Wrong Came Atomo button counter sequence during sending from Flipper

[RekordNadoev]

Describe the bug.

Version: Unleashed dev branch, commit 462ab25

During continuous sending Came Atomo, button counter changes from ~ 0x4Ch to 0x54h
Remote is working, gates opening but:
To program new remote to receiver with the old one, button counter should tick from 1 to 127 and begin from 0 with the same Pcl_Cnt. After this "zero-cross event" the receiver should switch to programming mode.

Reproduction

Select Menu => Sub-GHz => Saved => select any Came Atomo remote => Emulate
Hold central button to continuous send
Pcl_Cnt is static
Btn_Cnt is changing from ~ 0x4Ch to 0x54h (instead of changing from 0x0h to 0x7Fh and begin from 0x0h with the same Pcl_Cnt)

Target

No response

Logs

No response

Anything else?

No response

[xMasterX]

Hello! Can you share full RAW recording of the original remote so i can check btn counter?

[RekordNadoev]

I'll send it later. The original remote is at my home now.
I checked the counters with my second Flipper.

[RekordNadoev]

There is RAW record from the original fob in attach. (.sub renamed to .txt)
Came_Atomo_RAW-2024_11_08-21_29_01.txt

[xMasterX]

There is RAW record from the original fob in attach. (.sub renamed to .txt) Came_Atomo_RAW-2024_11_08-21_29_01.txt

Thx, will work on that issue next days

[xMasterX]

Yes, the issue is present, its possible to generate full sequence but it will take big buffer space and will require long button hold, since current implementation in subghz doesnt actually support sending different codes while holding a button, so for system like that - we generate full or partial sequence and put it all in buffer, while button is held it being transmitted as it was one code (but big one)
To fix this condition there are three ways i see,
1 - generate whole 127 codes and put them all in buffer (we may have not enough ram for that) (and zero cross event wont happen if you release button too early)
2 - start all parcels from zero? should be already done, may be not resetting properly, will check later
3 - use only specific limited number of values to trick receiver that it had got all signals but missed some in between (like skip big parts of the counter, send sequence like 0 5 20 40 50 90 100 0)

[RekordNadoev]

I think third way is the best solution.
Also I would try something like "10,50,100,125,126,127,0,1,2"

[xMasterX]

I think third way is the best solution. Also I would try something like "10,50,100,125,126,127,0,1,2"

Ill make some changes after release in next dev builds, and will ask you to test, thats fine?

[RekordNadoev]

Shure.