Jump to content

The Future of ML 5D3 Raw Video


gloopglop
 Share

Recommended Posts

What's going to happen in the future for Magic Lantern raw video, specifically on the 5d Mark III?

 

Where will that project be in a year, two years?

 

I'm in a position where I'd love to invest in a 5d3 and accompanying lens system if I felt that I could reasonably predict where the Magic Lantern team is going

 

Undoubtedly the firmware will improve by leaps and bounds

 

As a film maker working on my first feature which will not be shot for, let's say, several years, I just wish I could say I might have an idea of where the future could lead––regarding ease of use and speed in workflow primarily. I'm already thrilled with the image: that's why I'm interested

 

May I please ask for your subjective speculation?

 

To put a fine point on it: is it completely insane to invest in that camera system anytime soon given the rapid changes in the market, assuming that one wants it as a learning tool in the present, in order to do much more serious work years from now? Please consider that shooting stills is an ongoing need

 

Is there any reasonable chance that the ML team will abandon further development of raw video features in the near future?

 

I would love to hear your input. Thank you in advance you guys r v smart

Link to comment
Share on other sites

EOSHD Pro Color 5 for Sony cameras EOSHD Z LOG for Nikon CamerasEOSHD C-LOG and Film Profiles for All Canon DSLRs

If they stopped development tomorrow, the ML RAW hack will stand as one of the greatest free "upgrades" any hardware has ever received. The image is phenomenal, rivaling top-end digital cinema systems. Even when the 5d4 comes out and Canon defeats ML in hardware, people are going to keep using 5d3s. 

 

Any improvement going forward is going to yield limited visual improvements. However, as CF cards get faster, it might be possible that we see the hack allow for overcranking/slow-motion at 1080p. 60fps would be a good start.

Link to comment
Share on other sites

I think it has probably plateaued now... there will be improvements to workflow and higher continuous shooting at higher resolution (in crop mode)... but certain things just can't be improved on, as they have to continuously tell people who ask for full frame 4K, or 120fps.... (though they occasionally have a breakthrough with new code).

 

As Ben stated above though.... You could buy it now , install a stable version and never update it again and you would still have a great camera for the coming years. The 5D range also hold their resale price well, so we the next big thing comes along, you could sell and reinvest.

 

One thing, have you though about audio? I know they are making some strides on that side of things, but it might never sync perfectly

Link to comment
Share on other sites

Thank you so much for the feedback guys, that's what I was hoping to hear~! lol

 

The heart wants what the heart wants –– and my heart wants a 5d3 to f**k around with for the next year or so while I figure other things out for my lil production

 

I'm sold on image quality, moreover I'm in love with it.

 

What pitfalls are there that I need to be aware of? It's ultimate goal is being used in my lil low budg feature... which is in terms of production a bit like My Dinner with Andre meets Trainspotting

 

One thing, have you though about audio? I know they are making some strides on that side of things, but it might never sync perfectly

 

Hmmm that's something to be aware of, thank you ScreensPro

 

dont rlly know what I'm dealing with there, I'll have to do some googling

Link to comment
Share on other sites

As much as I love ML, and I've been working with it intensively for 3 months now, it is NOT a professional hack and I have no sense that it ever will be.  Each devs work independently to implement cool features that HE wants to work on; they work together only by chance.  There are conflicts and rivalries beneath the surface.  There is no ML leader, I can see, who sets priorities or goals.  However, yes, if you're happy with a build and it shoots fine there is no reason it won't work forever.  

 

It appeals to me because video is a hobby and I love photography.  I get the sense from other posters that followers are either old farts like me (relatively speaking) or young aspiring film-makers who can't afford professional equipment.  This isn't to say people aren't using it professionally; they are.  But you want to be realistic.  If you're shooting something important enough where you have actors showing up, grips, lighting, an audio person, etc., you're going consider renting equipment, or going with something like the BM cameras that are designed for this purpose.

 

As great as the video is from the 5D3, it remains a STILLS camera designed for stills.  

 

In short, the 5D3 on ML "CAN" do professional video, but you may not want to.

 

The real benefit I think you'd see from getting a 5D3 and doing RAW on it is the learning experience.  I've learned more about photography in the past 3 months wrestling with RAW than I could ever imagine.  In order to do it well, you have to learn about RAW non-bayered images, and all that goes into making them usable.  So if you feel you're still learning, and have the money, this is the camera you want.  (I don't have a 5D3 by the way, I bought a used 50D and EOS-M.  You don't have to spend a lot of money to work with ML RAW!)

Link to comment
Share on other sites

 

 

One thing, have you though about audio? I know they are making some strides on that side of things, but it might never sync perfectly

 

Btw, how are things on the audio front? have they put it back in the newest release? I am so waiting for them to sync the audio becuase there are many situations that I can use where I don't need to record the audio externally and internal audio would save lots of time.

 

you mentioned they are making strides? what are they? Thanks

Link to comment
Share on other sites

There are in my experience, maybe 2 or 3 devs that come to mind on the team that seem to represent themselves in a professional manner and give you a sense of comfort when it comes to what they say and produce. Alex and Gheggo come to mind immediately. There are also a handful of devs that know how to program, but don't instill much in the way of confidence. When I had a 6D, the thread for its development would often go freaking binary, and it just started to look like a young smart kid in rockstar mode playing around and talking in code to people who obviously had no idea what he was talking about. I decided to quit before I bricked a $2000 camera. Other than that, the "mature" crowd of ML are basically responsible for delivering the best hybrid camera on the market right now with the Mark 3. A 22mp still camera that shoots 14bit raw full hd. It sucks, because Canon threatened the free world with legal action to protect a $12,000 8-bit camera... the possibilities of the 1DX would've been awesome had ML decided to begin development on it with all of that processing power to work with.
Link to comment
Share on other sites

As much as I love ML, and I've been working with it intensively for 3 months now, it is NOT a professional hack and I have no sense that it ever will be.  Each devs work independently to implement cool features that HE wants to work on; they work together only by chance.  There are conflicts and rivalries beneath the surface.  There is no ML leader, I can see, who sets priorities or goals.  However, yes, if you're happy with a build and it shoots fine there is no reason it won't work forever. 

 

 
Thanks for confirming that my impression of ML development was accurate. It's an amazing achievement & congratulations to all concerned but I just wish they would develop it like other open source projects & put a stake in the ground on a version & fix & debug that before adding more features.
 
RAW video is such an interesting novelty that it seems to have diverted the developers from even looking at fixing broken features that are arguably of use to more shooters e.g. my pet peeve with Movie Restart not working on the 5D3. a1ex's word on the subject is "I have absolutely no use for movie restart, so I'm not going to fix it (especially since a single testing attempt is going to take half an hour)."
 
Link to comment
Share on other sites

Would I shoot a commercial project with Magic Lantern Raw on the 5D. Yes, I would, but I would have multiple 1000x CF cards that I know worked. Lens Rentals and other places are offer Magic Lantern 5D rental packages so it's safe to use. Audio might be a problem but...who really uses DLSR internally recorded audio anyway? I feel that the Magic Lantern features make the 5D Mark 3 an incredibly versatile camera. Crop Mode shooting is just icing on the cake. I can take a 70-200mm f/2.8 and make it a 400mm f/2.8 and lose nothing IQ wise. The detail is wonderful.
1404700_10151858309344597_519331260_o.jp

1403537_10151858308989597_220449546_o.jp
This was shot with crop mode on a 24-70mm f/2.8.

736163_10151858309949597_394563504_o.jpg

1403476_10151858309419597_461732373_o.jp
I can't see myself shooting h264 anymore except for wedding shoots or other event type of shooting. I do like what Blackmagic is doing. I went back and fourth between the 5D and the BMCC and finally settled on the 5D and I'm very satisfied with it.
Link to comment
Share on other sites

I think the future of ML will be strongly influenced by the next 7D2 release.

 

If Canon does not cripple it, we might get a true S35 RAW camera for $2000, it could seriously become THE most popular indie cinema camera and ML will thrive on that wave. 

 

The announcement of 7D2 is scheduled for March 2014.

 

Time will tell...

Link to comment
Share on other sites

Some more observations.  I've been trying to make the EOS-M a nice portable RAW camera (720p), as an alternative to the BMPCC.  One of the obstacles are the focus dots that appear in the raw frames.  They generally turn into pink dots after debayering (because most are in the red pixels).  Someone created "PInk Dot Remover" in Java which is certainly a very slick and effective piece of software  However, it doesn't get all the dots (interpolate around the "hot" focus pixels).  The first dev is no longer around, nor the second guy who took it up.  I also wanted something in Windows, just to make my workflow easier.

 

Anyway, I use a C# project open-sourced by g3gg0 (one of the lead rock stars) to get me going.  It has a function to read pixels from the RAW files, but not to write them.  I ask him if he can do it and he says he's too busy.  No problem.  I figure it can't be too hard.  Well, it takes me hours upon hours, and I can't do it..  I finally go back to him and another developer and ask them what the format of the RAW data is.  They don't seem to know!  Seems incredible to me.  How can they not know the basic format of pixel data in RAW files?  I believe the answer is that they took routines from elsewhere and modified them a bit without knowing exactly what these routines do (lots of bit-wise arithmetic).  They were able to get each pixel's value into a ushort and that was it.  As smart as they are, they are overloaded by technical stuff, which I believe even they can't get completely on top of.

 

I finally figured out that the data was encoded as a 14-bit stream where each other byte was swapped with the one before it.  If I had known the original format I could have figured this out easier.  The point of this story is that as much as the devs seem to know, the stuff is monstrously difficult on so many levels.  Knowing what I know now, I'm amazed ML continues to move forward.  However, if one or two of those key guys quite that would be the end, or severely slow down the development of ML.  I believe this is not unlikely.  Because...

 

Video is a narrative medium.  Yes, it's fun to take some RAW video of your garden, kid, tugboats, etc.  After that, no one wants to watch a video that doesn't tell some kind of story.  Story requires scripting, editing, actors, lighting, audio, etc., etc.  I don't know if this is heresy, but I can't see why someone would use a 5d3 to shoot a film when they could use a BMCC or any of the other dedicated cameras for that purpose.  Now that the BMPCC seems to get in stock I"m thinking, I have $850 in a used 50D (CF card+10-20mm) to shoot RAW, what am I thinking!  I've already watched 2 people on the ML forum sell their 50ds, or try to.  I may be next.  The other thing about narrative video, VERY FEW people actually do it.  That is a VERY SMALL market.  And many of them, like Andy, don't have the time or need for any kind of RAW-type workflow.

 

Yes, I believe the 5D3 is a better value than the BM cameras, but only if you're going to spend the time/effort to get the most out of the RAW feed.  If you just need basic improved dynamic range, the BM cameras make so many things easier.  And they will continue to improve.

 

Because the devs won't try to make ML stable on these cameras, or even allow the general public discussion of a to-do list, I can't see it growing beyond a "cool" hack.  Maybe I'm too close to it.  Familiarity is starting to breed contempt ;)

Link to comment
Share on other sites

maxotics this is exactly the shit that im talking about, the insight im looking for

 

and i commend your effort with the EOS-M, seriously

 

may i ask: how are you so smart about this stuff??? rocket scientist???? :o

 

at the end of the day, although of course the 5d3 shoots great stills, and i have a need for that, im ONLY looking at it as a video camera because i am soooooo subjectively in love with that image, both because i think it looks amazing and because i have a history of shooting canon slr raw stills... and thats just what it looks like. so the color science is very familiar to me in that respect

 

but yeah, blackmagic would be soooooo much easier to deal with!!!! ugh. thats why all this is driving me crazy

 

like i feel like shooting an indie feature with this "hack" is a little insane because (thank God) there ARE other options today that are affordable?? for great cinematic video??

 

and?? it fucking scares me that something wont work on the set and ill just be like "Well citizenkaden thats what you get for using this hack bullshit on a freaking movie you moron"

 

i just wish i could count on ml raw becoming stable smh :(

Link to comment
Share on other sites

@Citizenkaden, I've been developing software (mostly financial) for almost 30 years. That and more time than I should spend!

My original goal was/is to find a portable RAW quality video camera to match my SigmaDP1 and SigmaDP2s (I hope to eventually get the Merrill versions of these cameras). The BMPCC is the best solution. However, an EOS-M could also double as a very good all-around shooter. The Sigma cameras are very limited to either ISO 400 (and under), or BW in high ISO. And they are slow. I can put on a Canon Flash onto the EOS-M, put in P mode, and shoot any event with great results. The H.264 is also very good from the EOS-M.

LIke you say, you can't find the most stable build of any ML build (in Alpha for RAW). I asked nicely, prodded, lectured, etc., on the forums, all to no avail. I believe the core dev is taking a break. I haven't seen him recently.

BTW, I have the same problem with this site. I couldn't be more impressed by the work Andrew does. He does studies that corporations pay $10,000 a shot to do ("Figure out which camera and platform we should buy globally, etc.") We get it for FREE! However, he gets off track. HIs starting a thread about the Nikon DF, which doens't shoot video, is a distraction and raises my level of anxiety. I know that sounds odd, but I suspect many feel the same. If I want to socialize about cameras dpreview does a great job. But there is only ONE EOSHD! :)

I see no indication whatsoever (if anything the reserve) that ML Video RAW will only be a "home" video RAW solution for advanced hobbyists and professionals.

@ScreensPro, I don't know about your family, but my family will run from an F3 :)
Link to comment
Share on other sites

http://vimeo.com/66170436

 

This video is a commercial for the 5D3, you'll notice the advert link right below the video.  It basically shows someone who doesn't know (or intentionally) f***s-up the BMCC in post. 

 

 

Here is good video comparing 2 cameras where black magic is way better then mk3

http://vimeo.com/49875510

 

 

This is comparing the BMCC to 5D3 h264, not 5D3 raw.  H264 makes the 5D3 look soooo much worse than it actually is. 

Link to comment
Share on other sites

I know it is not raw, but have you looked into a 2nd hand Sony F3? I've seen some amazingly cheap prices, kits for $5k or so.... couple it with a BM recorder and you are shooting 10bit S-log at 13 stops DR with alot of nice video features and quality audio.

 

dude! EXACTLY. an actual video camera––a freaking great one

 

:o

 

::sigh:: idek

 

also, @maxotics - very interesting

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

  • EOSHD Pro Color 5 for All Sony cameras
    EOSHD C-LOG and Film Profiles for All Canon DSLRs
    EOSHD Dynamic Range Enhancer for H.264/H.265
×
×
  • Create New...