The visible loss of roll-off at only 85% confirms my hunch an experienced cinematographer like Holland would not have clipped the highlights, and secondly supports my perspective that the highlight handling on this camera leaves something to be desired. These types of tests are helpful to establish the practical versus theoretical limits of a camera. As a for instance, albeit unrelated to RED, it's recommended not to expose V-Log L beyond 80 IRE. Perhaps the Komodo has similar limitations. In looking at the side by sides, it's apparent the Komodo and Monstro handle highlights differently. It's risky trickling out off-the-cuff footage from pre-production models or prototypes, because you only get one chance to make a first impression. Canon and Sony nailed their rollouts of the C500 MK II, C300 MK III, and FX9 respectively, releasing dazzling footage only after their cameras were fully baked.
Really wanna love this camera, but nonplussed with the highlight rolloff in the test footage from the two Phils. There are parts of Grossman's white plane and Holland's forehead that have very scant data. Both of those cats are accomplished cinematographers, so I'm thinking it's more likely this cam's highlight handling is less forgiving, than them both clipping their respective highlights. Wonder if they're recoverable with some tweaking in Resolve, or if they're lost for good. The look reminds me more of a Red One, Ursa Mini 4K, E2-S6, Z6/7 or S1H than a modern Red. Maybe it's something that can be fixed with a firmware update? Hopefully it won't continue to be an issue with the production model. Otherwise, for the same amount of dosh, better off picking up an EVA1 + RAW recorder, an UMP G2, or a few BMPCC 6K's for that matter.
Kudos for reviewing gear indie filmmakers can actually afford. These are nifty little tube lights, but the lack of app is a significant drawback. They are missing some key effects (e.g. TV, fire), which would be nice to be able to program via an app. Also, it would be great to be able to save color presets in an app for instant recall. As is, you'll have to make a manual note of the settings in order to reproduce the same colors reliably. Otherwise, I highly recommend these lights. In addition to the hexagonal ends lending well to mounting via a Cardellini clamp, they also come in handy to tilt the unit upward while laying on a flat surface (e.g. table, ground), say for a background splash. If you prefer not to mount the light from either end, the Matthews MQ mount works quite well, too. Also, don't forget these are bi-color, in addition to RGB, and as such can fill a variety of different lighting needs. For instance, they cast a consistent spread across a portable green screen, and can also provide a nice hair light, etc.
"I'm sorry, Dave. I'm afraid I can only give this tune a six for danceability."
The day humans give AI the power to grade art is the day we cease being creative (and surrender to the Borg). Moreover, tech is the whole reason for humankind's decreasing ability to pay attention in the first place.
We got rid of analog, and now spend a lot of time trying to make digital look/sound less digital, i.e. more analog. Is there any reason to believe the boring detector will be any different? Will the result be a new (old) wave of cinema with visionary directors boldly daring... to linger? I can already hear the critics of tomorrow heaping praise on Director X, whose revolutionary style allows us to really lean in and fully soak up the nuance of a character.
It's safe to say we've strayed far afield from the simpler days of celluloid and a sharp blade, but is cinema any better for it?
Instead of dubious bells and whistles meeting with near universal derision, how about getting some fundamentals down first, BMD? Like ASIO driver support and ProRes export for Windows, ProRes RAW import, uniform import controls in media/edit windows, spectral analysis/editing or even the basic ability to save EQ/Dynamics presets in Fairlight, etc.
Casey's advice was generally very instructive. However, there's at least one cringe worthy aspect, and other opportunities for improvement. First, NEVER rely on a single audio source--EVER! You've got two camera angles, so why not two mics and two audio recorders? For the second mic/redundancy, I would recommend also laving up the talent. If you get into post and realize for some reason your single audio source has a catastrophic problem, all of that time was wasted. Speaking of time, rarely do you ever get this much time to hem and haw, and rearrange the entire set. If your talent is important enough to be interviewed, then they have a million other things to do, and will likely be impatient if you're meandering around like these cats were. Your talent then becomes irritated, and may start pressuring you to speed things up, which will not help their performance and will add counter-productive stress on set. The best advice given was depth, and choosing the least bad background. Instead of taking so much time re-arranging the set, I'd recommend shooting a bit more wide open (but not too much), say f2.0 instead of f2.8 as they did. For my taste, the background was still too busy and distracting, which can be mitigated with a somewhat wider aperture, and saves some time trying to re-arrange everything in the background just so. Also, the talent was perspiring, causing her forehead to shine brightly, particularly in the second camera angle. That's something you're going to want to control, either by having a make-up person who can dab the talent's forehead, or bringing down the key a tad. Lastly, there was no discussion of monitoring. I would definitely recommend having monitors, rather than trying to focus off of 3.5 inch screens. This can be as simple as an iPad with a wireless remote app allowing you to monitor your B cam, and a proper monitor (at least 5 inches) to monitor your A cam.
The BBC debuted UHD on iPlayer for documentary work last month (Planet Earth II). While Full HD is still aired, and will be for some time, their technical guidelines for delivery spell out UHD/4K capture specifications. Similar to how feature films still mastered in 2K are typically shot at higher resolutions today, it's a good idea for the poster to invest in new rather than old technology, because: 1) his work will be marketable further into the future; and 2) it will look better today downscaled to FHD.
In addition to the trend toward higher resolution capture and delivery, BBC's existing delivery guidelines exclude native capture on the Sony A7s and C100 mk i, since both are 8-bit 4:2:0. He would have to purchase an external recorder to be within their 4:2:2 spec. That adds bulk/weight for run and gun, and the additional expense could be put to better use for lenses, etc. Both cameras are still more expensive than the announced cost of the GH5 - for less in terms of broadcast specs.
BBC delivery guidelines also reference EBU standards in terms of camera suitability for capture, and for what it's worth the latter have rated Sony alphas as poor/unacceptable for broadcast. Two other gripes that I ran into in the field with the Sony alphas, that I forgot to mention above, are rolling shutter and overheating.
I've shot two feature docs and two short docs on Sony, and have now jumped ship, unloading my Sony gear. As indicated previously, I really appreciate Canon's color science, but have decided to go with Panasonic instead, as I find that Canon lags way behind in terms of features.