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 - saf34

#1
Thanks Skinny, interesting to hear your take on it - and super useful point on using vertical resolution to reduce the necessary crop in post.

To answer the question: the 5% was an arbitrary number that was 'enough' really, rather than how much I'd want to zoom.  To explain: the idea came from the question "suppose we regard the standard 1080p preview as the only 'usable' monitoring experience for a particular use case.  Then, what extra features, however small, can be squeezed out".  I have a series of films that are all handheld, incognito and require good on camera monitoring.  And so I wondered about whether it was possible to have the standard preview while still filming higher resolutions outside of the preview. 

Regarding stabilization, it was really all just about eliminating the need for zoom.  If using a translation-based stabilizer in post - as I often do - then the zoom is the only aspect that 'reduces' the resolution below 1080p.  Perhaps it's a really negligible point - it probably is - but I was curious nonetheless, as I always work in post with the zoom disabled and watch the 'dance' of the black bars.  The other practical point is that I monitor from the LCD screen when using 1080p (it's always to film agile moments) and so framing with stabilization in mind means a smaller image.  Which again, is okay, but I wondered.

Finally, there's something about 1080p which feels like a lower zoom bound these days.   For context, I have a RED Komodo, two 5Ds with ML and a couple of DJI gimbals.  It's probably just psychological, I admit, but there's something about knowing these 5Ds have the capacity to do 3.5K but having to zoom to less than 1080p for handheld stabilization that would be fun to get around.  Yes I can stick the 5D on a gimbal, but it changes the experience in a variety of ways, not least the incognito way of filming you lose. 

In retrospect, if this were easy to do then I guess someone here would have already set up a super 16 preview while filming 3.3K or similar.  Still... just a thought!





#2
Thanks theBilalFakhouri, got it.  I had my fingers crossed it was simply that no-one had ever thought the standard preview was useful while accessing more of the sensor - but your explanation clears that up perfectly.  Not a straightforward thing to do then!

Ultimately then it might be that there's not a huge amount of benefit over putting an overlay guide that blacks out 5% of the top, bottom and sides of the standard preview for framing and upscaling that 5% in post.    Or using an existing higher resolution with compromised preview but again with the overlay, and downscaling.

I'm definitely up for at least trying some reverse engineering of the liveview though.  Any starting pointers very welcome!  And goes without saying: all your previous work is super-appreciated.
 

#3
Hi Bruno, please don't misunderstand: I'm asking as it is a feature I would consider *contributing* to if it were feasible.  I don't take for granted the ML community's work, and I'm not asking anyone to do this for me.  Just if there is a limitation based on the data streams (no point me going down a dead-end path).

Yes, using a gimbal is great, but there is a reason digital IS exists in recent cameras too.

Stewart
#4
Hi, so, I have an idea and wanted to ask anyone who has already got their hands dirty with coding if it would be feasible.

The idea: Alex's 1x1 1080p 3x crop mode with correct preview, but have it record an extra 5% (say) of pixels from the sensor.  Why?  Then you could use stabilization in post without needing to zoom.  It would effectively be like digital stabilization in camera.

This might not be of interest to everyone, but it strikes me there are really two ends of the spectrum on how to use ML.  (1) seeking highest resolution in some way, and willing to compromise on preview, continuous recording, overexposed areas. (2) Using it as a Super 16 HD or FF camera with 14-bits and seeing what extra features can be squeezed out.

Personally, I use it almost all the time as (2) and it is amazing.  Am curious if the stabilization 'border' would be possible though!

What do you guys think?
Stewart
#5
I've just finished my first full length project using just 5D mk iii with ML, after years as my only camera.  It has been the most amazing ride, and I can't express enough how grateful I am to everyone involved. 

I'm curious though - having never used anything else - what others' experiences are of the entry level commercial cinema cameras.  A RED Komodo, or a Canon C300 for example.  Even a Fujifilm GFX100.  As I'm acutely aware I'm going to have to deliver 4k at least on my next project.

Has anyone experience with these to share, in light of the amazing ML 5D? 

https://vimeo.com/558756617


#6
Thanks for this - I'm certainly confusing myself with their unspecified log request, but on I think I'm okay re the Cinelog workflow.  The point of showing them the Cinelog-C was that if they wanted to use that source I'd have bought a second licence and given them a LUT to use to import.  The footage is good- our home grade using Cinelog LUTs for import looks similar quality to your link.

My confusion was based on that they asked for 'log' and I didn't know a) what log they meant (and they were closed) and b) the colorspace process in Resolve, having always used ACR.  I still am confused (and hopefully will stand corrected) about Cinelog with Resolve 17.  As the supplied transform from Cinelog for importing into Resolve (BMD Film -> Cinelog-C) is way off as compared to ACR.  My guess was that things have moved on since 2017, with BMD Film Gen4 being a different space than the import transfer that was written in 2017.  In any case, setting up Resolve as instructed on the Cinelog site isn't anywhere close to the ACR method. 

But yes, I'm definitely missing the point on what colourspaces we go through when importing DNGs directly into Resolve 17.  Personally, I edited straight from the DNGs in Resolve, and it was set to use the camera metadata - which resulted in a good picture.  But I was nervous as it seemed I was therefore already in Rec709 and I'm unsure of the consequences of going into Rec709 and back again to something like Arri Log.  So it made sense to me that we probably shouldn't go to Rec709 at all, and go straight from DNG to ARRI Log or whichever.  But I could see no way in Resolve to import the DNGs straight to BMD Film with the camera tone curve in order to output any of the possible log formats.

Incidentally, when I did eventually get to ask them about which 'log', they answered that I had to use .RD3 files and sent me a sheet of instructions.  But they also told me that ML is "only ever 8-bit up-ressed" which of course it isn't, nor would they accept DNxHR 444 - only ProRes 4444.  So I think I was a bit of a hassle for them from the start...

Anyway, many thanks for helping me out with this.  Really appreciated!
#7
No, they didn't
#8
Hi guys, thanks for the replies.  I'll put down my experience here, just in case it is useful to anyone else. 

First thing to say: I think they found Magic Lantern a bit confusing.  I delivered footage both as DNGs and Cinelog-C (using Adobe Camera Reader and the 5D mk iii Cinelog-C log profile).  They looked at both, and preferred as a rule to go with the DNGs just for being closer to source.  It was filmed at 1250 ISO @ 2.8 in low light, so there was only around 9EV dynamic range I think - so from their point of view I think they just saw it as 8-bit footage from an 8-bit camera really.  The best looking DNGs I found to be with 3x3 CS and aggressive bad pixel fixing, so I did this first.

The PITA was actually all with the NLE in Resolve.  I used Resolve for the first time for this, downloading v17.  I did the editing before I knew it would be graded, and so when we got in touch I found out they were on v16.  Fine, but I'd been editing in multicam objects and doing manual sync using retimes.  Then I discovered that you can't export the retimes.  Nor can you flatten a multicam clip if you've retimed it.  So no EDLs, now XMLs.  I could relink all the media to the Cinelog files instead of the DNGs and export a cinelog preconformed, but in the end they just installed Resolve 17 for me which was pretty accomodating...

EDIT: to answer "why log", basically they wanted a preconformed version in case they couldn't recreate any effects (stabilization, retimes, zooms, translations).  So in future I'll probably make sure to render down some cinelog masters from the get-go too

So if it wasn't for the mistakes with the NLE I made it would all have been fine basically!




#9
Hi guys, am hoping you might be able to help me out with my confusion.  I've been asked to send over a heap of footage in log prores 4444 to a grading house (huge shock).  It's currently in DNG form (converted from Canon 5d mk iii MLV) and I'd been editing in Resolve in Rec7090.

What is a safe way to get my timeline nicely output as log without losing data?  I was happily working in Rec709 until this bombshell hit, but now I'm confused about curves and color-spaces.

Specifically: can I import my DNGs into Rec709 in Resolve and not lose data (aren't we in 32 bit at this point?) - and then use Color Space Transform to output some log space version (BlackMagic Cinema-camera film Gen 4 for example?) 

Or should I go into the Blackmagic Film space first (but then which transform to use?  The output as is looks suspicious)

Any help would be really appreciated.  I have to send this out on a drive tomorrow...

Thanks in advance,
Saf34
#10
Thanks Kharak, sounds like it's about time I get a new workflow down.  Thanks for the tips, which I'll start playing around with.  Coming from a photoshop background, Cinelog in ACR was perfect at the time, but has perhaps made me a little lazy re Resolve etc!
#11
Thanks for answering - much appreciated.  My workflow is a quick setup from a few years back when I bought Cinelog-C:

1) Copy MLVs to folder on internal fast SSD (D:\MLV\*.mlv)
2) Run a batch file that runs mlv_dump on each MLV file in that folder and deletes the MLVs in turn.  So now I have D:\DNG\[filename]\*.dng
3) Import each DNG folder into After Effects using my Cinelog profile, and render out Prores 4444 masters D:\cinelog\*.mov

Originally this was all uncompressed 14-bit MLVs so this was my own way to 'compress' my masters.  Recently I've moved to lossless MLVs (why not) and because my MLVs are now 14-bit lossless and my DNGs are uncompressed, the intermediate DNG folder is twice as big as my MLV folder.  The final Cinelog Prores files are a little smaller than the original MLVs, but mainly they're just a convenient habit for grading.

Then I always just did a render of rushes from my final log masters.  But now - with such big sessions and multicam taking ages to transcode - I'd like to er... rush ... them through first *ahem*.

I'm just setting up MLVFS for the first time, but haven't read how much storage is used when mounting a folder of MLVs.  I.e. is the entire folder in temp files as DNGs or perhaps individual DNGs are pulled one by one from the MLV when required?  If the latter, this is much much better than my hotch-potch method.  Plus is seems it gives you a timestamp on the DNGs and Wavs, which is useful.

MLVApp is seductive too, but takes a long time to load in large amounts of footage.  Waiting for a frozen screen to unfreeze is stressful! 

Anyway, that's the context.  I'm a bit late to the workflow party, but ideally I'd manage to get a bunch of log masters, a bunch of Rec709 H264 rushes, and have them all with a timestamp rather than have to handbuild my timelines.

I'm veering towards MLVFS right now, but am still getting set up.  So I'm definitely interested to hear other possibilities from more experienced MLers!
#12
Quick question for those shooting *lots* of footage:

What is the space-efficient way to quickly render out low-resolution rushes for the director?

For example, yesterday I shot 5TB of MLV (14-bit lossless).  If I mlv-dump this lot to dng (my usual workflow to Cinelog) it's around 10TB. 

So my questions:
  - If I mount everything using MLVFS to render out low res, does this need 10TB of temp files?
  - If I import the whole lot to MLVApp, does this require 10TB of temp files?

Or how do others do this for long shoots?

Best,
Stewart
#13
Agh, yup, was too easy to be true... thanks.   Just thinking out loud: it's kind of useful as a 'magic focus'... almost... as it's not centered.  I'm not there yet with how the code works: is this the nature of the stream or could I tweak something to center it?
#14
Hi guys, just been reading through this topic for the first time after a long time on the stable branch.  Just a quick sanity-check for me I wanted to throw in re. monitoring (pprompted by @AlanSmithee's comment above).

In anamorphic mode, my HMDI out (on 1.2.3) is giving the unsqueezed live feed.  So this means with an external monitor with anamorphic desqueeze we can film RAW *and* monitor with a fully functional preview (i.e. not the black and white one shown on liveview)? 
#15
Thanks for answering Kharak.  I was being a total dumbass: it only appears when being played back in a resized VLC window...  It's been a long week!
#16
General Development / Help me identify the banding issue?
December 07, 2019, 04:16:54 PM
Hi all, just a quick one.  Could anyone help me identify the banding issue I'm having with my MLV file in the highlights?  I'm almost certain it's been discussed, but I haven't managed to find an image example that helps me identify it.

My workflow:

Canon 5d mk iii > mlv_dump to DNGs > After Effects

The mlv_dump vertical stripes fix seems to be working in mlv_dump fine.  And these stripes are not just vertical.

Image (zoomed) attached.

https://www.dropbox.com/s/3q0spbrfje1bvel/Screenshot%202019-12-07%20at%2015.04.22.png?dl=0

Thanks in advance!
Stewart
#17
Thanks to both.  MLV App worked out well for me. 
#18
Hi all, I had a little look around the forum but can't see an immediate way to do this.

I'd like to quickly extract the wav audio from my session's MLV collection. 

The reason?  After I did my 24-hour batch MLV -> DNG using mlv_dump, I have audio missing from a bunch of files (but not all).  So I need to investigate.

Is there a command-line tool for doing this?  I half wondered if there might be an option on mlv_dump but I can't see any.

Any help really appreciated.
Stewart
#19
Hi, thanks for the responses.  Makes sense! 

@Andy600
Just to get something straight in my head: lets say I have a scene where exposing to the right leads to an overexposed frame.  If we accept that exposing to the right and reducing gain in post is preferable to exposing 'correctly' at source from a noise persepctive, then the correct workflow is...

1) Expose to the right in RAW
2) Import DNGs ACR->AE at this exposure using Cinelog-C ACR profile
3) Reduce exposure offset in AE to the desired 'correct' exposure (as described in the manual)
4) Generate DNxHR 444 master

Did I get that right?

#20
Hi all, I shoot handheld footage on a Canon 5D mk iii.  And I'm wondering how others manage colour while previewing from live view?

Basically: I find that when I import footage that looks good on Live View, I always need to bump up exposure by about 1EV to get a similar exposure in the shadows in REC709.

I can just expose to the right of course.  But then I can't make aesthetic decisions from the live view in real time.  I'd much rather be able to see something that is close to the final result for this kind of work.

My workflow is:
- shoot RAW,
- Adobe ACR to Cinelog-C (usually have to bump up exposure by 1EV)
- After Effects to Cinelog-C DNxHR 444 master

How do other people deal with this?  Or am I missing something (I am pretty new to this)?

Many thanks,
Stewart
#21
Hey, first post on the forum after a year using ML on a 5dMkiii.  Has changed my life!

If anyone has a minute, I'd love any recommendations on a desktop PC I need to build.  My workflow is:

- Film handheld 14-bit 1920x1080 uncompressed MLV
- Go home & transcode to 12-bit Cinelog DNxHD for storage
- Edit, stabilise, grade & render H264 in DaVinci Resolve 14

At the moment I am using an old i5 HTPC, grading in Premiere Pro. The problem: I have to leave it for a day to render the DNxHD and can't even play back while grading...

Any opinions on the best value DaVinci Resolve System for a sensible transcode and realtime render?