I have a question. In a project with the ESP32, I use both cores, and I need both cores to read a GPIO ADC quickly. With a semaphore system, they could synchronize to read the ADC without any problem. However, for my code, I need both cores to read the same signal quickly, without waiting times. I thought about connecting the same signal to two different ADCs, which would solve the problem by eliminating the need to use semaphores. However, is there any problem if I do this? Due to impedance coupling, noise, or something similar?
I'm not an ESP32 user so can't advise from that perspective.
There is a risk that both ADCs give slightly different readings which might or might not be important.
ESP32 has beally bad ADCs, but you will get about the same results with both cores.
In any case, you have to wait while ADC doing that. So you could read the ADC by one core and use the same result in both.
Correct me if I'm off base here, but the ESP32 processors only have one ADC, no? Two cpu cores, one set of hardware peripherals? Page 2, here:
So then, if code in both cores attempt to read the same ADC input, there must be some arbitration mechanism that provides a reading to one core, then a reading to the other core, causing one core to 'stall' until a reading is available? Or is the hardware dedicated to one core only, and the other simply a processing engine?