@DirkVoorhoeve The footage itself shouldn't be showing any values under 16 or over 235 though right ? We are discussing the interpretation of the GH2 footage in ITU 709 here still yeah ? Am I missing something here, I've had a weird week. Pressing that +7.5v button is going to give you a 'what if it was broadcast' waveform. Correcting out the milky look as people have described is going to move it into full swing, this is alright to do obviously, but you probably want to convert it back if you do intend to broadcast it. If you only have a computer monitor then moving footage into full swing may be the only way to work.

Basically the YC waveform has little to nothing to do with this discussion really.

As far as it goes with Premiere interpreting GH2 footage I did a quick test with Nuke. I used ffmpeg to move an mts file into a mov container using -vcodec copy (cos Nuke won't read mts files) and also exported a sequence of tif stills from the same mts file using Premiere. The results were identical. Is this a valid test though, can someone with more ffmpeg experience confirm this, or will copying into a mov container alter anything else ?

@mozes alt-prtscr will only grab the active window.

@towi Why not assume it is not remapping at all, but just converting from the sensor's RGB values directly to the studio-swing Y'CbCr colorspace?

I think anyone who wants to: take their copy of Premiere, or download the free trial, set up two 1000 watt lights and video something really black (angled) and something really white (facing), expose three vid clips bracketed since the histogram overexposes slightly and you lose the blacks (for very good reasons, given the sensor), import into Premiere, uncheck the 7.5 IRE button, set the display level to 100 (to show the transients), turn off the chroma, and take a look. I'm sure--and I mean this--that ppl who have been doing this way longer than me can spot something I might miss. You have to of course use a scene that is brightly lit so that you have to pull down the exposure to get the whites in, otherwise you won't know if you are videoing the full dynamic range.

@DrDave ah right, heh, fair enough then. My concern came from the suggestions from the flowmotion thread being so similar to what you described, I got the impression there was some wacky workflow that had become accepted by some.

DrDave said "IRE:Adobe chooses to map IRE to YC, that's their choice. As soon as they do that, it becomes a digital equivalent."

O.k., but how do you map IRE to luminance values? 100 IRE is supposed to be white. Adobe mapped 100 IRE to 255. So they already screwed up, 'cause 255 is not always white, depending on what video standard you're using. Professional cameras almost all put white at 235. IRE doesn't become a digital equivalent just because Adobe mapped luminance values to IRE in some arbitrary way.

There's one very good reason to have a digital waveform monitor labeled in IRE, and that's if you are converting between analog and digital and you want to compare the digital levels to the levels on an analog monitor. But that depends on the digital monitor having the configuration options to show the levels correctly for whatever digital standard and whatever analog standard you are using. Vegas actually does that, with a waveform monitor option to interpret digital levels as either 0-255 or 16-235, and the option to set the black point to 0 IRE or 7.5 IRE. The digital monitor is a simulation of what the analog monitor would show.

IRE is a scale that shows you what the underlying signal voltage looks like in a way that relates directly to the luminance values that they encode. (as well as a bunch of other things besides luminance values, like sync and porches) We already have the digital equivalent of that, and it's the Y values themselves, on a scale of 0 to 255. IRE is a well-defined scale for voltages in analog video. Mapping digital Y values to IRE in an arbitrary and non-standard way doesn't help anything.

Okay, to try and illustrate what I'm saying I've taken a locked off shot and used nukes curve tool to map which are the points in the frame that have the max and min luma values (your white/black points). As you can see both points move around a fair bit during the length of a shot. Now its a good example as the overall changes in the min/max value do not change substantially. So, I guess with this shot using just the YC scope would be 'good enough' to use that approach. Other shots, meh, not so much. Shots with really light blacks or dim low whites, definitely not.

However even with this shot you can see that the actual black point (min value) alters a bit between one frame and another. Likewise in the footage there is a situation where the white point (max value) changes too. Now it doesn't seem like much of a change, but remember these figures are in linear float space, in an RGB scale the change is probably going to be a bit more noticeable. Now you can, as you do, set your white point/black point based on your YC scope but don't you really need to identify at what frames in the footage are your luma peaks and lows ? Am I missing something truly fundamental here ? Obviously looking at how the luma min and max jump around doing this with a pixel sampler is definitely going to be dodgy.

edit : Yeah, It got remapped, my bad, but what I'm saying still stands. Basically you don't want to look at the scope,just set your blackpoint to 16 and your white to 235, and its mapped right.

Adobe YC displays 0-255, and my footage falls between 0-255, but I guess it doesn't really matter whether the parts in the zero-16 and 235-255, plus the superwhites, are real or imaginary, since I presumably would use the same workflow for exposure, first setting the blacks to zero (which they are) and then adjusting the 255 level, then the gamma slider in the middle--all this using any effects tool in Color Corrector that does not remap. You can watch the YC SCope in real time as you make these adjustments. As to why they show up on the scope if the camera cannot create then, I can't answer that, but I doubt they are artifacts. Sotf clipping: in my training course the values are pulled down to 255, not 235, and you can again check the results in real time using YC. Now, the person in the video could be mistaken, of course. But I have had good luck with using the system as it is laid out. IRE:Adobe chooses to map IRE to YC, that's their choice. As soon as they do that, it becomes a digital equivalent.

@DrDave The GH2 will not give you any footage outside of 16-235, so it won't have superwhites/superblacks.

Soft clipping, which I think your talking about with your Adobe training courses, works when you have superwhites you want to pull back down into a 16-235 range without pulling everything down. Same goes for pulling superblacks up. You can soft clip to get a specific look too even if you don't have superwhites/superblacks. An over or underexposed shot is just that, you won't have any detail in those areas, just hard clipping.

@_gl Yep, in CS5.5 it does have handy icons that show which of its built-in effects are using YUV and float. Frankly some of the Premiere filters just weird me out, like the gamma correction filter. "The default gamma setting is 10. In the effect’s Settings dialog box, you can adjust the gamma from 1 to 28." WTF doesn't even begin to cover that one.

"But towi, that's my point: if a camera uses full swing, black is zero and white is 255, so there is nothing blacker than black and nothing whiter than white. Those cameras don't have superblacks and superwhites"

ah, okay - I've got it! Yes, that's correct, of course.

@balazer

"The Canon camera, I believe, does not generate superwhites either. The Canon just happens to define its white point at Y=255, compared to Y=235 for the GH2"

I have no Canon camera or Canon files at hand right now... but if I remember correctly Canon video files are fullswing.*

Regarding hybrid cameras I think talking about generating certain levels is not really accurate. I think photo cameras always capture the full range of luminance (0-255, just like in photo mode). The question is how the sensors full range signal is encoded into the video file in camera. The GH2 remapps its MTS files to 16-235 in camera (which I assume is the reason for the GH2's banding issue). Canon cameras do not remap... they pass the full range of luminance into their video files. So if you want so: Canon video files contain super blacks and super whites (= levels above 235 Y' and below 16 Y').

*edit: just found some 5D2 video files... yes, they are fullswing

I will say it again for those who weren't paying attention... IRE is a scale of voltages in analog video. It is completely meaningless for digital video.

How much sense would it make to describe the luminance of some digital video as 0.5 V? Of course it doesn't make any sense.

Some NLEs have waveform monitors labeled in IRE. But these are merely a simulation of what an analog waveform monitor would show. And that simulation is only accurate if a number of import, project and/or monitor settings are all made correctly. And even then, interpretation of IRE values would need to be made with respect to some analog video standard. Different analog video standards map luminance values to voltages differently.

If you want to talk IRE, you must qualify it by saying which software package you are using, which monitor screen you're looking at, and how your import, project, and monitor settings are set.

Otherwise, for digital video, if you want to talk about what is encoded in the video, the only things that make sense to talk about are the unencoded YCbCr or RGB values. And since the GH2 encodes YCbCr values, if you want to talk RGB, you must say how the YCbCr values were converted to RGB. There are lots of different ways of converting.