Black-water video shooting techniques?

[humu9679]

@bvanant I'm impressed by Nanette's BW video. Great stuff.

 

I cannot fathom how you handle focusing with the LX10 @bghazzal I was on the #SmilingSeahorse 18-25 Feb 2024 and we did one BW dive. My Sony A7c and 50 macro struggled to keep up with the critters -- shooting stills -- and my first BW dive was a big fail. Re lighting, it seems the lightbox technique over lights the whole scene, and I wonder if just bilateral side lighting would do the trick? Probably best not to light up everything in the water.

 

@christophe chellapermal It was a very nice trip on the Smiling Seahorse (SS). Franck did not come on this trip, and I believe the SS was the first dive operator to offer BW dives in the Andaman Sea.
 

[bghazzal]

10 hours ago, humu9679 said:

I cannot fathom how you handle focusing with the LX10 @bghazzal I was on the #SmilingSeahorse 18-25 Feb 2024 and we did one BW dive. My Sony A7c and 50 macro struggled to keep up with the critters -- shooting stills -- and my first BW dive was a big fail. Re lighting, it seems the lightbox technique over lights the whole scene, and I wonder if just bilateral side lighting would do the trick? Probably best not to light up everything in the water.

I was shooting in MF and following with focus peaking - I switched on AF to see if it helped but given the chaos it was easier to track with peaking (which I don't have with AF, unless focus is locked)

But once light position is set, it should be a question of keeping them in the light, like snooting a moving critter. But these are super fast indeed, and small...
 

My main worry is actually that I might have bumped into the LX10's lens limitations when it comes to size/movement for video...
Zooming-in, which I must do to get the proper magnification, works fine for filming tiny critters when they're stationary or slow, but when they're flying around like this, even if you keep them in focus long-enough, the speed of the background blur makes for an uncomfortable, sea-sickness inducing watch...

It's something I noticed shooting during the day as well - I have some stable, in-focus shots of small fast moving critters that are not nice to watch because they're zoomed in, and you have a background like a fast-flowing river, which is not a pleasant watch...

My hope for black-water is to try to isolate them more, so that the background is black, which might be a way to get around this issue.
 

However, a major issue remains: the movement patterns of such tiny planktonic critters (which I believe are actually feeding in the light) is very erratic.
During the day, the fast moving critters where moving in a relatively straight line, but these guys are way more random, which means more camera movement to follow.

Unfortunately, even if the shot is stable and in focus, camera movement is amplified when zoomed in / long focal length, and not really nice.

 

Solution would be not to zoom and use actual macro lens, but I can't on a compact... Tricky, tricky... Especially if i drop resolution to 1080p to get 60fps, meaning I can't plan on cropping in post...

 

Filming fast movement zoomed-in is really my main concern at the moment, not sure if I can find some way to make it work for fast-moving tiny critters (other than focusing on slower / bigger ones, if they exists!)

The subject-background-movement problem is one of those specific video-shooting issues, really marked for longer focal lengths / telephoto...

 

You're right for the light position - I need to find a compromise between a workable focusing area to track the critters and isolating them more.

Plan is to do some dry testing to see what the options are!

 

cheers

b
 

Edited July 15 by bghazzal

[Giancarlo M.]

19 hours ago, humu9679 said:

@bvanant I'm impressed by Nanette's BW video. Great stuff.

 

I was on the #SmilingSeahorse 18-25 Feb 2024 and we did one BW dive. My Sony A7c and 50 macro struggled to keep up with the critters -- shooting stills -- and my first BW dive was a big fail.

 

Unfortunately it is not a good lens, the AF is very slow and makes you miss many opportunities.
I experienced something similar with my R5 and the Sigma 70 macro Art, beautiful lens but terrifying AF, very difficult if you are not used to shooting consistently.
You should try, the Nikon 60 macro with the Monster Adapter, to some it works, to others it seems not to work well. 

[Giancarlo M.]

Unfortunately for Blackwater, you have to accumulate the necessary experience, the more you do, the more you will be able to bring home a good result.
There are many factors to consider, the moon, the tides, possible currents, surface wind, bathymetry. It is a dive that requires the diver to be a good diver and those who organise such dives to have the appropriate experience and the necessary equipment.
Turn to professionals and not improvisers.
I have read before about being tied to the boat for this kind of diving... it has been done in the past, but it is very dangerous, I strongly advise against doing such a thing.

[Davide DB]

13 minutes ago, Giancarlo M. said:

 

Unfortunately it is not a good lens, the AF is very slow and makes you miss many opportunities.
I experienced something similar with my R5 and the Sigma 70 macro Art, beautiful lens but terrifying AF, very difficult if you are not used to shooting consistently.
You should try, the Nikon 60 macro with the Monster Adapter, to some it works, to others it seems not to work well. 

 

Here we are speaking of video od course but what's the best Canon lens for AF in this kind of dives?

 

Seeing the Aquatilis project her in the Med, most of the critters are huge. Something I would shot with the 12-35mm (24-70 FF).

[humu9679]

2 hours ago, Giancarlo M. said:

 

Unfortunately it is not a good lens, the AF is very slow and makes you miss many opportunities.
I experienced something similar with my R5 and the Sigma 70 macro Art, beautiful lens but terrifying AF, very difficult if you are not used to shooting consistently.
You should try, the Nikon 60 macro with the Monster Adapter, to some it works, to others it seems not to work well. 

For BW I agree. The Sony 50 macro on my A7c was slow to lock on, but good for tracking. 

[bghazzal]

Regarding light positioning for black-water / bonfire type video, I've had some interesting info from Alexander Semenov, who shoots the amazing footage in the Aquatilis project

 

Critters shot this way are mostly bigger than typical BW subjects, and setup primarily using is a triangular edge-lighting set up with wide, powerful lights.
This is how he describes it, and helps shed a little light (pun) on the photos Davide posted earlier in the thread, which I'll repost below:

"Two huge lights a bit behind the housing - you need to have triangular "dark zone" right in front of the port glass with no light at all and then set up the lights in a way that crossing of light beams will be just in a few cm closer to camera than focal plane. It's a bit tricky, but works like magic. And more light is always better."

 

 

Edited July 24 by bghazzal

[bghazzal]

One question that really intrigues me, in current circumstances, is the relationship between the constant light output of video lights used for filming and video lights as lures.
 

For photography it seems pretty straightforward: a narrow beam focus light to spot, staying around the lures, and milliseconds of strobe lighting to shoot the critters.

 

But for video, shooting means turning on the constant output video lights...

Question is, in blackwater/bonfire type setups, how does this work when working with lure lights ? Won't the camera video lights become lure themselves?
Is the secret to always have more power on the lure lights than on the videast's rig?

This interrogation is directly linked to my current situation, where I'm in a very interesting place to try things of the bonfire kind, great opportunity to work on things, but I'm also alone in the endeavour, with no access to a set of lure lights.

I'm basically trying to shoot some of the larval planktonic critters which show up on night dives, to work on things, and there are plenty of subjects arounds (most common at the moment are larval mantis shrimps for instance, present even on a bad day).

So I tried using one of my 3 lights as a lure, but it's only 4000 lumen - so when I turn on the other two, the swarm of planktonic critters instantly reforms around my camera's lights, and it's very difficult to isolate subjects as it's heptic with life coming in super close.

Really not sure how to approach this and make it work. It's been easier to shoot planktonic critters in blue water during the day,because the sunlight means I don't have to deal with attraction to the lights...
I just shoot them as I would to create black-background shots, which works fine (especially now that I have a set of 2x 8000lm lights) but daytime plankton is much less consistent.

I've asked around and no one has powerful lights unfortunately (too many photographers 😁), so not sure what I could do.
It feels a little sad to give up on the idea, but maybe local waters are just too rich to roam in the water column at night....
i guess I could use one of the 8000lm lights as a lure (leaving me with one 8000lm and one 4000lm, but wondering if this would make that much of a difference since the rig's lights would still overpower the lure...

any thoughts?

 

cheers

ben
 

Edited September 14 by bghazzal

[Davide DB]

Given that you are working with 2 lights very near to the subject (basically a lightbox) you could work with a very low power.

 

Another solution would be having an on camera small light to spot critters and turning on the main lights only when framing focusing and filming.

 

[bghazzal]

I tried the small light to spot last time - I had the SUPE MS-10 snoot video light centrally mounted, and used the Bacscatter MW4300 as a lure.
But the beam of the MS-10 turned out to be too thin for these purposes.

After 20/30 minutes waiting for the swarm to develop around the lure only to have it reform on the lights, I gave up and mounted the 4000 lumen lure centrally on the kit instead, which is nice.
Macro mode beam is narrow enough, but not too narrow to be unuseable, and also it seems to help with the focusing / lighting from what I could see.

As for using the two lights on low, this was indeed the plan - I thought I would have them on low all the time - but based on what I could try in situation, what worked best to avoid backscatter (or rather critter saturation, as they're actually alive) was actually having the lights at 6000 lumen and using the side of the light cone instead of have them more central which illuminates a too large zone.




So I moved from the lightbox to a more open config, but this also requires more power to work.

Maybe this need more testing - if I remember the previous posts, some people (Nanette?) got nice results with 3000 lumen lights, but I assume this was in a true blackwater setup with a ton of lure lights...

After giving up on the single lure light, I tried just using the MW4300's beam to spot/attract critters, but the problem remained, everytime I had something and turned on the two lights, a swarm formed around them, drowning out the suject I was trying to isolate..


 

Edited September 14 by bghazzal

[bghazzal]

As a follow up to this thread on shooting blackwater video, I’ve been experimenting with “bonfire” / light trap style setups here in east-Bali, and can share a little clip compiling the first results of these tests.

My "bonfire" (the name is really over-the-top in this case) or "light-trap" lure setup used is very basic and rudimentary, as I’m doing this on my own and have very little lure lights to work with.

And yet it’s still been quite interesting, and brought up questions on lighting power, positioning, spotting and focusing, and importantly the relationship between video lighting’s constant light output and the lure lights, which I'll develop below.
 

Overall, the most important thing is that despite the lack of lure light power, it still works, subjects do show up, and my current interrogations are now very video-specific, and don’t necessarily apply with the sub-second strobe lighting used in still shooting, but do overlap in strobe / light positioning...
 

***
 

When I say my bonfire-dive setup is basic, it really is. It's not really a bonfire, more along the lines of a small campfire barely good enough for heating up a can of tomato soup... 😅
 

So not really the stuff of dreams, but then there's location, location, location, right?

Facing the deep waters of the Lombok Strait (which is also one of Indonesia’s major current throughflows), what I do is simply swim up to a mooring with with roughly 25m depth, where the plateau stops and the deep slope begins...

My bonfire crime scene, on the shores of the Lombok Strait

 

Sloping off into the strait, kumbaya, kumbaya...

Setting of from the beach into the darkness at around 7:30 pm, I swim out to the mooring (which is on a sloping bottom), and then hang my light rig, facing downwards, at 8/10m depth, and wait for the swarm to form and food-chain action to build up, which takes about 15 minutes or so.

Let there be (a little) light - my first lure light rig: a Backscatter MW4300 + 2x Archon dive lights...

 

Now I've broken out the spare dive lights, and I’m basically working with 3x to 4x 1000lm standard dive lights each with a 16° beam, and the Backscatter MW4300, which is a 4000 lumen light.

On that note, I still haven’t concluded if the MW4300’s wide or spot mode is more practical.
- Wide mode (max 4300 lumen, but I usually put it on second-highest power) has an 85° light beam, which creates a larger cone of light.
- Macro mode (1400 lumen) has a 14° light beam

The MW4300 in 85° wide-mode with a swarm forming



Will I have very little lure lights, this is actually enough to attract stuff, and it's been worth the hassle of going off into the darkness, bringing up some nice subjects, juveniles and larval forms.
Nothing fancy like a blanket octopus, but still enough to keep me occupied....
 

So this limited light power is enough to attract different critters here in east Bali, with regulars and also some more suprising encounters, depending on conditions.

Some visitors on the last 3 test dives

One thing is very clear through:  the closest critters are to the light source, the more active (and the faster) they are.
As with an outdoor light on a warm summer’s day, the light sources gets flooded with critters, generating  lighting fast feeding action and general frenzy.
Mantis shrimp larvae - which are everywhere in these parts -  go from slow cruising to backflipping frenzy depending on proximity to the light source, for instance.

Action building up right under the lights

The swarming has begun, time to break out the marshmallows...

 

One of the things I learned is that for video (and following subjects not lost in the mass of the swarm), it’s best to go a little deeper and on the edges of the light cone

It's where you'll find slower subjects (sometimes more interesting than closer to the light), which are not lost in the swarming mass, so easier to follow.
 

So for people looking to organise super-basic bonfire / light-trap dives with limited lure lights, I would strongly recommend finding a way to hand the lights over depth rather than putting them on the sand facing up.

Reason is that it’s just easier to hover in open water, and to go deeper rather than shallow-up to be on the edges of the cone of light.
Also, there are less chances of stirring up sand when swimming around to film - given the small critter’s feeding behaviour, the dives are quite active compared to standard macro shooting...
 

I also found myself directly under the cone of light a few times, so really, for limited light setups, hanging the lights is the way to go.

Lastly, a weird thing is seeing critters letting themselve drop in the water column, which you'll see in the clip, as drop with the critter - this vertical movement, as opposed to horizontal, is quite common in this context for some reason - not sure if they do the opposite on a sand-based bonfire setup...
 

For my "bonfire dives" I use a boat mooring which is anchored to the bottom at 25m, but given how little weight and traction the lights have, a buoy or smb floating on the surface with a few weights to anchor it to the sand would be just fine (though using an existing mooring is more environment friendly, and of course you couldn’t do this near a reef)

 

So all good... 
If conditions are right, you can have fun in the darkness without a massive amount of light (please note that I am doing this alone, so with a group of divers, well, the equation is probably different).

More light is surely better of course, but it's possible - to some extent - to wing it with a very basic setup, without a milion lumen of light...

 

 

*******

Here's a little clip based on these first three bonfire-dive style experiments:

 

 

 

I'll follow-up with some more video-centric observations and interrogations...

Edited October 5 by bghazzal

[bghazzal]

Now for gritty gritty video-centric questioning…

 

If I break it down, here are the main issues and interrogations these bonfire-style three tests have brought up for shooting video:

Video lights becoming lures...

 

The first major issue I’m having is dealing with my 2x video lights becoming lures themselves…

 

At first I thought this was mostly an issue of not having enough lure lights (so the swarm would reform next to the strongest light source, ie my 2x 8000lm video lights), but while this is certainly true close to the light source, this also happens on the outskirts of the lure light cone, where the more interesting / manageable subjects seem to be hanging out... 😎

Anilao blackwater drifting downline as invented by Mike Bartick vs. homemade bonfire rig... 😅

Turn on the camera's video lights, and a swarm forms, even at a good distance from the lure lights...
 

Meaning that just adding more lure lights probably won’t solve the issue - as soon as you turn on the camera video lights, the lights become lure themselves, so we have to work with and around this.
 

This is both a video specific issue and, I’m guessing, also probably something more specific to shallow bonfire dives.

I might be mistaken, but I imagine that open ocean blackwater diving probably doesn’t, in most cases, have as much swarming, at least not the frenzy you get to the shore, which would help a lot for shooting video, to isolate subjects.

 

Critters swarming onto the camera’s video lights means that subjects become more active as they enter a feeding frenzy (or whatever it is they do...), and also that, plus the mass makes it very difficult to isolate a subject efficiently.

 

Another issue is that the lights being on the camera, they come way too close to focus efficiently, even with my Inon +6 diopter.

So this video-lights-becoming-lures has been a major issue with my bonfire-style setup...

 

One solution would be to use the lights on a lower setting.
And this is precisely where I hit another hurdle, video light power vs. video light positioning...


Video light positioning and video light power

 

These dives take place in very rich waters, so not only critters but phytoplankton and particules, the works.
Like for still photography, backscatter is a major issue, as is lighting translucid critter correctly and not becoming a lure...
 

Looking at strobe positioning and also the very rare examples I have of people shooting BW video, I’ve narrowed it down to 2 positions which work best for dealing with backscatter.

 

- using the edges of the video light cones, after pulling the lights behind the port, lights facing outwards

 

outward facing video lights, with the frontally mounted spot beam, showing the crossbeam triangle

This is a position I need to experiment with more, as I was a little shy and afraid of attracting too many critters.
 

- ear-muff type lighting, where you create a sort of light box in front of the port, more or less angled inwards

slightly forward facing earmuff style light positioning, with shades on the lights

This second position is actually what I've been using the most.

However - and this is probably both a constant out video-light + bonfire swarming specific issue - I'm still very much figuring out what works best in this case...
 

As a caveat, having two powerful video lights is still new for me - before that I was working with a main centraly mounted MW4300 complemented by two weak video-lights on the side for macro and supermacro (and ambient light + filter for wide angle)
 

I recently got two Hydra 8000 lm lights, for this type of shooting, blue water and medium shots, so it's all still very fresh.

And unfortunately, the task-heavy environment like the bonfire one isn't the easiest to figure out what works, and land tests can only help as much...

I based myself on whatever info I could find online for strobes, eg:

capture from Nicolai Deutsch's Insider Divers YT tutorial series, showing a classic wide positioning



capture from Simon Lorenz's Insider Divers Youtube tutorial series, showing "ear muff" positioning



And also refering to the precious advice given by a couple of people actually shooting black-water type video (rare!)

capture from Alexander Semenov's Aquatilis / Ponza video shoot, "ear muff" style...

 

Back to the actual in-water results, lights angled out to use the edges of the of the video light cones works well for backscatter, and gives me more of a visual reference for subject tracking.

However, I then need enough power for the light to reach where my main focus area will be.

I do have sufficient light power with the Kraken pair, it works fine at 6000 lm for small critters BUT then the swarming happens fast, and is quite crazy.
I have less than 30 seconds to shoot before hell breaks loose, and my subject usually starts going wild before then, making following / focusing very tricky.

More than backscatter, it’s more abundance of subjects partying in my lit focusing area which is an issue...

 

Other issue is that even if my central subject is well lit, I’ll be lighting up others in the frame, really no way around it in this context – and one thing I’ve noticed - and which you can see on quite a few sequences in the clip I’ve posted above - is that critters have a tendency to let themselves drop in the water column....
I think this is related to feeding  by basically they come up to the light and drop, almost not moving horizontally.
This is ok to shoot, but you can see the downwards movement of the camera + critter dropping together as the other critters move up in the frame, motion blur which highlights anything lit up in the frame...

post-larval flounder - East Bali bonfire tests

 

With a more closed-in, inward-facing ear-muff style position (eventually adding shades to the lights to avoid lens flare, though this might actually be counterproductive in the case with my 100° beams, the jury is out...), I can use a much lower power setting, which means less swarming, and is better for staying the subject.
 

But for this, lens flare is risky, as I can’t really review shots as a still photographer would.

And even locking myself in a dark room with bits of translucid plastic bag hanging from fishing line to test things out (yes, I'm that desperate), it's very difficult to recreate in-water conditions, which means I'm trying to figure it out on the fly...

 

subject spotting/tracking before turning on the main lights
 

And other than that, there is another issue - subject tracking - with the lights not facing forward I’m basically blind until I get the subject inside the light box / focus area

 This is very very frustrating, as can I spot a subject above the camera, but then have a hard time getting it in the light box to follow it.

One way to try to get around this is to use a focus beam, like BW photographers do - a narrow forward facing beam, allowing me to narrow in on the critter before turning on the lights.
 

I've been playing with this, but this has proved tricky as well.
 

I’ve tried with a snoot video light (before it started acting up again, see here) with a powerful but very narrow beam (5°) - but while this works well, it’s also too narrow to spot with in camera efficiently.
 

I then tried with one of the Archon dive lights with a 16° beam, but it was a little too wide (and it is a dive light, even if I add a homemade snoot, not being a video light it will have a hotspot, and probably won't blend too well with the two lights...)

 

Using the MW4300’s 14° narrow beam sounds like a good middle-ground alternative, but then I lose one of my main lure lights... hmmm..

Nonetheless, this is what I’ll probably trying in the future, and the ability to spot / track critters on the outskirts seems more important than the power of the lure lights in this kind of setup

free-swimming tozeuma shrimp in the fray - East Bali bonfire tests

 

Depth of field / focusing area

 

Shooting small fast moving critters is tricky, focusing of course but also dealing with

subject abundance in the focusing area...
 

I shoot in manual and use focus peaking to see what’s going on, because depth of field is just to narrow for my autofocus to handle (which would be confused and probably won’t be reactive enough anyway). Tests on land showed that the auto-AF was unable to adjust properly.
 

Most of the shots are zoomed to my compact’s 50mm lens equivalent, which is a good compromise (more zoomed out is too shaky, hard to spot, and wider is too small)
 

I’ve been shooting mostly with a +6 diopter, as I found the extra magnification helpful in this situation, and also helps critters moving closer to the lens.
But zoomed out to 50mm, I don’t have much leeway and the diopter helps.

Or does it? basically the diopter's small magnification factor means I can zoom in less (which is less shaky, and potentially helps with DOF, but then we need to take into account the +6's effects on DOF as well), and focus closer (but very close to the lens, light position usually becomes a problem, repositioning, avoiding lens flare etc...)

This is also still unclear - recent on-land tests I did showed that at f11, I can focus almost almost the port glass even at 50mm, so I might be making it harder for myself with the diopter (more risks of flares etc), magnification is a double-edged sword indeed...

Of course it would be easier if I had a standard 60mm macro lens like a normal person, but where's the fun in that, right? 😉

down and down we go, dropping together in the water column... East Bali bonfire tests

 

For focusing, overall, it really depends on the critter’s type of movements.
Feeding shrimps are really erratic, but worms and fish are easier to follow, using peaking as a ref.

 

If I had light positioning figured out, it also would help give a better reference of the focus area, and generally make life easier as well.

At the moment it feels a bit like being blind at times, with the added difficulty of switching to the camera view and find the critter you’d spotted, and when to turn on the lights which is something I would need to work out.
Using a set framing reference, as it has been suggested, would not work in this case as I wouldn't be able to see it above or in camera, and more importantly the depth of field is just too shallow to use efficiently.

 

Exposure

400 ISO works well – but for shutter speed, shooting in 30fps (image quality at 1080p 60fps is just to low) I’ve resorted back to the 180° rule of shutter and shoot in 1/60 (*I usually bend the rule shoot at a higher shutter speed for macro/super macro, which gives me more flexibility and doesn’t actually result in a « saving private Ryan » look anyway).

basic settings for bonfire tests

 

 However, given the type of movement I’m shooting I’ve been wondering if cinematic motion blur isn’t a bit too much – maybe 1/80 would be a good compromise for a little more detachment, if exposure can handle it, that is.

Most of the shots were in f11 (which is the max on my compact) to f8, to try and maximise depth of field...

Which leads us back to light power and positioning, loop closed, over and out...😅

***
 

As a temporary conclusion, I'd greatly appreciate tips on light positioning and subject tracking, as this has been quite difficult to experiment with in-situation.

The issue is primarily how to minimise light power (to attract less critters) while still not being completely blind, while lighting up critters properly...

I'm really starting to think a frontally mounted video light would help with tracking (but it would need to stay on as I film though), but just not sure where to start...

cheers

ben

 

 

 

 

Edited October 6 by bghazzal

[bghazzal]

If I'm to summarize the information collected from BW video shooters, recommended strobe positions are mostly variations of the following:

 

 

earmuff-style positioning, for subjects at a certain distance from the lens/port (unlit background)

earmuff-style positioning, for subjects closer to the lens/port (unlit area between port and subject)

These two basic positions can also be angled in (for better rendition of transparency for instance) or out, like this for instance:

earmuffs, slightly angled out

 

 

For larger subjects / wider scenes, it's a variation of the typical wide-angle video light setup, but mitigating backscatter by either angling out or pulling the lights behind the port:

lights pulled in behind the port

 

lights more forward but angled out:

Other consideration would be video light verticality - lights level with the port or above (and angled down), for instance...

 

cheers

ben

Edited October 7 by bghazzal

[bghazzal]

As a follow-up, I've been experimenting further with blackwater type video on bonfire dives.

Lighting

Light positioning is still a challenge, as it's difficult to get visual feedback on the spot - results can be better or worse than expected 😅 still figuring it out, as always...

What is clear is that video-light positions shown above all work, but while having the lights facing forward as in the last two diagrams give cleaner results (less backscatter), in my short experience they also have two major downside for video

- light power needs to be set higher, which excites the critters, leading them to a plaknton feeding frenzy (moving faster), and attracts the swarm

- forward facing constant lights will definitely be a nuisance to other divers / image-makers who happen to be facing in the same direction

 

Ear-muff style side-lighting is not as clean in terms of backscatter, but has the advantage of allowing us to lower light intensity, which seems to better for the critters and fellow divers.
It also works better with translucent critters.

Focusing

The other main challenge for video is, of course focusing.

I don't have access to a fast autofocus on my camera, but frankly, I have my doubts as to how efficient a fast autofocus would really work *for video* in this case (but maybe I'm wrong and tracking would be seamless, just difficult to image this speed while recording)

Issue, of course, is that critters are small and are feeding, so move fast and often erratically, and while managing to lock-in and track a critter for a split-second strobe firing, for a minimum of 10 seconds of video, it's another story (which is why so much BW type video is slowed down 50% or more... which is fine and even great when it shows things impossible to make out otherwise, but also loses natural movement..).

And yet it is possible - difficult, frustrating, but possible - it really depends on the critter's behaviour and size of course.
One way to approach this is through manual focus, with peaking.
Lock the focus within the lit area, spot the critter and move in. Peaking lets you know how the focus is going, adjust with camera movements. 
The problem is the critter moving along the focus plane - which they do - coming closer to the lens then moving away, in and out...

This is where I'm at at the moment, trying to figure out what would work best in terms of magnification, working distance, and depth of field.

Life being what it is, I can't change lenses on my small-sensor compact, so I'm stuck with the lens I have, which in 4K starts at 36mm
After initial tests, what seemed the most flexible was working with the lens zoomed to a 50mm equivalent.

In terms of aperture, f/8 to f/11 - which is the smallest I can go - works fine for me - based on normal supermacro shots (this for instance, all shot at f/11) diffraction and resolution loss doesn't seem to be as much of an issue as it would be on bigger rigs, and this is video, so pixel peeping is less of an issue.

Zoomed out to 50mm or so, on my camera adding a +6 diopter (the Inon UCL-165) gives that extra magnification and drops working distance to the glass, which is appreciable (see below)
However, yesterday I also tested the AOI UCL-09 +12.5, which is tricky of course, but very nice and detailled when finally managing to nail the @!$#%@#%@ focus.

These are rough screen captures (screenshots from as my computer's preview) of clips shot with the AOI +12.5 at roughly 50mm (the lens vignettes under 46mm), f/11, giving an idea of the range I'm getting with this diopter

 

Other than getting a new camera (with 4k 60fps, a fast tracking autofocus and the battery life to support it for video, interchangeable lenses and my name on it in big phosphorescent letters) - which  isn't in the books for anytime soon - making do with what we have, means finding the  focal / aperture  / dof / working distance combo which is the most practical for video shooting.

One interesting aspect of shooting video with a small (1") sensor camera with a set lens, is that diopters behave a little differently than with a larger sensor, especially in terms of range.

This is Lumix LX10 / compact specific info, but just to give an idea, land tests give me the following working distance / focus ranges at 36mm (which is widest I can go shooting in 4K) and 52mm (which is  comfortable for BW style critters)

(as a reminder, the Lumix LX10 has a 1"sensor, and 24m -72mm equivalent zoom lens, f1.4 to f2.8
In 4K, it is 36mm to 72mm)

 

These tests were done on land at f/8
 

NO DIOPTER

36mm
min: glass / 5mm
max: 220mm + / infinity

52mm
min: 90mm
max: 260mm + / infinity

INON UCL165 +6

36mm
min: glass / 5mm
max: 220mm

52mm
min: 30mm
max: 210mm
 

AOI UCL 09 +12.5

45mm min (vignetting)
min: glass
max: 110mm

52mm
min: glass
max: 120mm

I also tested my beloved Nauticam CMC-1, but it is less flexible than the AOI UCL09 for this application, so it stays home. As you can see, the Inon UCL165 and AOI UCL09 diopters are interesting tools, allowing me to focus to the glass but also almost 10cm out.

But for video, more than working distance / minimal focus distance, the crucial issue here is depth of field, and efficient ways of maintaining focus within it for 10 second shots...

cheers
b

Edited October 28 by bghazzal

[Nikolausz]

I've just seen this video on Ikelite YT page. There are some interesting inside about the techniques used and some really great footage.

 

[Davide DB]

24 minutes ago, Nikolausz said:

and some really great footage

 

Yes, great footage.

It's easy to find incredible photos but few good video.

Looking for a Davinci AI plugin to remove swarm and Backscatter from video 😉

 

[bghazzal]

Thanks for that, always good to see some nice BW video clips!

Swarming is mostly a bonfire issue - it does happen in BW but to a much lesser degree, and much more occasional, hitting a big school of copepods for instances.

You can look at the light beams and how clean it is:

What strikes me here is two things:
first the nature of the footage critters: the video footage is mostly of salps and ctenophora, with some fish and a couple of more exotic critters.

This is interesting as while David Fleetham's commentary is about the vertical migration of deep dwelling species, we are shown images of critters which are encountered also commonly encountered in the shallows. Maybe it was a bad day, nature is unforgiving...
And yet even if the same species can be commonly encountered on shallow bonfire dives in certain locales, the great thing about being in open ocean like this is less swarming, and potentially better viz with less runoff and other terrestrial influences.

Also, from a technical point of view, these are some of the bigger and slower critters encountered on these dives, and are slo-moed above 50%.
I'm happy to see big-floats in there, which definitely help with stabilisation. Also not sure if these rigs are video focused or not - the main video light used here seems to be the wide video light, top mounted, and primarily functioning as a focus light.

cheers
b

Edited November 9 by bghazzal