You mean without these delays in the code, right?
Yes. No delays. Before I shared the driver to anyone for testing, I had none of the delays in the driver because there was no need to have it.
I assume that the upload takes at least double the time with these in place, correct?
It takes longer with .002 as the sleep delay, but probably not double. With a delay of .003 it is quite a bit longer. Unbearably long.
I'm not sure what program that came from, but it doesn't have the extra information that portmon gives us about how the serial port was opened. However, from the look of that, I would assume that the software is writing the whole block in one go, not doing byte-at-a-time.
Correct. It is not from Portmon. I bought a program that will run on a 64-bit Windows computer. My 32-bit laptop that I can use Portmon on is currently unavailable.
Given our relationship with Retevis, maybe we can get some information from them about any restrictions the radios may have in terms of digesting the data stream?
I can try.
Have you tried doing something like writing 8 bytes at a time, with a sleep in between or something like that?
I will give this a try.
I think that's a good first step. If we can knock it down to a delay every N bytes that'll be a huge improvement.
--Dan
Jim