Canon EOS M

Started by jordancolburn, December 30, 2013, 10:21:20 PM

Previous topic - Next topic

0 Members and 7 Guests are viewing this topic.

dfort

Yes, the EOSM platform is still blocked until this pull request is QC'd by a1ex. In the meantime you're more than welcome to use my branch:

hg clone -r EOSM__working https://bitbucket.org/daniel_fort/magic-lantern

travisennis

Thanks, Daniel. I followed the link that DeafEyeJedi (thanks to him, too) originally gave me which pointed to your repo. I pulled it down and built that and I've been very happy with it so far. I originally had the build from April on it and found it glitchy. Your build seems much more stable.

DeafEyeJedi

As expected and indeed it does work well on my M as well.

Well done, Daniel and Thanks for your hard work as always!
5D3.113 | 5D3.123 | EOSM.203 | 7D.203 | 70D.112 | 100D.101 | EOSM2.* | 50D.109

tupp

Do we have GOP controls yet on the EOSM?

Thanks!

Licaon_Kter

Not really.
Theres that IPP/ALL-I but that's it.

tupp

Thanks!  ALL-I is great!

Is this feature enabled, or does one have to compile?

dfort

Looks like it has to be compiled. The notes on what needs to be done is in this topic:

http://www.magiclantern.fm/forum/index.php?topic=15685.0

I followed dpjpandone's instructions but couldn't get it working. Maybe you can give it a try and report back?

tupp

Thanks, dfort.

It appears perhaps that Licaon_Kter knows something that we don't.  The dpjpandone thread that you link indicate that there were extensive GOP and flush rate settings back in August, but dpjpandone recommended having only the choice between IPP and ALL-I.  However, Licaon_Kter seemed to imply that now there is only the choice of either IPP or ALL-I (with a locked, default flush rate), in coincidence with dpjpandone's recommendation.

Were dpjpandone's suggestions incorporated since last August, and, if so, could that be the reason why your compile attempt failed?

Licaon_Kter

This is the dpjpandone build: https://www.dropbox.com/s/3d6zpxnndvhxpoa/magiclantern-Nightly.2015Aug21.EOSM202.zip that has it.
Although there is a catch, but I can't find the info now, something will happen (crash?) if you switch to photo mode when enabled or something.

tupp

Thanks, Licaon_Kter.

Is the "catch" the problem that is described in this passage (from the dpjpandone post in the thread linked above):
Quote from:  dpjpandoneIf i recall correctly, the reason this is not included in main, is because the settings persist until you restart the camera, not only this, but the camera must be restarted with the same card (with the current implementation) in order to clear the settings, this presents a dangerous situation, as someone who isn't aware of this might remove their magic lantern card, and the GOP/flush settings don't go back to default

If so, I might keep using the "unmentionable fork" and its GOP controls, as, even though it is outdated, it doesn't seem to have this problem.

dfort

Oh, you mean you want to go there. No problem looking into a developer's code, after all this is an open source project. If you want to use old builds from no longer supported forks, nobody is stopping you.

I tried adding dpjpandone's changes as he described them on his post but it didn't add the video hacks options to the menu.

This is something that dpjpandone as been asking for -- for over a year. It looks it just needs someone with the programming skills, motivation and most of all time to get it working in a way that won't leave the camera stuck in some sort of altered state.

travisennis

dfort, I pulled down the changes your made in your repo to try to get the video hacks to work. You are on the right track. What I ended having to do in order to get it to build properly was add the following

ML_SRC_EXTRA_OBJS = \
   video_hacks.o

to the eosm's Makefile.setup.default

That brought the video_hacks.o file into the build directory and the build then successfully completed. I haven't ran the firmware from the build on my camera yet, but I'm think it should work. I haven't been messing with ML long enough yet to be confident enough to be the first one to try it.

dfort

Got it, I took dpjpandone's notes literally including what is apparently a typo.

Ok--so I updated EOSM__working. If you don't feel like being a crash test dummy don't mess with video hacks. Otherwise, welcome to the bleeding edge. Note that I haven't loaded it on my camera yet but will as soon and I can get to it.

By the way if you read the very first post in this topic:

QuoteI moved some of the great  work 1% did to get ML functioning on the EOSM for firmware 2.02 back into the main ML code base.  I'm sure its missing some features and bugs from it's TL counterpart, but if you are interested in testing ML and helping make it stable and in sync with the code for other ML cameras for all base features, please comment here and I'll do my best to port 1%s bugfixes, or to test, create issues for new bugs and attempt to solve.  IMO, the main focus is ensuring all basic ML functions work well before branching out to things like raw video.

So don't be shy to dig into his code and bring back some features. I did an EOSM code cleanup that hasn't been merged yet but it is in my EOSM__working branch. In there you will find all the Tragic Lantern stubs that were either not used or removed from the ML main repository. That may give you either enough rope to hang yourself or the key to your favorite TL features.

[EDIT: News flash--a1ex just merged some pull requests including EOSM code cleanup.]

dfort

a1ex made some comments on bitbucket that should push the frame rate on the EOSM a bit further.

Quote

  • 45.02 fps in 3x crop standby (available for raw recording; limited to 45 as suggested by LK)
  • 34.37 fps in 3x crop H.264 (without video hacks)
  • 31.24 in 1080p 1x (can be pushed to 33fps in H.264 if desired)
  • 62.16 fps in 720p (can probably be pushed further)
  • 31.24 fps in photo mode (no big deal here)
  • 35.96 fps in 5x zoom (available for raw recording; can be pushed further if desired)

I'm a bit busy right now to test it but if there are any users who want to see how far they can push their little camera I added it to my EOSM__working branch and also compiled it and uploaded it to my bitbucket download area.

guiguinou

Hello!

I have been installing the last available built (april 2015), and i start having fun. Shutter bug is here but a power cycle solves the problem every time, so that's alright. I didn't pay attention before installing, but it looks like there is no way of doing some focus stacking on the eos M. Did I miss something in the thread? Is there any hope to see that happening?
I also got confused a little bit about what's going on in the last improvement. Mostly some video stuff right?

Anyway, thanks to the people working on the code!

j.diazmeco

Quote from: dfort on November 06, 2015, 03:35:17 AM
a1ex made some comments on bitbucket that should push the frame rate on the EOSM a bit further.

I'm a bit busy right now to test it but if there are any users who want to see how far they can push their little camera I added it to my EOSM__working branch and also compiled it and uploaded it to my bitbucket download area.

First time here, just got a second hand M in ebay. Compiled from your branch a couple weeks ago, and now downloaded and installed that zip. I noticed smthin, now I can go in CBR further than 3.0x, I think before I could just go up to 3.0x, no more values. ¿Am I right? Anyway no matter the value, when I enable bitrate info it seems to be stuck in q+-1.
Tomorrow will try to make some tests.

dfort

I keep adding what seems like interesting pull requests to my EOSM branch so at times it might exhibit some odd behavior. Of course ML in general exhibits some odd behavior. As do ML users--but I don't think there's a fix for that.

j.diazmeco

Quote from: dfort on November 09, 2015, 01:08:48 AM
I keep adding what seems like interesting pull requests to my EOSM branch so at times it might exhibit some odd behavior. Of course ML in general exhibits some odd behavior. As do ML users--but I don't think there's a fix for that.
Well, that's why I got a canon m, I'm odd...[emoji16]
Thanks for the work.

dfort

A few more improvements that will benefit EOSM users are in the works but the issue that is holding up nightly builds isn't fully resolved yet. Patience.

However, for the impatient who prefer a little more excitement I decided to merge lua into my EOSM__working branch. What the heck, it is very useful and it is a module so if you don't load it you won't even know that it is there. For those that still haven't set up a Magic Lantern development system I uploaded a nightly to my bitbucket download area:

https://bitbucket.org/daniel_fort/magic-lantern/downloads

Check out what you can do with it on the lua topic.

j.diazmeco

Quote from: dfort on November 11, 2015, 03:00:04 AM
A few more improvements that will benefit EOSM users are in the works but the issue that is holding up nightly builds isn't fully resolved yet. Patience.

However, for the impatient who prefer a little more excitement I decided to merge lua into my EOSM__working branch. What the heck, it is very useful and it is a module so if you don't load it you won't even know that it is there. For those that still haven't set up a Magic Lantern development system I uploaded a nightly to my bitbucket download area:

https://bitbucket.org/daniel_fort/magic-lantern/downloads

Check out what you can do with it on the lua topic.
Thanks! I will build it later and give it a look!!

alvinlok

Is it impossible to develop a magic lantern firmware for EOS M2? :-[
I really want to have the peak focus function on EOS M2 as it helps a lot in focusing when using manual focusing lens

dfort

Quote from: alvinlok on November 29, 2015, 11:48:43 PM
Is it impossible to develop a magic lantern firmware for EOS M2? :-[

Not impossible. It is being discussed in the EOS-M2 topic:

http://www.magiclantern.fm/forum/index.php?topic=15895.0

I doubt that a developer picked it up yet but there is still a glimmer of hope--thought the "official" response is something like, "Don't hold your breath."

dfort

I recently was introduced to the  "Auto adjust Kelvin + G/M" tool by Danne while doing some testing on his new MLV workflow. This is such a valuable tool and we should be using it more often. Basically the way Canon tells you to do a custom white balance is by shooting a still of a white or grey card then using that captured image to set the white balance. Those steps might make sense if you have a mirror flipping in front of your camera's sensor but since the EOS-M is mirrorless there shouldn't be a need to capture a still that you're going to throw away, especially if you shooting in crop movie mode.

Ok--so what I did was to make an extra copy of the "Auto adjust Kelvin + G/M" tool in the main shooting menu to make it more accessible. I put it in my EOSM__working branch that you can build from source via bitbucket or download. I also made a pull request so you can comment on it.

Here's how it works. The tool actually a submenu under White Balance in the Shooting Menu.



This makes it a bit difficult to get to with the EOS-M so I put an extra copy in the main shooting menu.



Aim at a white or grey card and it will sample the center 200 pixels to adjust the white balance in both the amber, blue and magenta green.



That's an Xrite Color Checker Passport though there are other options. Don't worry about focus, just make sure it is in the center.

By the way, all professional camcorders have a button dedicated to doing this--the EOS-M has a lack of buttons but this makes it much easier. No more excuses for not taking the time to set white balance before shooting!

DeafEyeJedi

Seriously now this feels like a Canon Porn site and digging 'em BIG time ... LOVE IT and thanks for sharing @dfort!
5D3.113 | 5D3.123 | EOSM.203 | 7D.203 | 70D.112 | 100D.101 | EOSM2.* | 50D.109

dfort

Still a work in progress. Fun experimenting on this little camera.

http://www.magiclantern.fm/forum/index.php?topic=16283

Did some changes - I updated the White Balance button assignment so it triggers "Auto adjust Kelvin + G/M" instead of incrementing the color temperature.

[EDIT: That broke something so I'm going back to the cluttering up the Expo menu. Note that you can also set up a custom menu but this is the way I've got my camera setup for now. Keep an eye on the pull request for the latest on what is going on with this.

https://bitbucket.org/hudson/magic-lantern/pull-requests/673/auto-adjust-kelvin-g-m-tool-in-the-main/diff ]