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 - markodarko

#101
Well, in an unexpected turn of events I'm now no longer using MLRawViewer for exporting to ProRes. Basically, I ran into a bug where it seems that individual audio slip data and In-Out export points are ignored during batch export and as I have better things to do than sit in front of my laptop and choosing the next file to render I started looking into things a bit deeper.

I discovered on this forum that MLRawViewer uses ffmpeg to render ProRes 4444 and that doesn't actually render as 16bit but 10bit (I checked on ffmpeg's website). I know the end result will probably be 8bit but throwing away billions of colour data before grading didn't fill me with joy. This on its own was enough for me to re-think my workflow and as of yesterday I've been trying out MLVFS, opening the cDNG files directly in DaVinci Resolve 12 Lite and creating timelines within Resolve that I can render as many times as I like - for example I could render ProRes 422 Proxy files for quick editing and then re-render as ProRes 4444 XQ for final output.

So far it has been quite a steep learning curve (and a slow one as I don't have a very powerful computer for running Resolve, but it doesn't need to be as I don't need to be able to play things in real time) but here's what I've found so far...

1. MLRawViewer's built-in LUTs are excellent. Being able to export as S-Log2 or Log-C is so easy. Opening cDNG files in Resolve however is a bit of a color management nightmare to the uninitiated (me). Having said that, I think I've got things sorted now:
- Resolve Timeline: Rec.709 Gamma 2.4
- Resolve Colorspace: BMDFilm
- 3D Input LUT: BMD Film to VisionLOG LUT (not as good as S-Log2 in my humble opinion but this is the best LOG starting point I could come up with, although this may change as I learn more about Resolve)

2. Resolve is slooooooow to work in on my mid-2010 MBP i7 with 8Gb RAM / 2TB SSD / 512Mb NVidia GFX, but that's to be expected judging by the specs posted on BlackMagic's website that you actually need to run things (and I guess that some of the speed loss will be due to MLVFS?). Luckily all I need to be able to do is create timelines, put clips on there, grade and render.

3. The image quality is superb (I'm not sure why people say that the debayering is rubbish in Resolve?). I rendered out a clip with the closest colour grade I could make in Resolve to match a clip which was exported from MLRawViewer and then graded in FCPX and I can categorically say that grading the native RAW cDNG files in Resolve produces a far superior result than files that are graded in FCPX from 10bit MLRawViewer ProRes files. You seem to be able to "push" the levels far more before the image degrades. Makes sense really.

4. My Resolve project will contain several timelines with clips that have been trimmed so that only the needed parts are rendered - saving disk space. You can also align audio to the clips.

5. A proper render queue. You can add as many timelines to the render queue and render out individual clips with any name you like which is one of the main features I was looking for so that  I can just press "start render" and not have to be there to open another file when one is finished rendering.

6. The whole workflow process takes longer than pressing "E" on MLRawViewer, but in my opinion it's worth it because you can grade the footage before you even start cutting (if only to colour balance) and not have to grade in FCPX meaning that FCPX will be more responsive as it won't have to render LUTs or colour corrections, just edits. When I've finished my edit I can go back to Resolve and do a proper grade and then simply re-render and replace the initial MOVs. (If I could play things in realtime I wouldn't even need FCPX or to render out the clips at all!)

Personally, I'm glad that MLRawViewer experienced that bug as I wouldn't have looked into this solution (or discovered that the ProRes files it produces are only 10bit) as I believe that although a little bit more time-consuming (due to my system specs) it is far more flexible in the long run.

Having said that, MLRawViewer is still an indispensable tool and I still stand by my initial statement a few days ago and that without it, shooting RAW would be a chore. Just being able to view what you've shot in realtime by sticking your CF card into your laptop is absolutely fantastic and I shall be using it in conjunction with MLVFS & Resolve for the foreseeable future I'm sure, just not for exporting from MLVs.

Cheers,

Mark.
#102
Hi there,

I'm not sure if you're (baldand) still active but I would just like to say that I think that MLRawViewer is the ONLY reason it makes sense to shoot in RAW with ML from a workflow perspective. It's such a powerful application once you get to know it. At first it doesn't seem like much but once you learn all the shortcut keys and discover the power of setting In-Out points, matching external audio and batch processing an entire folder it makes the RAW workflow a piece of cake - let alone the ability to immediately preview and SEE what you've just shot in REAL TIME by plugging your CF card into your computer (which makes checking takes painless) or the fact that it can output ProRes 4444 in a variety of LOG formats and hence cutting out so many other extra steps in other programs!

I tried many different solutions when I was researching the best workflow for dealing with ML files after shooting them and, bar none, there is nothing that comes close to saving time and effort than MLRawViwer. Most solutions involved converting to DNG files then opening in DaVinci Resolve to apply a LOG profile and then exporting to ProRes 4444 from there - but even then you'd still have to align your audio in the NLE. Your program does it all, elegantly and without fuss.

So... Thank you so much for developing this program. You've obviously spent a lot of time on it and it hasn't gone unnoticed.

All the best,

Mark.
#103
Thanks for all your help. This was all a misunderstanding on my behalf.

I've since done some experimenting and found that if I set the canon menu to 60fps and set fps override to 45 in ML's menu then I can record continuously at the "squeezed" 1920x648 which is excellent and perfect for my needs.

Sorry for the misunderstanding.

All the best,

Mark.
#104
Ah, so 40fps is possible at 1080p, but not continuously. Got it. What about 29.97 fps at 1080p, should this be possible continuously?

Thanks for your help,

Mark.
#105
Hi Walter,

Maybe it's me who has misread this post...?

http://www.magiclantern.fm/forum/index.php?topic=12801.0

Cheers,

Mark.
#106
Hi guys,

Firstly, thank you for all your help with developing ML. I bought a 5D III in November specifically to use ML after trying it out on my 5D II and so far it's been great! However, I tried shooting at a higher frame rate than 23.98 fps yesterday and could not record for longer than 10 seconds or so at 29.97 fps, let alone at 40 fps like others are managing.

What am I doing wrong?

I have a komputerbay card as recommended here and I'm running v1.13 of ML from November as I was told that v1.13 was more stable than v1.23.

Global Draw: off
Preview mode: Hacked
Sound: off
Record size: 1920x1080
Aspect: 16:9
Canon menu setting: 29.97fps ALL-I
Buffers etc: as default installation

Even with the display set to "hacked" so that it doesn't update the display, the only time I can get continuous recording is to set the aspect to 1.85:1 but if I try and use the display at all with "hacked 2" or "auto" then no joy.

One last thing... I'm using 2 cards. The SD card which has got ML installed only and the kompubay UDMA7 128Gb CF for recording only.

If someone has experienced the same problem and has a solution then that would be great as I was under the impression that a 5D III with ML should record 1920x1080 @ 29.97fps with preview with no problems according to the speed chart I've seen? (Do I need v1.23?)

Thanks again for all your help.

Cheers,

Mark.

#107
Hi Guys,

I've read this entire thread. I currently have a 5D II with ML installed (thank you!) and will be getting a 5D III soon to put ML on it. As far as I can discern the general consensus in this thread is to go with the v1.1.3 firmware and ML build over v1.2.3 but both nightly builds have been updated quite a lot since a1ex posted this thread back in March 2015:

http://www.magiclantern.fm/forum/index.php?topic=14704.0

May I just ask, is it still the general consensus to go with v1.1.3 over v1.2.3? I won't be doing much (if any) slow-mo shooting and will be predominantly shooting 23.976 the whole time. If I do need slow-mo then it'll be subtle so I'll probably just use a higher shutter and Twixtorize it in post.

My prime concern is stability and having no pink / corrupted frames (something I've experienced on the 5D II but have overcome by turning off Global Draw when recording and using a mask on my Viltrox DC-70 HDMI field monitor to frame the shot) but unfortunately this thread seems split between which is more stable. Some citing v1.1.3 as the most stable, others the opposite.

Obviously, my other prerequisite is to be able to use my HDMI field monitor without issue. I take it that's nothing to do with v1.2.3's ability to output "clean" HDMI and v1.1.3 will still output to a field monitor with no problems?

So, what's the general consensus now, at this moment in time, with regards to which build and firmware to go with?

Cheers,

Mark.
#108
Hi Guys,

Done a quick search but couldn't find this request so I thought I'd add it.

For those of use doing pieces to camera on our own (or "video selfies" to get down with the kidz  ;)) it would be great if ML could use Canon's IR remote in a more useful way on my 5D II.

At the moment, setting the 2s switch at the back of the remote (and setting the 2s on the camera) can start & stop the camera in video mode - fine. However it cannot focus - and if you're standing in front of the camera, it's pretty useless if there's only one of you.

What I'd like is the following:

1. In Canon's Live View menu, choose "Live View Face Recognition Focus"
2. Choose a setting in ML called "PTC mode" (or similar)
3. When in this mode, pressing the IR remote would cycle through the following:
- first press: focus
- second press: start record
- third press: stop record

Any subsequent presses would just cycle through that.

Would this be possible? At the moment I'm having to use Canon's EOS Utility via USB to a MacBook Air to focus and record in H.264 for all my solo piece to camera items but that means I can't record in RAW - which is my goal. Having this simple cycle of Focus-Record-Stop via the IR remote would mean I could then record in RAW and wouldn't have to carry my MacBook Air around with me for those times.

Thanks for your help,

Mark.
#109
Ah. Thanks @dmilligan.

Oh well, good to know. Worth asking all the same.  :)
#110
Thanks for the link, Walter. Good to know!
#111
@dmilligan: Thanks for the explanation! I hadn't realised that the endian was different. Hmm - how about this then (assuming that the endian conversion time you mention takes several seconds per frame in camera during recording only - as the processor is already busy - and not when idle) - a module which runs when you're not recording and converts the MLV file into a folder full of DNG files? There's plenty of time between pressing record that this could happen for the last shot that was taken - well, in my case anyway. :D Perhaps something like this:

1. User loads the "Create DNG folders instead of MLV" module
When in this mode, ML does not record to a single MLV file but instead creates a folder of individual frame files. Let's call them GND files as the endian is backwards.  ;)

2. When the user finishes recording the module shows the "busy" message on screen and converts the endianness of the frames that were just shot into correct DNG files.

That way we'd have a choice - either create a movie as an MLV (and have to convert to DNG in post) as we do now but with the benefit of immediately being able to press record once we've finished recording, OR... have a folder of separate DNG files + audio at the expense of some delay time after recording for the endian conversion process to happen.

Thoughts?

@Licaon_Kter: Interesting. Does MLV_REC continue the recording (audio and video) without any frames dropping?

@reddeercity: Yes, highest resolution with at least 24 fps (the 21fps choices are no good for my needs) are beneficial, but mainly I wanted to save time in post. I figured if the camera spat out a folder of DNG files instead of a single file then there'd be one less step to do in post, and with many many short clips, that one step could save a lot of time.

#112
I think you misunderstood my point. :) The individual image frames of the video do not themselves exceed 4Gb so if they could be saved as individual DNG "frames" instead of all of the frames wrapped into one MLV file, then the 4Gb limit would never be reached - in the same way as it's not reached when you you take 1000 RAW images in photo mode.

So, if shooting @ 24 fps for 1 minute for example, a folder would be created and within that there would be 1440 individual DNG files along with the audio WAV instead of *one* MLV file encapsulating the whole movie.

Does that make sense?
#113
Hi Guys,

I literally just stumbled across Magic Lantern a few hours ago and have been reading and playing with all it's glorious goodies ever since! Great stuff, thank you so much for everyone's effort.

One thing I noticed though is that I'm using MLVFS to mount the MLV file in Mac OS X which works great, but as it mounts the MLV file pretty much instantly I'm guessing that the MLV file is just a container for the individual DNG files and the audio, is that right? If that's the case, instead of recording to a MLV file in camera, would it be possible to just save out the individual DNG frame files to a new folder each time we press record instead of streaming them to the MLV container?

If that were indeed possible then there'd be no 4Gb file limit on my 5DII(!) :-D, plus the extra step of converting to DNG would be avoided.

I'm guessing that this ISN'T possible else it would already have been done, but thought I'd ask just in case.

Thanks,

Mark.