#412 Open

Bitbucket cannot automatically merge this request.

The commits that make up this pull request have been removed.

Bitbucket cannot automatically merge this request due to conflicts.

Review the conflicts on the Overview tab. You can then either decline the request or merge it manually on your local system using the following commands:

hg update 7a3b5fa3f4c6
# Note: This will create a new head!
hg merge f7947b627e33
hg commit -m 'Merged in iso-research (pull request #412)'
  1. Alex

From www.magiclantern.fm/forum/index.php?topic=10111

  • modified adtg_gui to show DIGIC registers and some more advanced options
  • iso_regs (for 5D3 only)
  • raw_diag (portable)

All 3 modules are compatible with both 5D3 1.1.3 and 1.2.3.

The only issue may be with the modified adtg_gui (the one from unified is a little more lightweight, this one is heavy on both RAM and the number of menu items).

Comments (15)

  1. Audionut

    With adtg_gui loaded (no raw_rec or mlv), zooming 5x in photo LV causes weird display. 10x causes LV freeze.

    Taking the first shot in raw_diag "compare 2 shots" or "compare 2 shots HL", and then powering off camera before second shot, causes the module to show the graph after the first shot, and then break raw_diag completely where none of the options work. Restart camera enables all other options again, but attempting to compare 2 shots, causes the issue again.

    Deleting RAWSAMPL.DAT and restarting the camera, solves the issue.

  2. Audionut

    Ran into an issue yesterday, where the iso_regs module was loaded, but not displaying in the debug menu. I think it happened after waking the camera from sleep.

    My build is 11 days old though (needed to check for show stopper bugs before shoot), so I'll update tomorrow and see if I can find the same bug.

  3. Audionut

    The weird frames I described here: https://bitbucket.org/hudson/magic-lantern/pull-request/394/5d3-123-port-almost-ready/diff

    Can be reproduced 2 ways.

    With adtg_gui loaded, and dual_iso loaded and enabled (didn't check if the problem also occurs with dual_iso only loaded, and not enabled).
    With adtg_gui loaded, and digic registers enabled.

    In both situations, no registers had been changed. These weird frames occur on the second, and subsequent images, in high speed continuous shooting mode (didn't check other modes). This problem effects DR calculations also.


    Raw files with this issue can be found at the 5D3.123 pull request.

    Another issue that is much less frequent, and as yet, I haven't been able to find a way to consistently reproduce, is the same image gets written to CR2, on subsequent photos. That is, take photo, problem occurs, all subsequent photos are the same image, with increasing WB. This issue occurred after taking a screenshot of a weird frame, but I haven't done sufficient testing, to confirm any actual relationship with the screenshot function.




    And 2 of the resulting CR2s.



    Note: _46A4032.CR2 is an actual valid image. And _46A4034.CR2 is a problem frame, trying to make me believe I can freeze time :P

    I am at changeset: d76742634eee+ with 5D3.123 and iso_research branches.

  4. Audionut

    Starting the camera in movie mode LV with either raw video module loaded, results in the debug menu modules not being shown in the menu (adtg_gui, raw_diag). The modules are still active in their last state, ie: raw_diag continues to function, they simply do not display in the menu.

  5. Audionut

    Getting compile errors in adtg_gui.

    adtg_gui.c: In function 'log_iso_regs':
    adtg_gui.c:1070:9: error: implicit declaration of function 'is_native_movie_mode' [-Werror=implicit-function-declaration]
             len += snprintf(msg+len, size-len, "lv=%d zoom=%d mv=%d res=%d crop=%d ", lv, lv_dispsize, is_movie_mode(), is_native_movie_mode() ? video_mode_resolution : -1, is_native_movie_mode() ? video_mode_crop : -1);

    I assume you removed 'is_native_movie_mode' in some of the latest changes to unified.

  6. Audionut

    On 6D.116 (didn't check 113) merging to 497b816 is fine. After this, adtg_gui and dual_iso no longer perform any actions. I can change registers and the patch manager seems to show these changes, but nothing actually changes except the UI.

    I seem to recall having issues with patch manager previously, but didn't have the time to debug back then, and obviously forgot to revisit the problem.

      1. Maqs

        Just double-checked the stubs and they should be fine. Strings match and the functions are similar to those for 5D3.113, so I don't know why this doesn't work. I'll try to look into it, but it may take some time, as I'm really busy at work atm.

        1. Audionut

          Revisiting this. Enabling adtg_gui and taking a photo results in card led remaining active. Pressing buttons on rear of camera enables backlight of LCD. Requires battery pull to restore functionality.

          6D_iso-research.png 6D_iso-research_1.png

            1. Audionut

              Before. It works as expected now, although those two entries still display red.

              "Cache is locked down (not exactly clean)."

  7. Audionut

    Setup the windows 10 environment, cloned unified, updated to this branch.

    '[ CC ] util.o
    [ LD ] magiclantern
    installer.o: In function copy_and_restart': /home/main/magic-lantern/installer/6D.116/../../src/installer.c:246: undefined reference tosync_caches'
    fio-ml.o: In function FIO_WriteFile': /home/main/magic-lantern/installer/6D.116/../../src/fio-ml.c:554: undefined reference tosync_caches'
    ../../src/Makefile.src:198: recipe for target ‘magiclantern' failedmake[1]: *** [magiclantern] Error 1make[1]: Leaving directory '/home/main/magic-lantern/installer/6D.116'../../platform/Makefile.platform.extras:53: recipe for target 'installer_check' failedmake: *** [installer_check] Error 2’