Eachine CG023 protocols - what are they?

The weird thing is at Devo 10 is working with 4100 period

At least I finished to install the environment

Why this error?

Trying to build latest DeviationTx Repo, or Goebish Repo.
Using GNU Tools ARM Embedded 4.8 2014q1

Right! I installed latest Pyton
Thanks!!!!

How many could be the shortest safe period value to test?
We found with 2050 the delays are little noticeable, but what lowest value we could test?
Sorry about hammering the thread, but it's weird in Devo 10 is working perfect at stock packet period, and with the same code the 7e needs shorter period...

PS. Finally my environment is building perfect . Thanks guys!

Hi, comrades! I have a new interesting idea and new test results on yd-829.

I started with idea - may be the problem is not in devo7e itself but in modules difference between those I have there in devo7e and devo10?.. In 7e I have that '+PA' module from BG. And in 10 I have module from wltoys stock TX. So the last is (or I think it to be) without PA.

So in 7e I have 'has_pa-nrf24l01 = 1' setting on, and in 10 it is commented out. And I really had problems previously if it is set other way. But for this test I reversed this setting on both devo's. Also I installed 4100 firmware on both to be exactly sure to receive highest delay issue.

The result on devo10 - nothing changed, no delay at all. On 7e with 4100 - it is nearly ok, delay is very low, similar to delay with 2050 firmware or even less. With 2050 firmware I do not feel delay at all.

So... Certainly it is not a solution. As has_pa-nrf24l01 setting do have some meaning, I believe, and should not be altered for this reason. But there can be some incompatibility with protocol code (or its chip emulation level) and PA handling code.

I expect this can be helpful.

goebish wrote: @Durete, it should be safe to go as low as 1300, that's what the green CX-10 protocol uses, but this is not a proper solution, it should work with 4100.

Yep, I agree with you. I don't feel confident knowing something weird is going
I will build with 1300 period, only for testing...

With 1300 period, as expected the delay is almost the same as with 2050 period, or very slightly noticeable.

Hi guys. I know it's a little bit late as this has already been tested properly but just wanted to report that I managed to give it a go this morning and my finding are essentially the same as what has already been reported. 4100 was kinda flyable but it was weird to control the quad. This delay feels as if there is some communication interruption or something. Like as if not all the packets are being received/sent or just being lost. What I mean is that when I moved the stick in one direction the quad would start flying in that direction pretty much without delay but there would be some jerkiness in the movement meaning that instead of going smoothly in one direction it would start smoothly and then suddenly accelarate (jump). But I guess we can say this is just a delay.

I thought that with 2100 it was flying quite good actually. Didn't have much time to test it but it felt almost OK. If I didn't have an original radio and wasn't aware of the issue I could think this was how the quad flew. Didn't have time to tes 2050 or anything below that.

What I also noticed (this probably has nothing to do with the issue discussed here and was just a coincidence but I thought I'd mention it) that the quad seem to be drifting in one direction quite badly on the 4100 and noticably less at 2100 even though I haven't changed the trims or anything. I didn't notice any drifitng with stock controller. It proably just needs some calibration though.

I don't see any drifting issue. The quad drift sometimes when you release the stick but is because the delayed communication.

Repeated my own tests with disabling PA on Devo7e - and can't reproduced the previous result. Now delay is the same no matter this setting. Meanwhile on Devo10 there is still no delay.

I was going to post when I saw your post.
I repeat your test some minutes ago with my own build (4100 period and PA disabled) and didn't notice any difference.
The same delay as with PA enabled.
My 7e has the same PA+ module from Banggood as your.

goebish wrote: Thanks czajunia, which Devo model and nRF24l01 board do you have ?

Sorry, forgot to mention this. I have 7E and the PA+ module from Banggood.

More tests...
Since I am testing 2 protocols under development, I compiled my own build using latest DeviationTX repo + Goebish YD-829 + First test version for Hontai from Hexfet.
This time I compiled using 4100 period, as stock TX.
I found using Fixed ID we have the delay at commands always, but with blank Fixed ID, and after some power resets (TX and Quadcopter) sometimes I managed to bind/fly almost perfect.
If I'm not wrong, when we don't have Fixed Id, the TX is calculated using some unique internal serial number from TX.
Could be the delay issue related with the TX ID?
@ Goebish, since I share my SPI captures, and we have my TX id and RF channels related, please could you build a test version using as TX id and RF channels my owns?
Only for testing purpose, because I have the hunch this weird issue is related with this values.
Thanks!

Just tested. You were right No change at all.
I'm run out of ideas
I noticed something that in previous tests I didn't recall. You can noticed the delay even at floor, before take off, making fast changes at throttle stick. But when you are in the air and past some time I have the sensation the delays are gone... I don't know...