New owner concerned about status of ML on 6d

Started by l_d_allan, December 13, 2013, 10:31:18 PM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

l_d_allan

I've been a happy user of ML on my 5d2 and T3i for a year or so now.

Yesterday, my 6d with 1.1.3 arrived, and seems to work fine. I'm cautiously eager to install ML.

But ... when I got ready to proceed:

  • broken nightly build for several links
  • no nightly build at this time (Dec 13) ... nothing shows up for 6d on Nightly Build page
  • attempted to read 57 pages of "Tragic Build" thread, with increasing concern
  • now reluctant to install the version that I did download  (file name indicates Dec 11th)

I'd really appreciate feedback / replies from 6d owners with ML working well on their camera.


Marsu42

Quote from: l_d_allan on December 13, 2013, 10:31:18 PM
I'd really appreciate feedback / replies from 6d owners with ML working well on their camera.

I was in the same situation, and it's somehow good to see you're as confused as I was :-)

The unfortunate situation: 6d is not supported by Alex' "Magic Lantern" anymore (he hasn't got a 6d), but is in a separate fork "Tragic Lantern" by 1% who - read up on the reasons - doesn't merge back the TL changes to ML. Alex considers the TL port as unsafe and I don't doubt it given his experience - but fact is TL currently the one that works at all and I've been using it since months, however doing nothing fancy with it but your basic auto iso, histogram & overlays.

* TL now has an own subforum, for installation refer here: http://www.magiclantern.fm/forum/index.php?board=58.0
* 1%'s repo with 6D some downloads is here: https://bitbucket.org/OtherOnePercent/tragic-lantern-6d/commits/all
* There are also TL nightly builds somehwere if you don't want to compile yourself.

DTSET123

Whatever is happening between Alex and 1% is very sad :(  I have been using TL since day one and I never had any problems with it. I hate to take sides, and I understand Alex's concerns, but there is no one else working on 6D (at least that I know of) so I will take whatever 1% will put out.  I hope devs will find common ground soon!!!   

l_d_allan

Quote from: Marsu42 on December 14, 2013, 12:03:00 AM
The unfortunate situation:

Very. But working with "software jocks" has a well earned reputation of being like "herding cats".

And working with volunteers also has a reputation of being like "herding cats".

So ... working with volunteer Devs?   The mind boggles.

QuoteA1ex considers the TL port as unsafe and I don't doubt it given his experience - but fact is TL currently the one that works at all

One of the Great Strengths of ML is that a1ex and other Devs are clearly "eating their own dog food". And that is perhaps as true or more true for 1%?

Quoteand I've been using it for months,

Thanks for the feedback. I'd really like to hear from some other "early adapters" and "bleeding edge pioneers".

Quotehowever doing nothing fancy with it but your basic auto iso, histogram & overlays

That concerns me. Hope I'm not coming across as "entitled", but I really, really value (need?) ETTR, Dual-ISO, intervalometer, extended BULB, and so on for night / full-moon panos. Like tomorrow in western Kansas for grain elevators. Brrrr.

Quote
* TL now has an own subforum, for installation refer here: http://www.magiclantern.fm/forum/index.php?board=58.0
* 1%'s repo with 6D some downloads is here: https://bitbucket.org/OtherOnePercent/tragic-lantern-6d/commits/all
* There are also TL nightly builds somehwere if you don't want to compile yourself.

And it gets worse? I'd maybe have to compile myself? Builds are "somewhere"?

I was able to download a 6d build from the main "Nightly Build" webpage several days ago. Comments appreciated on that Build's stability, features, etc.
magiclantern-v2.3.NEXT.2013Dec11.6D113

Really, I'm not meaning to be unappreciative and disrespectful of what 1% has accomplished. I'm sure it's remarkable ... but I've been spoiled by ML on my 5d2 and T3i / 600d.

a1ex

Quote from: l_d_allan on December 14, 2013, 07:29:00 PM
I was able to download a 6d build from the main "Nightly Build" webpage several days ago. Comments appreciated on that Build's stability, features, etc.
magiclantern-v2.3.NEXT.2013Dec11.6D113

I bet nobody tried them in the last month or two, so I've removed them from the download page.

Last update on 6D codebase in the main repository was done by me, on 2013-09-24. I've tried to merge some changes from 1%, blindly, without being able to test them on the camera. Of course, without any sort of feedback I can't maintain the 6D port, so I moved it to "unmaintained" state until somebody will jump in and bring it back to speed.

l_d_allan

Quote from: DTSET123 on December 14, 2013, 07:04:03 AM
I hope devs will find common ground soon!!!

+1

I've participated on several FOSS over the last decade or so. And pretty much all of them had serious problems with co-operation among devs.

To my shame, I'll admit being a detractor to at least one project in terms of team unity. My bad.

I've gotten dependent of FOSS, and then watched helplessly on the sidelines as they blew up for unknown .. to me ... issues between "the powers that be".

From Wiki about F. Nietzsche:
"The will to power describes what Nietzsche may have believed to be the main driving force in humans: achievement, ambition, the striving to reach the highest possible position in life; these are all manifestations of the will to power."

And from a country song ... paraphrase of Rodney King:
"Why can't we all just get a long ... neck?"  :o


l_d_allan

Quote from: a1ex on December 14, 2013, 07:31:31 PM
I bet nobody tried them in the last month or two, so I've removed them from the download page.

Thanks for the feedback .. you've address several of my issues.

QuoteLast update on 6D codebase in the main repository was done by me, on 2013-09-24.

Is there a link to that "somewhere"? Or vanished "into the ether"? I think it might provide most (all?) of the functionality I've salivating for.

But ... with you unable to "eat your own dog food" on that Build, would you recommend it?  Is the Dec 11th build a Tragic-Lantern build that managed to be visible?

QuoteI've tried to merge some changes from 1%, blindly, without being able to test them on the camera. Of course, without any sort of feedback I can't maintain the 6D port, so I moved it to "unmaintained" state until somebody will jump in and bring it back to speed.

Thanks for your efforts on that.  Due to utter ignorance on my part about what has happened, I will try to carefully express sympathy and empathy for a frustrating situation ... to all involved ... including the non-Dev end-users like myself.

I was just on the BhPhotoVideo taking a look at what would be involved to exchange my 6d for a 5d3. Ouch ... $1200, and for me, I suspect the 6d is better than the 5d3. YMMV.

BTW: if a ML member asks nicely, it may be possible to get a 6d body for ~ $1375 USD. I think their "normal price" is down to ~ $1500, with a Bunch of knick-knacks you probably already have (battery, monopod, case, SD card, etc). They might take off $125 if you get the body only. Or not?


jordancolburn

When (if?) someone starts maintaining the 6d build, I think the important thing is that end users move over and test, if everyone is using TL and nobody can vouch that ML doesn't have bugs for the 6d, there's not really a reason for it to exist.  Also, as a very, very, newbie dev, I found it was pretty easy to update ML to work for the EOSM, so if anyone has even moderate coding skills don't be afraid to step up and help maintain it.

Audionut

Quote from: l_d_allan on December 14, 2013, 07:57:26 PM
Is there a link to that "somewhere"? Or vanished "into the ether"? I think it might provide most (all?) of the functionality I've salivating for.

https://bitbucket.org/hudson/magic-lantern/commits/all

Specifically, looks like this one.


Marsu42

Quote from: jordancolburn on December 14, 2013, 09:24:10 PMAlso, as a very, very, newbie dev, I found it was pretty easy to update ML to work for the EOSM, so if anyone has even moderate coding skills don't be afraid to step up and help maintain it.

It would be great if it would be that quick and easy ... it is when merging works, but what if it isn't? In this case you need "at least" moderate coding skill, ml knowledge and a lot of time to track bugs and adapt ML changes to a camera alex doesn't have and cannot really help with.

1% has most of this this skill, and so far he also had time - that's why 6d/7d are merged to him. Btw the reason he's not doing the job supposedly is that he's not fine with just "maintaining", but he wants to quickly merge his own changes, even if cross with the main ML dev decisions. Edit: I'd would like to point out again that this is not directed against 1%, what he's doing is perfectly valid and well within the bounds of oss and community behavior.

Looking at how much time 1% invested into this, don't underestimate the job - I really don't want to discourage anyone, but this isn't done by a couple of people occasionally looking at the code, at least unless the 6d is at 100% stability and feature level just like ML.

jordancolburn

Sorry, I don't mean to underplay it, but it really was just a lot of recreating 1%s changes by hand (where they applied) to the ML code.  He's done a lot of awesome things that really took time and research and trial and error, it will just take some time for people to apply those to the main ML branch in a way that it can move forward with the widest camera support.  It will take time, and I am starting with moderate coding ability (I work doing web development, but have my MEng in EE, with some courses in C and assembly stuff), but the barrier to getting the code compiling and committing changes wasn't as bad as I thought and would just like to encourage others to dive in if they have an interest.

Levas

Quote from: a1ex on December 14, 2013, 07:31:31 PM
I bet nobody tried them in the last month or two, so I've removed them from the download page.

Last update on 6D codebase in the main repository was done by me, on 2013-09-24. I've tried to merge some changes from 1%, blindly, without being able to test them on the camera. Of course, without any sort of feedback I can't maintain the 6D port, so I moved it to "unmaintained" state until somebody will jump in and bring it back to speed.

Hi Alex,

Are you saying that if you get enough feedback on 6d builds,  you will maintain nightly builds of the 6d ?
I think there are (now the nightly builds are gone) enough 6d users who are happy to report feedback.
Is it an idea to start a new 6d development topic and claim in the first post that there should be at all time a minimum number(10 ?, 20 or more ?) of active users who download and use the newest builds and give feedback in the topic about bugs ?

P.S. happy magic lantern user on the 6d since september  :D
I just signed up for the forum this month because it looked like 6d development was dying and users needs to be more seen in the forum  :'(

a1ex

I prefer this to be done by a 6D owner.

I'm not afraid of doing it (since I did the 600D and 50D ports via skype), but it's not easy, since each camera has its own set of quirks. These are best handled by somebody who is actually using that camera. There are people in the community who have both a 6D and the necessary skills, but time and motivation are probably an issue. ML is a spare-time project, after all.

Levas

Quote from: a1ex on December 30, 2013, 11:14:17 AM
I prefer this to be done by a 6D owner.

I'm not afraid of doing it (since I did the 600D and 50D ports via skype), but it's not easy, since each camera has its own set of quirks. These are best handled by somebody who is actually using that camera. There are people in the community who have both a 6D and the necessary skills, but time and motivation are probably an issue. ML is a spare-time project, after all.

Ah I see, no 6d, so you're builds would be none tested/checked at all, not even by yourself.  :o

In that case, is there anywhere still a link to the "magiclantern-v2.3.NEXT.2013Dec11.6D113" build, which is mentioned in this topic ? Can't find it anywhere and the latest build I is the tragic lantern build from 30 november. I'm experimenting with the new sound recording function with MLV raw recording, but it only works with the first clip, after that I get a "sound recording failed to stop, state 4" error.
Reading through the topic, the build I'm using could be too old (tragic lantern from 30 november)

As long as there's still some progress on the MLV recorder and sound recording module for MLV I'm still interested in new builds.
If it won't take too much of you're time, would it be possible to make, once a month a newer build and only post the link on this topic, where you mention that there's nothing tested at all.
And if there's nothing new happening on the MLV module or dual iso module (think there still working on that one too?), then a new build won't be needed I guess.
The other magic lantern function on the 6d are already working fine, so won't expecting something new there. Even the raw recording (not the MLV type) works fine, no progress needed/possible there.
I think that's the main reason why most people are satisfied with the already existing builds.
But sound recording with MLV recording would be something new and fantastic, that's why there's some need for newer builds ::)


a1ex


Levas


l_d_allan

Quote from: Marsu42 on December 14, 2013, 12:03:00 AMThe unfortunate situation: 6d is not supported by Alex' "Magic Lantern" anymore (he hasn't got a 6d), but is in a separate fork "Tragic Lantern" by 1% who - read up on the reasons - doesn't merge back the TL changes to ML.
I'm still unclear on the status of ML / TL on the 6d. The above quote was from several months ago.

  • There have been Nightly-Builds of ML for the 6d for a month or so.
  • I inferred from that ... 6d was supported by ML ... and that the 6d version would behave much like ML on 5d2, 600d, and others.
  • I also speculated that the TL "fork" had been merged and "unforked" .... perhaps that was "wishful thinking"?
  • And ideally, that the devs for ML and TL were working together in peace, harmony, and effectiveness.
  • However, I notice that TL builds are still showing up (latest Feb 28, 2014) ... which surprised me.
  • So ... is there still parallel development for the 6d by ML and TL devs?
  • Are there pro's and con's for ML vs TL?
  • My less than informed speculation is that the ML version may be safer and standardized, while the TL version may have some more features reflecting the interests and priorities of the TL dev(s). Or not?

Audionut

I'd rather not have any more politics of the development situation discussed in threads all over the forum.
Suffice to say.


Tragic Lantern is a personal fork.  Just like you, me, and everyone else is capable of creating our own personal forks.

This forum is for the support of Magic Lantern.  First and foremost.

The 6D has no support from the core developers.  That is to say, they don't have a 6D in their hands.

The 6D has been supported again in the nightlies, under the sole prevision that it will be maintained to the standards set forth by the Magic Lantern development team.  Currently, this maintenance is conducted by 1%, with some smaller contributions from users.

Based on the development progress I have seen in recent weeks, I would expect the 6D to be on an equal development cycle with the other ports at some stage, in the near future.

Rather then speculating, and worrying about things that do not improvement the development progress of the 6D.


Regards.

l_d_allan

Quote from: Audionut on March 06, 2014, 01:49:24 PM
I'd rather not have any more politics of the development situation discussed in threads all over the forum.
Suffice to say.

Ok. Sorry. I can modify the earlier post, or delete.

Or ... since you are a moderator, please feel free to modify or delete as you feel appropriate.

l_d_allan

I still concerned about the status of ML on the 6d.

There for a while, there were regular ML nightly-builds. I switched from TL to ML.

About two weeks ago, I attempted to get an updated nightly build, and noticed a series of "build failed". That is still the case. The latest build that was successful was May 21, 2014.

I'm reluctant to switch back from ML to TL. Is the 6d no longer being actively developed / supported?

a1ex

Quote from: a1ex on June 22, 2014, 06:10:13 PM
Before resuming the nightly builds, I would like somebody to compile the code, run the stubs test (Debug menu), and post a screenshot of the Free Memory dialog. And, of course, some general checking to make sure the port is not totally broken.

I'm still waiting for these tests.

l_d_allan

Quote from: a1ex on June 25, 2014, 03:29:54 PM
I'm still waiting for these tests.

Sorry, I looked at the last 20 or so posts on that long thread, but not far enough back to see your comment.

Levas

Quote from: a1ex on June 25, 2014, 03:29:54 PM
I'm still waiting for these tests.

Are there any users out there who own a 6d and can compile ?
Anybody ?


l_d_allan

from another thread:
Quote from: a1ex on June 22, 2014, 06:10:13 PMIf nobody volunteers to fix FPS override in the next two weeks, I'll resume the nightly builds without this feature.

Not meaning to be impatient or unappreciative, but ... hmmmmm ... it's been a bit longer than two weeks since a working Nightly Build for the 6d has been available.

TLN

What is current status of 6D firmware?
I'm thinking about to upgrade my 5d2 to 6d, and this makes sense. I'm running Nightlies and it work flawlessly on my 5d2.
Wonder what features are missing in 6d firmware and how stable it is? I've read that rhere's no FPS overrive in 6d, and that's what I use(not often, but this is fun thing).