Hello all - We've just used the GH4 for an in-water segment on a low-budget feature (couple of stills here - one from the GH4, one from the BMPCC). It's ridiculously difficult to gauge exposure and best settings when you're in the water, so we did it a bit by eye and a bit by guesswork. Looking forward to seeing how it compares to the BMPCC though when we get to the grade. Any tips on best settings for future in-water stuff? You tend to get massive variations in exposure, as half the lens is under the surface and the other half is often seeing a huge amount of sky! The Facebook page for the film is www.facebook.com/dsrmovie - which will have more GH4 in-water pictures at some point, and a trailer that will use that footage quite heavily.

Cheers,

Alex.

@zachasurp thank you that is awesome news, 99% of the time I downscale the 4k anyway. One thing regarding having no monitor - isn't there an HDMI out for loop through to a monitor on the Ninja star?

As far as I know the GH4 can´t record internally 4k and have 10bit output.

@rrroger I am hoping my theory is correct but wondering if anyone has used the unit and can confirm?

I am interested in getting the ninja star and watched the video on how to record 10-bit out of it I just have one question I can't seem to find: when you record the 10-bit and select 4k to 1080p, is it recording 4k, down converting it and sending it to the recorder? Is it essentially the same as recording 4k internally and exporting it from the editor at 1080p, but in 10-bit 422? I am just wondering if the detail captured when recording in 4k and editing in 1080 would be present in a 10-bit 1080p file from the atomos. Thanks!

So I went out today to do some lens test and grabbed my tripod but unfortunately forgot the quick release mount. So I had to stabilize these shots in post. I love to take my cameras and film trees at some point because years ago I switch from Canon to the GH2 because I never liked the mushy trees that Canon would produce in my footage. So I know if a camera can make distant leaves on trees good, then I know It will be good to go. When I got home I wanted to see how some new camera settings and some new lenses will look with my filmic color grading treatment. I will post another video soon about my findings, right now I'm still testing. What do you guys think so far with this video. Do you like the look. Panasonic GH4 Panasonic Lumix 20mm 1.7 lens

@shian quote: "I have no idea what the camera is actually doing to the image...I'll admit that right now." Strange comment when you are taking such a solid position on flat profiles.

@c3hammer makes a valid point about fog obscuring detail. I think the issue here is that there are two factors to consider: 1. Fog lifts shadow areas into grey, making detail in shadows more visible to the eye. 2. Fog masks actual scene detail.

I think we should clarify what we mean in plain language to avoid misunderstandings like this. Non technical emotive language does not communicate very well.

@shian has some very important things to say: quote "I never base any conclusion I reach off of a pre-set theory. I tend to be more scientific and base it off observable and repeatable results."

Finally, most people seem to have trouble understanding when to use Master Pedestal settings. My opinion is that it is useful for this very situation. If fog or smoke is lifting black areas into grey, lowering the pedestal will bring the blacks back down and increase headroom for highlights. The gamma response or curve remains untouched.

I think the settings need to be adjusted to accommodate the requirements of the scene. Too many people seem to want one 'do-it-all' setting that will work with all scene conditions. Remember that there was never just one film stock or process that worked for everything!

@shian

The flatter you make your image the less stress you put on the codec. The more color info, contrast, and vibrance you make the camera compress, the harder the codec has to work to capture the info.

Depends on what you mean as putting "stress" on the codec. With the GH2, what I considered stressful images were shooting conditions that undermined the reliability of the video codec, provoking it to either crash or resort to Fallback Mode. While dynamic range and vibrance were definitely stress factors, sharpness, fine detail, and overexposure were even more significant. However, I think it's safe to say that none of the profiles called "flat" would cause reliability issues.

If you use "stress" to refer to bitrate-intensive coding complexity, the effect of a flat profile depends on the balance between highlight and shadow details. An H.264 codec uses far more bits to encode highlights than shadows. With a sharp, well-exposed, high-contrast image, most of the bitrate is devoted to encoding the high-contrast details of bright sharp objects. A flatter profile will likely reduce the dynamic range of these details and require fewer bits to encode. This, however, may free up more bitrate for encoding subtler shadow details. If your flat profile also boosts shadow levels to the point where low-level noise becomes visible, it will take more bits to encode their random motions.

Whether this produces better image quality in post depends on your aesthetic concerns. If your priority is midrange skin tones, you'll want to devote most of the H.264 codec's bitrate to the mids, and sacrifice as much shadow and highlight detail as you can tolerate. If you want to preserve as much shadow detail as possible, you'll want to flatten the highlights in order to boost shadow levels, and grade the shadows down in post. If you want to preserve highlight detail, you'll expose to the right to eliminate all highlight blooming, and disregard the shadow details that are clipped into black. In all cases, matching an appropriate profile to the type of image you want to capture will produce the best results.

@shian Wow !!!

Shooting through fog/smoke doesn't lift shadow detail in anyway. It masks detail and thus your observed lower bit rate.

Your statement "Whenever fog or smoke were introduced to an image thereby bringing up the shadow detail and lowering contrast in a very similar way to what the camera now does internally, the data rate dropped... significantly and consistently." is utter nonsense. It may or may not increase shadow exposure, depending on your lighting, but fog and smoke surely obscure detail. You're filming fog. Fog doesn't have much detail. That's why shooting in flat light is so boring and dangerous depending on how fast you're going.

Your next statement that follows "Now this tells me either Digital Compression Theory doesn't work the way you think it does, or the GH2 and GH4's internal computers have the algorithm all backwards..." is even more nonsensical.

And based on that nonsense, you move on to this "Something about lowering contrast, flatter images.... lower data rates, not stressing codecs... Pull your head out of your ass and find something better to do with your time than argue about whether or not it actually increases the fucking detail."

I got your point perfectly and I think you're completely mistaken. It doesn't increase the detail and you're wasting bit rate on something you and what you're selling are throwing out anyway. That there seems to be the waste of time here.

@shian. I've gone back and forth on which profiles I like. I'm with you on Cine D being the best option. You may have already mentioned in earlier posts, but what do you think of Master Pedestal? I just moved mine back from +10 to 0 since I'm always bringing the mids back down in post anyway. I notice when I'm matching GH4 footage to C300 or BMCC Log footage that the mids are always higher on the GH4.

@shian Thanks for sharing your view on the flat image data rate concept. Quick question if you don't mind. Where do you draw the line on how flat you're pushing your image? For the music video I had mentioned earlier in this thread, I found that at a -5 contrast in Cine-D, moving the highlight/shadow curve at all introduced too much noise. You seem to be getting away with an extra +2/-2, though. So what, in your experience, caused you to push it that far (or even to stop pushing it there)? I'm obviously just speaking about curves, as raising master pedestal only raises the lower spectrum of the IRE and risks highlight clipping (more people need to realize that...) Thanks, Shian.

^^^^ Does it raise the shadows or not, numbnuts? As a cinematographer, do I give a fuck if it actually increases the detail? Does it in any fucking way affect the final result? Did you miss the point entirely? What was my conclusion? Remind me. Something about lowering contrast, flatter images.... lower data rates, not stressing codecs... Pull your head out of your ass and find something better to do with your time than argue about whether or not it actually increases the fucking detail.

@shian you're the man :-)