see if it's possible to get precise timestamps on them.