Topics

[UPDATE:]

UPDATE: It is definitely to do with Exposure Override when at certain apertures (f/6.4 & f/4.5 for me). Easiest solution is to ensure that you have Exposure Override set to "OFF" and you shouldn't come across this problem.

---

Hi guys,

After a search all I could find on pulsing images was a long time ago here and that seemed to be for H.264.

Having just returned from a day's shoot (in the rain) for a short film I'm working on, I was disappointed to see that a long shot (a 3 minute shot of fields and clouds) was "pulsing" at around 100bpm (according to my tap tempo metronome) in MLRawViewer. So, I thought I'd see what it was like after conversion in AE, and same problem.

It's not a "flickering", it's definitely a "pulsing" phenomenon so MLVFS's deflicker feature doesn't seem to do anything with it no matter what values I put into the deflicker box.

So, a couple of questions, if I may...

1. What I would like to know is the cause of this so that I can avoid it in future. According to the link above it's to do with setting Exposure Override to "ON" and using specific aperture settings. Is this correct? If so I'll just set that to "OFF" and not worry about it (I have currently got it set to ON). However, if it is something else then I'd appreciate it if someone could provide pointers for how to avoid this pulsing in future as the footage I have as it stands now requires a re-shoot unless I can somehow fix it.

2. Is it possible to fix this so I don't have to re-shoot? I've tried AE's Color Stabilizer set to "Brightness" and that removed the pulsing but left flicker, so I'll play around with that in the meantime, but if anyone has a magic trick for fixing this pulsing I'd dearly appreciate it.

Incidentally, the lens used today was a Canon 35mm f/2.0 IS and the footage with pulsing (I've not checked all footage yet) was shot at f/6.4.

Thanks for your help with this,

Mark.

Hi guys and gals,

I've just uploaded my Sensor Crop vs DOF video which you may or may not find interesting, but even if you don't, the footage was shot on a Canon 5D III with Magic Lantern RAW at 1920x820 and can be found at the following times if you want to skip the footage:

2:56
4:45
11:33

Direct link is here: https://www.youtube.com/watch?v=MC6nM3S4S1k

I'm really loving the After Effects approach to convert the RAW cDNGs via MLVFS, I must say! 

Cheers,

Mark.

Hi folks,

Has anyone used the Tascam DR-701d with a 5D III / ML RAW? I'm intrigued to its HDMI IN feature which is supposed to start recording audio when you press record on the camera so both files apparently start at the same time.

Not that syncing audio is a pain, just curious if it works or not.

Cheers,

Mark.

[@DeafEyeJedi]

After my comparison video last week between RAW and H.264 where the H.264 files seemed to hold up pretty well to the RAW files providing you didn't waver too far from what was recorded in terms of color grading, I thought I'd download the Adobe CC trial thing to see what difference ACR would do to the RAW processing - if any - based on a comment @DeafEyeJedi once made on this forum where he mention that AE produced better results.

Holy bajoly. He wasn't wrong! There is quite literally a world of difference between processing the RAW files in DaVinci Resolve and in ACR through After Effects. The amount of control you have before the file gets baked into pixels is astounding. I've only ever previously used ACR for stills photography and that was way back in CS5. Since then I moved to Lightroom for my stills so using ACR for video is quite an eye opener.

The images are sharper (Resolve's sharpen filter is pretty basic to say the least) and just seem to "pop" more, not to mention how easy it is to manipulate the image in RAW-land and how much more control you have on the RAW file before processing. A powerful feature I like is the fact that I can dial in some luminance NR if needed before the file is processed where it's using all the 14 bit color palette to produce far better results than applying NR after the fact. Well, at least to my eyes.

Anyway, I just thought I'd share this personal revelation but of course the downside to this is now I can see how powerful this workflow is, and how much more beautiful the images look through this pipeline, I'll begrudge ever shooting h.264 ever again! 

Mark.

[EDIT:]

Hi Guys,

Following on from this thread I thought I'd upload my findings between RAW and H.264.

https://youtu.be/cMRhmigjbnI

Enjoy! :-)

Mark.

-----

Shot using Canon 5D III with
Canon 70-200L f/4 IS for the outdoor shots
Sigma 35mm f/1.4 Art for the indoor shots

EDIT: I'm not sure why the YouTube embedding isn't working. Am I doing it incorrectly?

[QUESTION:]

Hi guys,

QUESTION:
Is ML doing something behind the scenes on a 5D mk III to improve the quality of the H.264 video over what I'd get from Canon's All-I implementation or would the result be exactly the same as if ML was not installed? (Yes, I could test this but it's late here now and a pointless test if one of you nice developer folk already know the answer.  )

----

BACKGROUND:
After watching Act of Valor the other day for the first time - which was shot 70% on a 5D mk II / H.264(!) as I'm sure most of you are aware - and marvelling at the quality they got out of H.264, I've been running some test of my own between RAW and H.264 on my 5D mk III.

I have to say that I'm gobsmacked at how good the results are. Having seen so many RAW vs H.264 comparison videos on YouTube I previously disregarded using H.264 because all of the videos I saw showed a night-and-day difference between RAW and H.264. In my tests however I can hardly tell them apart (I must be doing something wrong!). Only in a few side-by-side shots can I see the extra detail but it's not by much. In isolation I'd be none the wiser.

Anyway. This got me thinking - I've been shooting H.264 through the Magic Lantern UI so that I can use focus peeking etc. and hence my question. I plan on doing some more tests tomorrow - low-light stuff mainly at high ISO etc but right now I'm encouraged by how good my findings have been and how incredibly useful Magic Lantern is even without shooting RAW.

Thanks for your help and support.

All the best,

Mark.
p.s. Is there a reason that Auto ETTR only works for shooting RAW and not H.264?

[70%]

Hi guys,

I propose that the spotmeter has a new function called "EV Difference" and how it works is as follows:

1. Meter for foreground with the spotmeter
2. Press "some button" (selectable in settings?)
3. Meter for background
4. Press same button again
5. The EV difference is displayed on the screen (e.g. "-3EV Difference" or "+1.5EV Difference")
6. The number stays on screen until "the button" is pressed once more to cancel the function.

Here's the background...

As the spotmeter goes from 0-100% (or IRE) I think of it as going from 0-11.5 stops in my mind (5D III) and therefore if I have my foreground lit at 70% it equates to approximately (11.5/100)*70 = 8.05 EV so if I want the background to be 3 stops darker I will need to light it at (8.05-3*100)/11.5 = 43.91% according to the spot meter.

That's all well and good but I can't do that math in my head hence the request.

I'm guessing that for this to work accurately, ML will either have to know the maximum EV that the camera has at a given ISO so that it can "assign" that to be 100%, or somehow figure it out? For a rough and ready solution there could just be a "Max EV Value" that the user could select from and that would probably do for most cases.

Does this make sense? Thoughts are appreciated.

Cheers,

Mark.

Hi guys and gals,

Since playing around with 1:1 crop mode a few weeks back I've been pseudo-obsessed with thinking about perhaps re-shooting the three short scenes I've done so far in my movie @ 2560x1072 (instead of the original 1920x804 that they were shot at) with the mind-logic that it should (surely) look better scaled down to 1080p rather than recording 1080p natively. Surely.

However, none of this was with any testing, just hypothetical logic in my ickle brain, so tonight I thought I'd try a quick and dirty shoot to test out the noise between 1:1 and 3x3 binning after A1ex's reply in this thread. It's clear to see at 3200 ISO that 5x zoom mode is decidedly more noisy. In fact, in contrast the 1920x804 non-crop image is very useable at 3200 ISO. Still plenty of detail.

Note: All I'm testing here is noise.

All I did with these shots were bring them into Resolve, add a Rec.709 color space and desaturate them to make them legal in the vectorscope:

1920x804 @ 3200 ISO on 1920 timeline


2560x1072 @ 3200 ISO scaled down to 1920 timeline


1920x804 @ 3200 ISO scaled up to 2560 timeline


2560x1072 @ 3200 ISO on 2560 timeline


1920x804 @ 3200 ISO scaled up to 4K timeline


2560x1072 @ 3200 ISO scaled up to 4K timeline


1920x804 @ 3200 ISO scaled up to 4K timeline (left corner)


2560x1072 @ 3200 ISO scaled up to 4K timeline (left corner)



Now, obviously there are differences in the setup of this test. Even though both times I focused on the nail varnish bottle (not mine, honest) there will be differences in the DOF as these tests were done using the same lens with me just moving back a bit to try and frame the shot the same as the non-crop image in 1:1 crop mode (mainly because I don't have a super-wide lens to get the same perspective in 1:1 crop). Not scientific at all.

If I was doing this scientifically (and if it wasn't so late now) then I'd use my 35mm lens for the 1:1 crop mode and then match that shot with a zoom lens at around the 100mm mark without moving the tripod etc. I may do that tomorrow if I've got time and add a selection of different ISOs into the mix too.

Right now though, at 3200, and with regards to noise, even with the different DOF it looks clear to me that the advantages the extra photons bring to the table in 3x3 binning mode ("normal" recording) far outweigh the extra pixels of the 5x zoom recording. Not just on the 1920 timeline but when upscaled to 4K too. It'll be interesting to see if that's the case at lower ISOs too.

Not rushing out to re-shoot that footage at the moment though...

Hi guys,

It occurred to me the other day when doing some high-fps test shots on my 5D III with ML that the vertical image is "squeezed" and needs to be "unsqueezed" in post to bring it back to 1080p. This lit a bulb in my head. A lightbulb, if you will... Ping!

So... would it be possible to squeeze the HORIZONTAL pixels by 1.34x to mimic the anamorphic lens squeeze so that it's still recorded at 1920x1080 but when it's stretched back out it becomes 2581x1080? If it's possible then perhaps a variable multiplication factor choice could be used, i.e. 1.34x for 2.39:1, 1.32x for 2.35:1 etc.

If this is indeed possible then the higher pixel count would be a great boon for ML and cinematography as a whole and would do away with the need of buying an anamorphic lens. Obviously you forego other advantages of an actual anamorphic lens such as horizontal flare, shallower DOF etc but at least we wouldn't be throwing away pixels when shooting at 2.35 / 2.39. Ya know?

Answers on a postcard to...

Mark.

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.

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.

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.