Hey guys. I have been using these settings for the past 3 weeks to film a large corporate video, and they have been phenomenal for me so far. I'm using the Portrait Setting -3, -2, -2, -4, -2. Highlight -3, Shadow +1. I can't show you any stills or footage yet because the project is a rebranding and has to stay under wraps for a while. For me these settings are perfect. Skins tones are beautiful when you have the right WB. To be fair, every scene was well lit with the Apurture LED lights.
@tonalt I though "Combining Cine-D with totally flat settings" was a reference to CineD AND and -5 in color and contrast AND using Shadow/Highlights (at their maximum values?). It seems clear to me that, after CineD (which is already a flat profile), everything is compressing information even more. Is that correct?
I do find CineD (all other settings at zero or, at most, -2 in contrast and/or color) a good starting point, leaving space for color correction, but then I not really a specialist...
@jopereira, I too find Cine-D with all other settings default (except possibly hue, and yes, contrast +/- 1 or 2 depending on your lens) to be perfectly usable. But I am also not opposed to using Cine-V with slightly more modifications, or Portrait with some heavy-handed shifting. I've used each of these successfully in various productions. So really my experience has been one is not better than the other, but that they are a trade-off on post time versus the extremity of the look you want. If you want a clean, solid, punchy look with minimal post time, the Portrait based stuff will give it, at the expense of being less pushable later, with the other end of the scale being Cine-D which lets you put a really different, strong stylized look on your footage with more post work being necessary. I shoot Cine-V or -D for music videos, quite often, and shoot Portrait more often for commercial-style and narrative work which requires a fast turnaround (such as doc projects, etc.).
Cine D isn't for everyone. But if you are familiar with it and know how to use it, it can produce great results, which I've proven.
http://www.personal-view.com/talks/discussion/comment/185947#Comment_185947
But I am a bit of an anomaly. My approach in post is different. And I did notice that I couldn't get anything resembling good results in standard grading modules. FCP-X grading module by itself was useless. I had to use CGP. Of course with CGP in AE and Davinci I was able to get great results. But trying to do it solely with FCP the image just fell apart. Even in Photoshop (for still grabs) the compressed areas fell apart. But as always, I and CG excel at getting the most out of compressed images. That's what it was designed to do.
@StudioDCCreative Funny you mention lenses, because until now I haven't realize why I put -2 in contrast. My lens is a Panasonic 14-140... What settings do you use in Portrait mode and what do you do in post with those?
@shian You know what you're doing and you get great results. Others are also achieving great results with different settings. I'm learning from you all. I'm a home video enthusiast and all I want is a good, real to life image but also a change to grade it to my taste if necessary (a flexible image). I don't care if I have to capture with settings that require a post work as, most of the time, once you get the look you want, all clips will receive the "same" treatment.
@jopereira, I've come to Portrait -5, -5, -3, -5, -2, Highlights -3, shadows 0, and I shoot on a Sigma 18-35 most often. When I use the Rokinons the contrast goes to -3, and when I shoot with my Dog Schidt Optiks the contrast sits at -1 to 0, depending.
I'd forgotten, but is it "Dog Shidt Optiks" or "Dogg Shidt Optiks" ... and those lenses ... such amazing glass out there ... ;-)
Neil (loving the term ...)
Optical "Iper-16" Anamorfico (Anamorphic) lens mounted on a Nikon 50mm F1.8 Kipon adapter . GH4 4K UHD photo mode 3328x2496 4:3 29.97P Cine V 0-5-5-2+2 High+1 shad+5 mas ped +11 CC MBL basic mojo
@rNeil, it's http://dogschidtoptiks.co.uk so I'll let you draw your own conclusions. ;) And yes, fantastic lenses to start with, even better after the good Mr. Gale has had his mitts on them.
Hey, I'd heard that before and thought is was just a joke phrase. That's some Serious Shidt, man! ;-)
Neil
I'm looking for the best video settings to get good pictures out of a 4k footage. I compared frame grabs using the "4k photo" video mode (which forces mp4) and the .mov video mode. I didn't find big differences BUT the .mov frame grab looks slightly better, so why do we need the 4k photo mode (only for 3:2)? Any suggestion about this? Thank you
It looks like V-Log is on it's way in the next few weeks.
No, Anamorphic in 24p is. V-Log is indefinite, and still not officially acknowledged by Panasonic, although the lid is apparently off the beta users.
I'm shooting a short next weekend and will be using the GH4 on set. I will only receive the camera on the shooting day so I've been trying to research what settings should be best for this task.
The end product will be shown in cinemas here in Iceland, and the subjects will mostly be dancers lit up with spotlights and with backgrounds that fade from midtones to black. So aiming for good skintones and little banding in the background gradient is my main goal.
Here come the settings that I'm planning to use, based on various recommendations I've read on the web, and below are explanations on why I've decided on a certain setting. These settings were selected for maximum amount of information to work with and to eliminate banding. However I haven't seen any thoughts about skin tones and what settings would be best for good rendering of them.
I would greatly appreciate if anyone experienced with the camera would tell me if these settings are indeed what I should be using, or if there is some other settings I should go for.
Luminance: 16-235 I'm not quite sure if I should use 0-255 and just make sure my exposure is perfect on location (I would then have to completely trust the GH4 screen), or if I should use 16-235, which should leave me with some room if I accidentally blow out the highlights or want to pull in some shadows. I assume that shooting 0-255 would give me less banding issues, given that I've got the exposure right.
Master pedestial 0 I've heard that lifting this up (putting in higher numbers) will only lift up the shadows, resulting in information loss. So leaving at 0 is probably best.
Highlights and shadows +1 and +1 I've read allot of contradicting advices on the net. I guess it boils down to what you're looking for. The background gradient that I'm mostly afraid that will create banding issues, will go from black to little bit over the midpoint, if that makes any difference regarding how this setting should be.
Photo style: Natural or Portrait I have the option to use Cinelike D. According to the manual of the camera which describes this like so: "Gives priority to the dynamic range by using a gamma curve designed to create film-like images. Suited for editing.". Now that sounds really promising, but from what I've heard is that this profile can create problems with correct color rendition especially for skin tones, so I've been advised to use Natural or Portrait instead. Any thoughts?
Noise Reduction -5 Planning to do my own noise reduction in post. If the internal noise reduction is good though, then this might be unnecessary. Anyone have experience with that?
Then we come to contrast and saturation, which allot of people recommend setting to -5 and -5 for the flattest profile. I'm really not sure whether that's helpful or not, so any input on that would be great.
I'd be greatly thankful for any advice on how to get this right, since I'm in the uncomfortable situation of having little time to test the camera before shooting starts. So getting this right the first time is my only option :)
kind regards Frímann Kjerúlf
@dreamspy, for your situation I would NOT shoot very flat. Flat reduces the colours used between gradients when recording, which increases banding / noise in the shadows due to macro-blocking in the codec. I'd initially start looking at the portrait settings that are elsewhere in this topic. I've had very good luck with that profile in dark scenes with bright specular highlights - at least as good as any other profile and the noise I got was quite filmic instead of looking digital-blocky - e.g. sensor noise as opposed to compression noise.
I would say this: WATCH OUT for your highlights. I'd record 0-255 (that's what I do, for a variety of reasons which I won't get into yet, but after several camera tests I've found this to give me the confidence I need for the proper highlight and shadow protection) but if you choose 16-235 this procedure will still be of benefit to you. Protect your highlights zealously. I use a Highlight curve of -5 with contrast boosted to +5 in the profile settings - this separates the shadows' gradients more and gives the codec more difference to record rather than mushing it all together, while protecting the highlights and giving them a gentle clean rolloff (highlights have more bits allocated to them so they don't need as much separation between the tones, they need more gentle falloff). Set your zebras to 90% or so, and during your test shoot put the actors at the brightest points with their bare skin or white clothing reflecting maximally to the camera with the lights at the highest setting they'll be during the actual performance, adjust your aperture to this value so that the zebras just click on, and you'll have 10% latitude in case things are a bit brighter onstage.
Try this setting during a tech rehearsal and see if you like it! You might want to shoot a totally zeroed-out CineD in comparison. Grade them side by side afterwards and see which one adjusts to your liking the most.
Went to Paris on vacation. Did some run and gun filming. Nothing fancy. I wish I could of brought my good lenses but they were too heavy. Shot with portrait setting. Mostly 96fps. 14-140mm and pancake 20mm.
What do you GH4 users use for NR? I have used with 4k NR -5 because I think GH4 NR makes blotchy or mushy result. I get more noise with NR -5 but I think that the image is still more natural and better defined.
Usually -3 for noise
BTW, speaking of settings and since I'm just a child as far as know-how is concern, should we expose not to let anything overexpose or is just fine if those overexposed "spots" are not in our main subject and small enough not to have concerns about it (like reflections or some kind of hotspots)?
Balance your needs ... specular highlights, which are bright reflections like off chrome or other polished surfaces, will normally go to pure white. No camera has an unlimited dynamic range ... so you always have to pick where the most needed range of tones are that your cam can handle, and expose for those. So ... you need to expose so highlights you NEED to have detail still have it, and the shadows have enough exposure so you can lift things a bit at need without TOO much noise. You will have a bit, but again, how much noise is (for you our your "clients") too much? You have to define that.
Takes practice and experimentation. And as noted up above, if you find yourself shooting say a v-log camera, you may need to "protect" your shadows by exposing them just a little higher as the cam will lower the contrast and record more into the highlights. If you're shooting most M4/3 cameras in their normal modes, you have to balance the shadows against the highlights pretty much by what's in the scene you're capturing. By knowing your camera, any monitor you're using, against what your recorded videos will bring up in your editing/grading programs.
Neil
Seems like a nice piece of advise. Thanks. I also "find" a tool in Magic Looks (Auto Shoulder) that will help me deal with highlights clipping much better when a raise in overall image brightness is needed.
Hey! Check out my newest video using the GH4.
Everything was shot on 96fps using Sigma 30mm 1.4 with Speed Booster. Color corrected using Filmconvert and Kodak Lutz.
Many say that shooting flat increases noise. I have found that when I shoot video flat (Natural -5 contrast) there is less noise than with Natural 0 contrast. When increasing contrast the sensor grain looks more contrasty too like when adding sharpness. Sharpness is only added local contrast. If I set master pedestrial to +15 the contrast of grains is very low and it is difficult to see noise. Image is still too foggy then.
Of course if I adjust contrast back to normal in post the noise is back there. With my high contrast plasmaTV I dont have to adjust flat image back to normal. It looks very good and natural with flat camera setting. When there is a low contrast scene it looks natural and with a high contrast scene colors and shades pop nicely.
I have to say, to my surprise, that I also feel the CineD results look almost what I want out of the camera. Perhaps a little contrast with Magic Bullet Looks to pop a little more, particular evident in color saturation. Do I have my monitor and home cinema projector with too much contrast? I don't really think so, but I could be wrong (I have contrast and brightness adjusted so I can see the full scale, no more, no less).
It looks like you're new here. If you want to get involved, click one of these buttons!