Menu

Show posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Show posts Menu

Messages - JohanJ

#1
Camera-specific Development / Re: ML on EOS-M2
January 08, 2019, 11:06:07 AM
Quote from: dfort on January 07, 2019, 12:29:43 AM
I re-enabled raw histogram because several modules weren't building without that feature enabled. In addition, I had LiveView screen freezes even with the non-raw histogram.
LV did not freeze with (RAW) histogram enabled in photo mode for the nightly builds between 14th of Dec 2018 and at least 27 of Dec 2018, but it was freezing for the ISO research builds during this time interval. Yesterday I was loading the latest nightly build from January 7th and now LV is dead again with RAW histograms enabled (I cannot tell for the builds between 27/12 and yesterday, I was off there).

@dfort, you stated here
Quote from: dfort on December 21, 2018, 08:51:57 AM
Ok--so it looks like the "C" build is the most stable ..... This is actually from Danne's crop_rec_4k_mlv_snd_isogain_1x3_presets_EOSM2 branch with some of the lines in raw.c commented out.
C build was from magiclantern-Nightly.2018Dec14.EOSM210 and LV was stable with RAW histogram. So maybe some lines in raw.c had been activated again by merging in different new developments?
#2
Scripting Corner / Re: Lua Scripting (lua.mo)
January 02, 2019, 06:45:57 PM
Test results for 100D using build lua_fix.2018Dec23.100D101


  • API_TEST.LUA went through w/o any problems. Logfiles: LUATEST.LOG
  • Stubs test in selftest.mo: 14 failed. Logfile STUBTEST.LOG
  • Memory benchmarks in bench.mo:  BENCH1_100D.JPG
All files can be downloaded from here

One strange artefact identified using long-press Q/SET button: the ML overlay symbol for long-press Q/set remains visible after Q button was released. You can see this orange little ring as a sticky symbol even when scrolling through Canon menus or even in Play mode when reviewing a picture. One has to (short) press SET again to let the symbol disappear. Two screenshots available in the download folder.
#3
Scripting Corner / Re: Lua Scripting (lua.mo)
January 01, 2019, 09:09:41 PM
Test result for 60D using build lua_fix.2018Dec23.60D111

running API_TEST.LUA ended with ERR70 crash when the script was processing movie tests. Crashed in movie mode in funktion movie.stopp. I had to pull the battery.

Logfiles:
LUATEST.LOG
CRASH00.LOG
LOG000.LOG


Stubs test in selftest.mo: 1 failed
Logfile STUBTEST.LOG

Memory benchmarks in bench.mo
BENCH1.JPG

All files can be downloaded from here:

One observation during field tests:
shortcut keys for focus patterns do not work anymore.

EDIT: Shortcut keys for focus patterns work as defined as long as back screen is off (neither Canon info screen nor leveling meter should be active). I guess that is as designed, so ignore the previous observation.

#4
Camera-specific Development / Re: ML on EOS-M2
December 29, 2018, 11:26:39 PM
Quote from: Danne on December 29, 2018, 11:21:49 PM
Did you restart camera and went into the unique setting without turning adtg_gui on? After selection turn on adtg_gui.
Ah, my bad. Was not aware of that I can enter ADTG menu w/o activation. Now it works

Edit: still the same result,only Evf tasks logged

00f00000:     8b7 ISO=100 Tv=100 Av=20 lv=0 zoom=1 mv=0 res=-1 crop=-1 task=Evf pc=46284 addr=416d949a Analog ISO (most cameras)

#5
Camera-specific Development / Re: ML on EOS-M2
December 29, 2018, 10:59:22 PM


Quote from: a1ex on December 27, 2018, 12:05:14 PM
There is another way to log photo capture alongside with LiveView - in adtg_gui, in the Advanced menu, set "Unique Key" to "Register + caller task". That way, still photo and LiveView registers will be logged as separate entities, rather than grouped together (even if both tasks will override the same register).

Caveat: you have to enable that option *before* enabling ADTG registers (i.e. it's not something you can change during a logging session). Go to the submenu first, while the stuff is still grayed out, change the unique key, then go back to enable the main menu entry.

Tried to set "Unique Key" to "Register + caller task" before doing anything else in the  ADTG Register Advanced submenu but as soon as I hit any key in Unique Key submenu I get the following yellow  message in ML bottom bar: "You can no longer change  this, sorry. Restart the camera."

Tried both dfort's latest nightly and the ISO research build. Both come up with this message.  adtg_gui.mo was the only thing active, everything else in ML menus like global draw etc. was inactivated before. Has this message been hard coded and the needed function been removed?

Sent from my SM-T719 using Tapatalk

#6
Camera-specific Development / Re: ML on EOS-M2
December 26, 2018, 12:10:47 PM
Quote from: dfort on December 25, 2018, 05:39:17 PM
I believe that Tv and Av are the property values and will vary depending on how the camera is metering at that point in time.
That's is totally correct and it was  provoked in my test scneario just to show that the value we are looking for is changing when exposure (Tv) is changing. a1ex commented here

Quote from: a1ex on December 24, 2018, 04:00:03 PM
In photo mode LiveView, Canon firmware uses exposure simulation (i.e. the actual exposure values are not necessarily the ones dialed in their menus, but something with equivalent brightness). That's why CMOS[0] appears to depend on shutter speed - because Canon firmware is actually changing ISO....
Look at the task name - "Evf" is Canon's task for LiveView on recent models. For still photos, it's "ShootCapture"....
If the camera cannot operate without returning to LiveView, you have an option to disable logging in LiveView, in the Advanced menu. That option was written specifically for the EOS M, which had the same issue.
I ran all the tests with this advanced feature "disabled logging in LiveView", still the result in ADTG.LOG shows task=EvF (=LV) only, instead of task=ShootCapture - which makes me conclude:
- LV is still active and the address we are searching gets continously overwritten by Canon firmware changing ISO due to ExposureSimulation
- the advanced Logging fetaure written for EOS M does not work as expected for EOS M2

By the way, I inactivated Exp.Sim in ML exposure menu for all testing, w/o any effect, Canon firmware seems to use it anyway.
#7
Camera-specific Development / Re: ML on EOS-M2
December 25, 2018, 04:27:08 PM
Quote from: uizin on December 25, 2018, 11:26:10 AM
I made many changes to the settings, and now i don't seem to be able to save any log to the SD... Still experimenting!

Same happend to me with build magiclantern-Nightly.2018Dec23.EOSM2103. I reinstalled the build after having low level formatted the card but no luck. ADTG GUI does not write a file anymore. No idea why!

I used a new card with the latest iso-research.2018Dec23.EOSM2103 build instead and got out a new file with all registered logged. But here we have again another value
00f00000:     803 ISO=100 Tv=200 Av=20 lv=0 zoom=1 mv=0 res=-1 crop=-1 task=Evf pc=46284 addr=416d93f0 Analog ISO (most cameras)

As you can see Tv is higher than in post #388 which leads to anoher value for CMOS[0].

In other words, all tests are still in Evf mode for LV and Canon firmware takes over. Looks like the advanced settings in ADTG_GUI.mo do not work for EOSM2.
I am pretty much stuck here. Maybe @a1ex has an idea how to continue.

Here is the entire ADTG.LOG if needed.

#8
Camera-specific Development / Re: ML on EOS-M2
December 25, 2018, 12:50:22 PM
Quote from: uizin on December 25, 2018, 11:26:10 AM
Also, I set the advanced options before enabling the recording (so the counter on the right remains to 0 uniq / 0 until I take a shoot.
Just to make it clear for me, are you in photo mode or do you take a picture in movie mode (talking about recording)?
#9
Camera-specific Development / Re: ML on EOS-M2
December 25, 2018, 11:30:51 AM
Quote from: dfort on December 25, 2018, 05:17:07 AM
Look on your SD card under ML/LOGS/ADTG.LOG

Here is the line I'm looking for (this one is from the EOSM):

Canon EOS M 2.0.2
00f00000:     803 ISO=100 Tv=160 Av=35 lv=0 zoom=1 mv=0 res=-1 crop=-1 task=ShootCapture pc=16d14 addr=4048124c Analog ISO (most cameras)

Sure, that's what I did. My point is that there is not a single row with "task=ShootCapture", task=Evf only, what ever I changed in the procedure.

Here is a part of the file matching the Analog ISO line

00f00000:     86f (was 893)      ISO=100 Tv=30 Av=20 lv=0 zoom=1 mv=0 res=-1 crop=-1 task=Evf pc=46284 addr=416d9456 Analog ISO (most cameras)


And here is the entire file.

EDIT: I changed the link to the LOG file, I had published the wrong one. Now correct values with "Logging disabled in LV ". The NEW link should work now. Also changed the extracted line in code.
#10
Camera-specific Development / Re: ML on EOS-M2
December 25, 2018, 01:26:21 AM
That is interesting. @uizin could you describe how you get these log entries? Are you testing in photo mode?

I ran the following test a couple of times but the file adtg.log did not show a single line with the task = "ShootCapture", only lots of Evf entries. My procedure was:
- work in photo mode 2, M expo, Exp.Sim off.
- load adtg_gui.mo
- Debug: ADTG Registers on
- Enter ADTG registers submenu Advanced
- Set Disable logging in Live view
- Set Auto Log registers after taking a pic
- leave ML menu
- take a picture
- message on the display that 35 registers have been logged in a .ca. 4k file

Did the same test w/ "Disable logging Off" but the result was approx. the same. What am I missing here?

Sent from my SM-T719 using Tapatalk

#11
Camera-specific Development / Re: ML on EOS-M2
December 24, 2018, 04:13:07 PM
Thanks a1ex! M2 like M is in permanent LV mode. I will check the advanced features later, now it is time for Xmas celebrations.

Seasons greatings from snowy Sweden to all of you!
#12
Camera-specific Development / Re: ML on EOS-M2
December 24, 2018, 03:32:06 PM
Quote from: dfort on December 24, 2018, 02:34:24 AM
I also did a big cleanup of the test builds so please try out the latest EOSM2 test builds at the top of my downloads page.
Using your latest build Dec 23 from last night still comes up with the same error ISOless PH err 8 . I made some more tests and had en observation which makes me wonder if we are looking at the correct register.

Some thoughts:

  • The test results in movie mode posted here where taken in Manual video mode to be able to set ISO 100. Exposure constalation  is static and the identified addres was 416d93f0.
  • In photo mode 2 my first tests in post #362 were taken in M exposure mode too, shutter speed was 1/30 sec and the result was 416d93f0, too. Today I think that was a kind of coincidence (as the stills were taken under same exposure conditions as in movie mode).
  • The photo mode result posted late yesterday in #374 were produced in Av mode. I got two different results and the first was not reproducable so I took the second one.
  • Today I repeated these tests in Av mode with the latest build and got different values again - but reproducable: 416d949a, 416d9478, 416d93f0 and others. Theses values simply changed depending on the light in the scene and the shutter speed automatically changes in Av (I did not get that yesterday night, it was simply dark)
  • Switched to M exposuer in Photo mode 2 and chose 2 sec shutter speed. Result is 416d9478 whereas at 1/30 it is 416d93f0
Meaning the CMOS[0] register changes values in depencence of the shutter speed. So, are we looking at the right register?

#13
Camera-specific Development / Re: ML on EOS-M2
December 24, 2018, 01:24:09 AM
Quote from: dfort on December 23, 2018, 09:39:44 PM
I updated the latest build with the latest adtg_gui module. See if that works. magiclantern-Nightly_with_adtg_gui.2018Dec23.EOSM2103
Ok, now photo mode and video mode show different results. A strange thing was that in photo mode I got even 2 different values, one before taking a picture and another one after taking a picture (both having dual iso activated). But ever since then the second value got kind of persistent and I cannot reproduce the initial one (though I have a screenshot, but ... never mind). So here are the new values for COMS[0]:

Photo mode 2--> Value after taking a shot -> since then kind of persistent
Evf:46284:416d949a v=2231(0x8b7) nrzi_dec=3877(0xf25)

Video mode --> this is the same as in previous tests based on iso_research build
Evf:46284:416d93f0 v=2051(0x803) nrzi_dec=4093(0xffd)
#14
Camera-specific Development / Re: ML on EOS-M2
December 23, 2018, 08:28:04 PM
Quote from: dfort on December 23, 2018, 07:08:25 PM
You don't need to load dual_iso to find the right values. Set the ISO to 100 and follow Bilal's instructions. Try using the iso-research for this. Do screenshots and post them if possible.
I did all this already following Bilal's  instructions using your the ISO research build from your repository.  The result is posted in #362. Same values for photo mode 2 as well as video mode made me suspicious whether I did right or not so I repeated a couple of times but in the end it was no difference.

Quote from: dfort on December 23, 2018, 07:08:25 PM
In the meantime I posted a new build that has PHOTO_CMOS_ISO_START and FRAME_CMOS_ISO_START set to the same address. Let's see if that clears up the ISOless PH err(?).
Unfortunately the result is the same as posted here
#15
Camera-specific Development / Re: ML on EOS-M2
December 23, 2018, 03:45:39 PM
Made another observation: the value for CMOS[0] is alway the same 416d93f0 when

  • Dual_iso.mo was not loaded
  • Dual_iso.mo was loaded but not activated
  • Dual_iso.mo was loaded and activated
Does that make sense?
#16
Camera-specific Development / Re: ML on EOS-M2
December 23, 2018, 01:58:36 PM
No more linking error but ISOless PH err(8)
Now with error code 8 instead of err(15) both in photo and video mode.
#17
Camera-specific Development / Re: ML on EOS-M2
December 23, 2018, 01:21:14 PM
Quote from: Danne on December 23, 2018, 12:41:53 PM
Could you try this module?
https://bitbucket.org/Dannephoto/magic-lantern/downloads/dual_iso.mo
I put the module into Dforts build December 21 and started in photo mode but the module fails to load with the same error as posted here. Dfort made a minor change in reply #355 which fixed it.

There was another message repeatedly printed to the console:

Black 1/5: mean too different (8152, ref 8013+-0.00)

Maybe worth to look into it too, this message does not show when going back to dfort's original DUAL_ISO.mo
#18
Camera-specific Development / Re: ML on EOS-M2
December 23, 2018, 11:44:05 AM
Quote from: dfort on December 22, 2018, 10:10:41 PM
Follow Bilal's instructions to find the addresses and I'll plug them into the code.
Well, I hope I got it right. ADTG-GUI came up with same values both for video and for foto mode:

Evf:46284:416d93f0 v=2051(0x803) nrzi_dec=4093(0xffd)

At least I can find these values in Critix post too.
#19
Camera-specific Development / Re: ML on EOS-M2
December 22, 2018, 07:57:08 PM
Thanks Walter. I can see that dfort was involved in this discussion. let's see where it brings us from there.

Sent from my SM-G930F using Tapatalk

#20
Camera-specific Development / Re: ML on EOS-M2
December 22, 2018, 07:33:45 PM
Quote from: dfort on December 22, 2018, 06:04:53 PM
In order to find the addresses missing in dual_iso we need to use the iso-research branch. Lucky for us the patch manager was recently merged into iso-research so we should be able to use the patches needed for the EOSM2. Not sure if the LiveView hack Danne posted in Reply #330 is needed. You tell me -- iso-research build posted on my downloads page.

In fact the latest change in magiclantern-Nightly.2018Dec21.EOSM2103 made it possible to load Dual ISO without crashing but it cannot be used. When activating DualISO in Expo Menu you get the following message:
ISOless PH err(15)

I will look into the iso-research build later. I would like to finish tests on the current standard build first.
#21
Camera-specific Development / Re: ML on EOS-M2
December 22, 2018, 07:23:51 PM
Some more test results based on build magiclantern-Nightly.2018Dec21.EOSM2103

I tested all functions in Expo menu in M mode, most of them even in Av & Tv as far as applicable. Everything works just fine except the following:

- Expo. Presets -> Press SET: SET activates Canon LV menu (might be as designed for M/M2?). With setting "Press Info" toggeling between two modes works fine (but you can't reach Canon Info screen anymore, of course).
- Expo.Override crashes the cam in LV if ExpSim is activated same time! To get out there I have to pull the battery! At least 4 different crash logs are written on the card.


CRASHLOG0.LOG

ASSERT: 0
at ./LvCommon/LvGainController.c:893, Evf:17bd0
lv:1 mode:3

Evf stack: 1be758 [1be8a0-1bdca0]
0xUNKNOWN  @ ca14:1be898
0xUNKNOWN  @ 36684:1be870
0x0003637C @ ff0d6258:1be858
0xUNKNOWN  @ 363ac:1be848
0xUNKNOWN  @ 478e00:1be820
0xUNKNOWN  @ 36434:1be800
0x00017B84 @ 1363c:1be798
0x00001900 @ 17bcc:1be790
0x0044C478 @ 44c57c:1be758

Magic Lantern version : Nightly.2018Dec21.EOSM2103
Mercurial changeset   : 577a7023a9f9 (crop_rec_4k_mlv_snd_isogain_1x3_presets_EOSM2) tip
Built on 2018-12-22 01:17:20 UTC by [email protected].
Free Memory  : 384K + 1911K

CRASHLOG1.LOG

ASSERT: !IS_ERROR( TryPostEvent( this->hTaskClass, this, EV_READOUTDONE_INTERRUPT_EVF, NULL, 0 ) )
at ./Evf/EvfState.c:520, **INT-D9h**:11bdc
lv:1 mode:3

debug_task stack: 2abfb0 [20fbd8-20dbd8]
0x0044C478 @ 44c57c:2abfb0

CRASHLOG2.LOG = CRASHLOG3.LOG = CRASHLOG4.LOG

ASSERT: !IS_ERROR( TryPostEvent( this->hTaskClass, this, EV_SETPARAM_INTERRUPT_EVF, NULL, 0 ) )
at ./Evf/EvfState.c:503, **INT-E0h**:11b9c
lv:1 mode:3

debug_task stack: 2abfb0 [20fbd8-20dbd8]
0x0044C478 @ 44c57c:2abfb0


CRASHLOG5.LOG

ASSERT: !IS_ERROR( TryPostEvent( this->hTaskClass, this, EV_VD_INTERRUPT_EVF, NULL, 0 ) )
at ./Evf/EvfState.c:545, **INT-6Ah**:11c40
lv:1 mode:3

debug_task stack: 2abfa0 [20fbd8-20dbd8]
0x0044C478 @ 44c57c:2abfa0

==> Pulled battery!



- Expo.Override seems to work fine if ExpSim is off

Besides these observations all functions in Expo menu are stable (as they are in Overlay menu as well as in Focus menu). More to come!
#22
Camera-specific Development / Re: ML on EOS-M2
December 21, 2018, 11:10:15 PM
Quote from: dfort on December 21, 2018, 08:51:57 AM
Found the problem. The EOSM2 uses lens properties like the other Digic 5 cameras while the EOSM uses a different set of lens properties. Let's see if the December 20 build that is on my Bitbucket downloads page is any better.
I can confirm that lens focal information works perfectly with the new December 20 build! I will continue testing with this build (let's call it D) and will post results asap.

I promised earlier to post the error log from the crashing DUO-ISO module. Here we go:


Locking cache
Lens moving (1561,  0)
Scanning modules...
Load modules ...
  [i] Load:  dual_iso.mo
Linking ..
tcc: error: undefined symbol 'mlv_set_type'
  [E] failed to link modules
updating Movie Tweaks -> Movie Logging
updating Movie Tweaks -> Time indicator


#23
Camera-specific Development / Re: ML on EOS-M2
December 21, 2018, 12:29:26 AM
Just trying to get an overview about the different builds published by @critix, @danne and @dfort so far. I feel a bit unsure how I should continue testing on which build. Took some RAW pictures and tested basics with all OVERLAY functions again on all three of these versions and came up with the following observation (again from a still photo point of view. Not that much MLV skills here - yet):

A build magic-lantern-crop_rec_4k_mlv_snd_isogain_1x3.2018Dec07.EOSM2103 (critix)
B build crop_rec_4k_mlv_snd_isogain_1x3_presets_EOSM2_2018Dec19.EOSM2103 (danne).
C build magiclantern-Nightly.2018Dec14.EOSM2103 (dfort)

A & B both show the same freezing effect in LV photo mode 2 as soon as overlays are activated based on RAW depending parameters (Histogram and/or Spotmeter - as discribed in post #321)
C does not at all freeze in LV photomode even with activated histogram in RAW RGB, LOG while shooting RAW stills!  That's good!

So from that perspective I tend to continue testing w/ build C! But there is another minor observation, it is about lens information shown in the ML Overlay bar:

A build shows correctly focal lenght 22mm, also a varying focus distance as well as the DOF range when activated for LV.
B & C builds do not! Focal lenght is 0mm and focus distance is constantly 56m, DOF range is not shown at all, even when activated for LV. Something got lost in these builds!

Last but not least all builds have in common:
- DUAL-ISO module crashes for all builds directly after re-booting (some error messages are sent to the console. Will post them later)
- Leaving ML menue always requires doubble-press Canon menu button to get into photo mode with all overlays shown (see #308 comment 5)

Based on Build A I also did a bunch of tests for all funtions in Shoot / Focus / Display menue - mostly w/ positive results (as far as I recall). But I will wait for your hint what build should be the one to continue working with before spending time with texting (unnecessarily).

#24
Camera-specific Development / Re: ML on EOS-M2
December 15, 2018, 07:10:02 PM
Focus peaking worked just fine with the build I was checking out ;) I am travelling this weekend. Back on Tuesday. Will continue testing then for sure!

Sent from my SM-G930F using Tapatalk

#25
Camera-specific Development / Re: ML on EOS-M2
December 12, 2018, 02:21:13 PM
Quote from: uizin on December 10, 2018, 02:15:44 PM
Yes @JohanJ! I had that setting and confirm that it works also in C2 now. I think I will keep the custom setting to quickly select the ISO, otherwise I'll need to browse in the menus...
Right. That comment was helpful! My first thought was that you are probaly not shooting stills in RAW format. Maybe you do. But the freezing has to do with RAW parameters for certain  ML overlays!

Both "histogram" and "spotmeter" have parameters for RAW data analysis. As soon as you set RAW depending parameters the LV tends to freeze w/ moving objects or half-shutter pressed moving or simple panning. Tipping half-shutter again refreshes LV.
High ISO values seem to speed the freezing. But here I am not sure I have to test more carefully.

Anyway when using overlays with RGB or non RAW based parameters there is no freezing  at all. I can even add more overlays same time - no problem when shooting stills.

Hope this can help.

Sent from my SM-G930F using Tapatalk