There are a lot of influencers out there, many of them know what they are talking about, some not, but unfortunately think they do which is made even more unfortunate as the video industry in particular is full of variables and opinions, each with their own raft of technical needs, creative constraints, fans and detractors.
As a stills shooter over many years I am starting to see the cracks as a new video generation, as they are often learning from each other, copying, making the same mistakes over and over (applying LUT’s to stills, using incorrect terminology etc), but it took me a while to be confident enough with video to start recognising these as mistakes.
What profile do you use, Log/HLG/Cine-D, colour depth, do you grade with or without LUT’s and when in the process, do you match cameras, filtering, lighting, camera technique? These all have their space, but what if you just want to get good footage out quickly and consistently with enough room to fix those things that will inevitably bite you?
Considering the end point, which is likely Youtube or a web site, the time it takes to consistently process the mix of elements involved, sometimes just getting the front end right fixes a lot of later steps.
I recently put together a video from several parts made with three brands of camera and graphic elements. What made it easy for me and something I learned from this was the Canon footage was processed either in camera of after, but either way, it was nice and easy to work with, bright, natural and colourful, finished basically. My own footage shot on Flat profile was less beautiful OOC, so I had to match what I was given.
Heresy I know, but what if the nay-sayers like Markuspix are right and the end is the only justification of the means. Maybe good work can be had without employing a LOG profile.
My issues with V-log are many.
It is a pain to expose for, needs special treatment in camera and/or after and at the end of the day, tends to be noisier due to it’s highlight tuning, it usually looks to me like a faux filmic, murky, overly warm-toned facsimile of genuine professional cine-video, but that may be processing.
We see a lot of it, we accept it, it is what it is, but I feel that either we don’t know how to get it to where it needs, have become normalised to it’s look, or by it’s very nature cannot get it there at all. Why is it a camera on Standard mode with no changes looks right to me, when graded LOG footage often looks stylistic and odd?
What you gain in the case of the S5’s is 14 stops of dynamic range, which is better measurably than say the 10-12 stops in other profiles, but not nearly enough to fix the unfixable. There are at least 5 more stops that have to be either avoided, mitigated in some way with lighting or angle or just lived with.
A bit like how sound makes or breaks a video, blown highlights can kill one to, so basically concentrate on what highlight detail you need or want, then let the rest fall where it may. This may mean almost black shadows, but they are way more acceptable than blown highlights, just look at some of the latest movies, which are full of bottomless black.
HLG, Flat and Cine-D are “lite” versions of LOG, giving you a small boost in DR, with a much smaller work load after, but still need processing.
After a lot of research over several years and my own meagre experiences, it seems that Standard or Natural profiles on a Panasonic camera, even an old G9 mk1 can give me all the quality that I or my clients need, but there are other considerations.
The file below was one of the first I shot on Standard profile. I was using modified Natural on a G9.1 (-5 Contrast/sharpness/colour), but was struggling with highlight muddiness I have been seeing from a lot of graded LOG footage. I realised afterwards that the same people recommending this came from the LOG camp.
Out of desperation I switched to Standard, no other changes and lo-and-behold, clean, crisp, neutral and punchy files were produced, good enough to lift straight off as stills (without the running bar). Shadows may be lost, but if you expose for them, highlights are retained, just don’t go looking for hellish scenarios.
The very cool thing about the G9 is the 10bit/422 colour depth out of camera, something other makers reserve for several levels up. This was even available in 4k/60 for ten minute clips.
My videos matched my stills (from RAW), my stills matched my video and my videos, as long as I got white balance roughly right, matched each other. I could turn around a 2 minute video with several elements in one hour from shoot to upload and do my stills processing while it uploaded.
The trick was simply to make sure the histogram (no wave form on the old G9.1 without the paid upgrade), was just within the highlight range and I could often eye-ball that. If that meant deep blacks, then so be it.
I will use everything I can to make sure the footage is otherwise the best quality it can be, so I always use 10 bit/422 colour, something Panasonic pioneered in hybrid cameras and still offers at the lowest price point, my best sound possible and my best technique.
Other tricks.
Panasonic offers i-dynamic range and separate shadow/highlight control that can be combined. These are designed for standard profiles and can be used for stills or video. They can significantly increase DR as long as don’t over stress them to avoid that HDR look.
This is not HDR, it is DR expansion. This is one of the advantages of over three decades of engineers trying to get jpegs right, which by its very nature has spilled over to the video equivalent.
This also looks cleaner and more natural than doing it in post (not to mention easier) and all the other benefits of base profiles are retained like improved noise control, colour and contrast.
Filtering.
A soft focus filter like a Black Mist takes the edge off of strong highlights. The look and stronger is very in at the moment, but it will pass, so keep your footage timeless, but avoiding overdoing it.
Lighting.
If you want nice videos in controlled or semi controlled situations, you will likely use lighting, so many problems are solved. Even a simple reflector makes a huge difference. Often when looking at video comparisons of different video profiles, the subject is put into a horror situation like backlit against a washed out sky.
Here is the thing, and this goes for all such tests, any decent cinematographer will avoid that automatically, because it looks like s%#t, so basically they are measuring what will happen when you stuff up to see if the camera will save you.
Like a lot of things in life, don’t do it badly and the bad bit goes away.
Sound.
Poor sound has killed a lot of decent footage, but superior sound gives it a boost.
*
The advantages of this process are speed and consistency, the only down side is are all the attention needs to be paid at the business end. The reality is, even if I was using LOG, I would be manually setting white balance, exposure and having to allow for the needs of the profile all while attending to light, camera angle, sound etc.
I would also be working in an unfamiliar space, something that flies in the face of 30+ years of stills experience and for what?
If I need a get-out-of jail card, I have likely stuffed up more than the extra 1-2 stops Log can save and still struggle with the end results anyway.
My growth path is now re-aligned mostly to the capture end, where I want to be anyway, less to processing.
The huge advantage of this is the availability of 5 matched cameras*.
The only variation on my mind is the possible use of Flat profile for trickier situations, but I will look at the DR thing first.
I am a stills shooter with a passion for producing decent commercial video on the occasions it is needed. If that changes, I will change as needed.
*2x G9.1, G9.2, S5.1, S5.2