GH4 GH4 Luminance level setting

W

wwjd

Well-known member
[since I did not immediately find a thread about this using search...]
On GH4, any reason to NOT capture all video with luminance level from 0 - 255 ? If one's device CAN record 256 levels, why not allow it a full 256, and adjust levels in post as needed?
I'm aware many cameras are set 16-255, and BROADCAST wants 16-235.... still, in my original source, won't I want to give it as much space as the camera allows?
Maybe even tightening levels in post to 16-235 would allow you more levels, maybe even less banding?
My end goal is getting as close to shooting a hollywood movie as I can with this GH4.

I am kind of a newb to the tech stuff, so, please go easy on me. Set me strait as needed.

Referring to the LUMINANCE LEVELS setting in GH4 menu.
 
wwjd said:
[since I did not immediately find a thread about this using search...]
Maybe even tightening levels in post to 16-235 would allow you more levels, maybe even less banding?
Click to expand...

I think that's right theoretically but not sure. If you do shoot 0-255 make sure your editor doesn't cut off the extra bits when converting to 16-235:
https://vimeo.com/92613144
 
yeah, no, my editor will PUSH everything in instead of slicing off everything lower than 16, above 235, so that is nice.
I don't know what the "rules" are, but seems to me using the maximum potential of the gear is the best way to get the most out of yet.
I'll check out all the links

edit: k, this popped out at me: but the GH4's white is standard set to 255 to have more room to play in grading.
Exactly what I was thinking. You can reel it in later. I'll keep reading....
 
Last edited:
I just saw this. According to him its better to shoot 0-255...

the EOS codec hard clips at 100 IRE. Now if 255 = 100 IRE, I'm somewhat ok with this. However, if 235= 100 IRE, I'm not. In 8 bit color space you need every last step of range you can get. Why throw away 20 more steps in gradation, especially in the high end ? My JVC HD100 lets you shoot over brights to 110 IRE, and thats where I have the camera set. When you have to grade compressed 8bit material, every little bit counts. On the positive side, the codec does allow pure 0 IRE level blacks, adding 13 steps on the bottom end.
Click to expand...

http://www.steveoakley.net/template_permalink.asp?id=172
 
hey great find. I fully agree with that guy. even about the saturation. in 8 bit, squashed saturation is harder to bring back to life - my experience as well. We're NOT shooting with ARRI or REDs here, we have to make compromises that WORK. I plan to stay 0-255 shooting video (since I have this option now) and see how things go. I bet it will be just fine as I can adjsut in post. As long as I am capturing everything I want in the shot, reducing sat or contrast won't help. Now, if the shadow detail is crushed or whites blown, yes flatten it out to fit.

But I think every littel BIT* helps. *pun intended :)

Hoping more experieced peeps sound off about this. I think it is kind of important in this bleeding edge camera.
 
That Steve Oakley article was great, thanks for posting it. Along with wwjd, I'm beginning to think that Saturation 0 is a better setting.
 
Agreed - that article made me question the validity of doing anything to the profiles at all other than turning down sharpness and noise reduction. Gonna try shooting with CineLike D as is (no highlight/shadow curve adjustments, no contrast changes, etc.).
 
k, i been reading and asking other places, and come up with some things. First, CINE-D already flattens things a pinch so I don't think I need to use shadow/highlights curves. I might use MASTER PEDESTAL LEVEL a few times to raise the bottom zero to 2 or 5 above... maybe... and I think maximizing this camera means making settings changes PER the scene being shot. Or find a medium effectiveness to leave it all for run and gun stuff. Example, shooting a dark scene, maybe I'll bump the pedestal up a couple to make sure I am seeing the dark detail I want to see (I love dark details!)

I am still not sure if this camera shoots a full 256 levels or stretches a lesser amount OVER the 256. I'm going to try to video a grey gradient and then check it using video scopes in Vegas, but recently discovered they are not all that accurate. So I'll go with best guess there.

FCPX dropping bits: I would say find the scopes in there if it has it and check. A simple histogram is a good place to start. It should list levels from 0 to 255, and you can see a stait line up/down at 16 if it cuts things out.
 
here's my VERY UNSCIENTIFIC, and INEXPERIENCED tests:

https://copy.com/Oas0yd6opE47

1. the original gradient file I stole off the net somewhere (I don't know how accurate it is, don't know how accurate my monitor is - like I said, all very unscientific)
2. scopes from Vegas Pro 12 for the original file
3. scopes for the camera filming that file on screen, set to 0-255 CINE-D
4. camera set to 16-235 CINE-D
5 16-235 STANDARD setting (I was curious)

I viewed the gradient file full screen, set screen to darkest and brightest settings to fill spectrum, pointed camera at screen and defocused it, dialed in the settings to maximize the GH4 onboard histogram.
My settings were CINE-D profile, f5.0, shutter speed 10, iso 200, WB set to whatever looked the most like the same color from the screen to the GH4 LCD

FWIW: to me, it looks like the 0-255 on the camera does provide more range of bits or lumens or whatever, based on the count on the scope in vegas going from 0-255. I have a couple plug ins that crunch down
levels from 0-255 to 16-235 but NOT just chopping off the ends, which would look hideous. Or I'll leave it 0-255 while editing for the IMAX release. ;) ;)
 
Thanks. It looks like 0-255 is well outside normal IRE (the top chart?). So I guess its up to the editor to recognize that and crunch it down. Btw AfterEffects I hear is not supposed to be able to do that, only Premiere. Too bad because I prefer AE.
 
I just posted another thread about this but it relates directly to this thread. I discovered that the "16-255" Luminance Level is the one that is causing all the issues with unnecessary clipping of highlights. When the Gh4 is set to "0-255" instead, Premiere maps it it correctly and highlights are not clipped.
 
On the GH1 and GH2 (and presumeably the GH3), AVCHD files are tagged with Metadata that indicates that the luminance ranges from 0-255. You can confirm this by putting Adobe After Effects in Color Managed mode, and setting its internal color space to "Rec 709 (0-255)". After importing an AVCHD file, you can check its Interpret Footage dialog, and AE will indicate that its colors have not been remapped, because the video's color space matches AE's internal color space. If you instead set AE's internal color space to "Rec 709 (16-235)" and reload the file, AE will remap the colors, compressing the file's 0-255 range into 16-235. (Of course, if you are in 32-bit pixel depth, AE will do this in floating point with no loss in image quality.)

With the GH4, I'd hope Panasonic has taken care to mark each video file with metadata indicating its correct color space setting. If not, users will need to keep track of the proper color space for each clip, and override After Effects' interpretation of color data when necessary.
 
Lpowell said:
On the GH1 and GH2 (and presumeably the GH3), AVCHD files are tagged with Metadata that indicates that the luminance ranges from 0-255. You can confirm this by putting Adobe After Effects in Color Managed mode, and setting its internal color space to "Rec 709 (0-255)". After importing an AVCHD file, you can check its Interpret Footage dialog, and AE will indicate that its colors have not been remapped, because the video's color space matches AE's internal color space. If you instead set AE's internal color space to "Rec 709 (16-235)" and reload the file, AE will remap the colors, compressing the file's 0-255 range into 16-235. (Of course, if you are in 32-bit pixel depth, AE will do this in floating point with no loss in image quality.)

With the GH4, I'd hope Panasonic has taken care to mark each video file with metadata indicating its correct color space setting. If not, users will need to keep track of the proper color space for each clip, and override After Effects' interpretation of color data when necessary.
Click to expand...

Ok, so I went through this process with AE, and found the color management items indicated, found the 'embedded profile' was Rec 709 Y'CbCr, as opposed to Rec 709 (16-235).

I exported a frame, to a psd file, and found that the full 0-255 values seem to be present in my 'test shot(grey card, black cases, small blown out area of my meter plexiglass cover, etc...) all mapped to what I expect.

So, the question is... how does that relate to the Premiere IRE waveform display, which shows 0% to 100% for my imported 'movie' file (MTS), but when I use a 'constructed' file, with the a pattern of 0-255 values showing grey values and 0, 255 bars, etc... I get a below 0% reading for image value 0, and a +100% for the 255 "super white" found in the constructed MOV file..

Does this mean that Premiere, seeing Rec 709 in the MTS file, does the conversion to 16-235 for the 'video' material, where as uses 0-255 for the 'constructed' MOV sequence that has no assigned profile?

Further, does this mean that one should not do much 'image work' in Premiere, but how does one export a cut sequence to work in other tools and still maintain the full 0-255 range?
 
Last edited:
you guys have advanced beyond me at this point... I'm just curious about utilizing the maximum amount of range inside the GH4, before editing starts. But I understand what you are saying is totally related to the levels, so carry on. :)

Does all this mean 0-255 is best setting to shoot video in this camera?
 
Lpowell said:
On the GH1 and GH2 (and presumeably the GH3), AVCHD files are tagged with Metadata that indicates that the luminance ranges from 0-255. You can confirm this by putting Adobe After Effects in Color Managed mode, and setting its internal color space to "Rec 709 (0-255)". After importing an AVCHD file, you can check its Interpret Footage dialog, and AE will indicate that its colors have not been remapped, because the video's color space matches AE's internal color space. If you instead set AE's internal color space to "Rec 709 (16-235)" and reload the file, AE will remap the colors, compressing the file's 0-255 range into 16-235. (Of course, if you are in 32-bit pixel depth, AE will do this in floating point with no loss in image quality.)

With the GH4, I'd hope Panasonic has taken care to mark each video file with metadata indicating its correct color space setting. If not, users will need to keep track of the proper color space for each clip, and override After Effects' interpretation of color data when necessary.
Click to expand...

So for AfterEffects, is it safe to say this is a non-issue since illegal values automatically get re-mapped? (AE works in 32-bit too, i think).
Or does this apply only to AVCHD files which get tagged (.MOV / H.264 get tagged also?)


WWJD said:
Does all this mean 0-255 is best setting to shoot video in this camera?
Click to expand...

You could watch the input/output levels like he does in the video if shooting 0-255.
 
j1clark@ucsd.edu said:
I exported a frame, to a psd file, and found that the full 0-255 values seem to be present in my 'test shot(grey card, black cases, small blown out area of my meter plexiglass cover, etc...) all mapped to what I expect.

So, the question is... how does that relate to the Premiere IRE waveform display, which shows 0% to 100% for my imported 'movie' file (MTS), but when I use a 'constructed' file, with the a pattern of 0-255 values showing grey values and 0, 255 bars, etc... I get a below 0% reading for image value 0, and a +100% for the 255 "super white" found in the constructed MOV file..

Does this mean that Premiere, seeing Rec 709 in the MTS file, does the conversion to 16-235 for the 'video' material, where as uses 0-255 for the 'constructed' MOV sequence that has no assigned profile?
Click to expand...
Internally, Premiere Pro uses a luminance scale that ranges from -20 IRE to +120 IRE, regardless of whether an imported clip is encoded full-range (0-255) or studio-swing (16-235). Unlike After Effects in Managed Color mode, Premiere does not remap the luma levels in each clip to match its internal color space, it finesses the discrepancies behind the scenes. The way it does this is clever but often confusing.

H.264 videos recorded by DSLR's use the Rec 709 color space with an embedded metadata tag called the Full Range Flag. When clear, this flag indicates that the file's 0 IRE level is set at 16 and 100 IRE level is set at 235. When set, the Full Range Flag indicates that 0 IRE is set to 0 and 100 IRE is set to 255. These levels are expressed in 8-bit pixel depth, which is the case for all Panasonic. H.264 videos.

In Premiere, however, the pixel depth of each Sequence can be set to either 8-bit or 32-bit floating point. (This is set by the Maximum Bit Depth flag in the Sequence Settings dialog. You can optionally set Maximum Render Quality for each Sequence as well. Doing so will require more RAM for compositing and will likely slow down Premiere's real-time performance.) Bit depth can have a significant effect on the quality of rendered transitions and effects, some of which require the Sequence to be in 32-bit mode.

When you drag a clip into a Sequence's timeline, Premiere takes note of the clip's Full Range Flag, and sets its 0 IRE and 100 IRE levels accordingly. As a result, both Full Range (0-255) clips and Studio-Swing (16-235) clips will be displayed as ranging from 0 - 100 IRE. Even when you stack two such clips in overlapping tracks, Premiere will calculate their relative scaling and opacities and display the composited levels on the same unified 0 - 100 IRE scale.

With this system, confusing results can occur when imported clips are flagged incorrectly. A 0-255 video whose Full Range Flag is clear will be interpreted as ranging from -7.5 IRE to 110 IRE. Likewise, a 16-235 video whose Full Range Flag is set will be interpreted as ranging from 7.5 IRE to 92 IRE. While this is not a problem for Premiere to accommodate within its -20 IRE to +120 IRE internal range, it can produce incorrect contrast and black and white level clipping in the exported results. Unfortunately, in Premiere you cannot manually override a file's incorrect Full Range Flag in its Interpret Footage dialog. (You can, however, do so in After Effects and export the corrected file to Premiere.)

These scaling discrepancies are moot when working with a Sequence in 32-bit mode, and are usually negligible in 8-bit mode as well. When rendering a video, however, you definitely want to do so in Maximum Render Quality at 32-bit pixel depth. Adobe gives you separate options for setting these flags solely for rendered media output in its Sequence Export Dialog. As a result, there's rarely a need to put a Sequence into 32-bit mode while editing. You can work in 8-bit mode seamlessly with all types of footage, and composite it all down in 32-bit mode in your final render.

With respect to the GH4, I'd recommend sticking with full-range 0-255 format, as it's the standard used in other DSLR models from Panasonic, Canon, and Nikon. Studio-swing 16-235 format would be more useful for compatibility with older videocams that shoot in that format. I'd stay away from the oddball 16-255 format, as I suspect it's not properly identified by the video file's metadata tags.
 
You must log in or register to reply here.
Back
Top