T3i - Enabling RawVideo/MLV_Lite module cause 'Busy - Please Wait' message on display

[mungewell]

T3i running 'magiclantern.2025-06-20.600D.102'

When the RawVideo mode is turned on, the 'Busy - Please Wait' message appears and the display refresh becomes very slow.

[reticulatedpines]

Does this happen in LV, or only in some modes (menus, etc)?

[WalterSchulz]

Looks like a duplicate: #183

[mungewell]

Message appears in (rather under) the Live view overlays. Only when Raw video is set to be 'ON'.

The 'SRM' setting make no difference, On or Off still results in message.

[reticulatedpines]

Did you restart the cam after changing SRM usage to off?

[mungewell]

Ah, yes... I need to remember to reboot the camera.

That said, the error message does NOT go away (with either SRM on or off), but the display refresh rate is much better with SRM on (makes sense).

[reticulatedpines]

Unless Walter can repro this, there is nothing I can do. I suspect it's related to your settings, and SRM usage. The "Busy - Please Wait" message is a symptom of conflicting SRM usage.

[WalterSchulz]

Oh, this is gonna be fun!

====
Started cam in movie mode with ML defaults
Enabled MLV_LITE module
Restarted cam
-> No error message in liveview
Enabled "RAW video" in Movie tab
Left ML menu
-> "Busy....Please wait." message appears in upper part of LV screen
Accessed "RAW video" submenu and set "Use SRM memory OFF"
Restarted cam
-> No error message in LV

(Now the fun part)
Access Movie tab and turn "RAW video OFF" and ON again
-> After entering LV message "Busy....Please wait." appears again!

Restarting cam and message doesn't appear again.

(Also bringing up again, that this is an issue with Midsummer Edition only. It works with "SRM memory ON" in legacy builds)

[reticulatedpines]

Tried a crop_rec derived old build? I suspect it'll have a similar problem, for the same reason.

[WalterSchulz]

Can't! crop_rec_4k_mlv_snd was exclusive to Digic 5.

[reticulatedpines]

That could explain it. Ugh I hate our past decisions. The branch was never exclusive to D5. I guess that means we chose only to publish builds with it to D5?

Can you check behaviour of crop_rec_4k_mlv_snd on some D5 cam, to confirm/deny existence of this problem there?

[reticulatedpines]

For future ref, code around this commit changes SRM a lot: reticulatedpines/magiclantern_hg_02@4d4ba86

Which was taken into new repo for the other stuff in crop_rec that we actually want.

Along with that, we changed the strategy for SRM usage, which is my guess as to what is causing this problem.

[WalterSchulz]

Tested with legacy magiclantern-crop_rec_4k.2018Jul22.650D104.
Error not reproducible.

[reticulatedpines]

Thanks. That defines the scope fairly well. What's your call on whether this and #183 are dupes?

[WalterSchulz]

Same issue.
We have a new side effect documented here when SRM memory is toggled OFF-ON-OFF, though.

I think I was unable to reproduce error on other D4 cams.
EDIT: Confirmed. No error with 60D and Midsummer Edition.

[reticulatedpines]

Alright. It's something to investigate the next time anyone looks at fixing the mess that is our SRM code.