Hello all – I find myself bringing the subject up in various threads (notably here, here  here) so instead of further derailing innocent threads with my ramblings, I figured I'd try to regroup the discussion here.

 

I’d love to discuss and share experience reports on a specific shooting technique now often seen as antiquated and unneccessary due to technological advances made by recent action cameras:

 

shooting with “red” filters in ambient light, in a flat profile, and manually white-balancing the footage in postproduction

 

  
What you’ll mostly read/ hear everywhere nowadays is basically that recent camera white-balancing capacities have improved to the point that the "balancing" effect of filters on a camera lens are no longer necessary underwater, or even detrimental to the overall quality of the footage.
 

While there is definitely a very good case for this in the shallows / snorkeling zone, where spectrum alteration is not as marked, I’m actually less convinced by the results what I am seeing at depth when it comes to ambient light shooting.

Of course, speaking of colours, grading etc, is entering into slightly dangerous, highly subjective territory – but let's say it - to my tastes, I find that in most cases, if the ambient-light underwater action cam footage I see is certainly usable, sometimes even quite ok with recent models, fine in the shallows,  I’m much less convinced by the overall balance of colours at common diving depths, especially in the 10m to 20m zone (as a disclaimer, I do have a fondness for the soft "ambient light" look in wide-angle, ambient light-dominant scenarios)
 

Personally, I’ve been shooting in flat with filters since the GoPro4, following this up on the GoPro 6, and 7 models.
 

But the change in housing design with the upgrade to the GoPro8 has made it more difficult to use filters on the GP8 and up, a reality bolstered by the widespread idea that recent model’s auto white balance capacities underwater rendered filter use obsolete.
 

I actually held back from upgrading the cam primarily for this reason, and also because I was not fully convinced by the improvements brought about by each yearly update - but now, generations later, there is probably a reason to do so - but what of the filters?

the filter + flat profile combo

 

Now when I say I shoot with filters, what I really mean is that I shoot ambient light with filters in a flat profile. This is important, as it's quite a different approach to simply slapping on a filter and letting the camera figure things out for itself underwater.

 

By that I mean that I’ve been using the ProTune settings since they were implemented to basically stop the camera from trying to make white-balance and colour adjustments.

Human control (freakism...) - the opposite of AI if you wish 😁

In GoPro's ProTune mode you can set the cam to a relatively flat profile (on mine, which is a GP7, it's called "native" for WB, and colour profile to "flat"), which works great for grading in post, and most other action cam models have similar "flat" settings (log, etc)

 

As always with flat profiles designed for grading, this gives a rather dull-looking result out-of-the box (read not great for near instantaneous uploads to your socials of choice...)

However - and this is where it gets interesting in my opinion - the recorded colour data is there, just waiting for you to work on it in post.

 

This allows you more flexibility to set the white balance (sand, bubble, tank, white little rock or even WB slate if you have one) in post-production instead of at depth, and also work with the colour and overall colour balance.

Clearly, this is more dedicated approach - it  does mean more work in post, and also means having access to editing software which gives you the option to work on the footage (Adobe Premiere, DaVinci Resolve, Final Cut Pro, Sony Vegas, iMovie all work, but there might be others).

So yes, more work, but since I always work on the footage anyway, I find that this approach generally gives a much more balanced ambient light colour cast, and actually in the end not that different from having performed a manual white-balance at depth...

Which is why I would argue, to the risk of sound vintage retro and old-fashioned, that lens filters, especially combined with flat-profile shooting, still have a role to play underwater video, especially on action-cams, since they allows us to make-up for these camera's lack of manual-white balancing functions at depth.
 

Yes, action cams cannot manually white-balance at depth – but shooting in a flat profile with a filter actually gives you a rather similar flexibility, by allowing you to set/adjust the white balancing point and colour balance in post.

However, having tried to work on footage shot in non-flat profiles, it just doesn't work as well in post if it's not been shot flat -  adjustement options are very limited, since you are pretty much locked-in to adjustments made in camera by the camera software. 

Shooting in flat is something of a radical choice, but if you intend to work on the footage in post, I would recommend shooting in flat whenever possible, for maximum flexibility and more balanced results.
 

And in this shooting scenario, the physical "red" filter in itself plays an important role   - as it does for ambient light with manual-balancing cameras - by, very schematically,  "evening out" the balance of the data recorded in the flat profile.

Using a filter at depth means that there is not as much of an adjustment/stretch to be made to rebalance the wb point and colours of the footage - the logic when rebalancing flat-filtered footage in post is basically the same as when manually white-balancing at depth, and end results actually somewhat comparable.
 

Of course camera capacities / IQ come into play, which is why I would be very eager to see footage shot in flat profile with a good filter on more recent, advanced action cam models!

 

Again, this is very subjective territory - as are most colour / grading considerations - and what follows is a very subjective statement, but for shooting at depth in full-ambient light or ambient light dominant scenario, I would argue that a (good) filter really helps, and that shooting in a flat-profile, will allow you to get better balanced end results (with the “ambient light look” ) than using no filter. It's worth losing some stops of light and/or definition for this
And I would argue that this is also the case for quite a few cameras capable of manually white balancing at depth as well...

 

 "good" filters

 

I’ve mentioned “good” filters a few time – what I mean by this is that I also believe that filter quality matters, an aspect which is often overlooked.

 

Sure, any “red” filter will help to some extent, but if we are talking about obtaining a well-balanced end result, then filter design, and its action on the spectrum, is also important.

This is also a little controversial, but I’m not very convinced by the results obtained on commonly referenced filters like the PolarPro or even Flip system (which seemed, awhile back to give slightly dayglo results...) – this is also true for other cameras.
As example, on my Lumix compact (which I manually white-balance at depth, the difference in results between a generic “red filter” (the Howshot filter I first  used with my Inon UWL-100) and well-designed filters like the Magic Filter or Keldan are massive, for instance (eg Hotshot generic vs. Magic vs Keldan )

You need to find something that works in your conditions, and gives you a good base to work with post.

 

For more on this, I would recommend reading up on the fantastic tests and experimentations done by Interceptor121 in his blog and by Dreifish in this fantastic Wetpixel thread (read it while you still can!)

All this to say that actual filter design does matter underwater,  it's not just a question of sticking any red piece of glass or plastic on your lens -  this is true for any camera, and this statement applies to action cams as well.

 

Furthermore, I think the true value of using filters at depth come to light when combined with a flat profile, for white-balancing in post - this crucial aspect is very often overlooked in the discussion on filters.

If planning to work on / grade the footage, flat profiles are the way forward for shooting ambient light at depth, especially in the zone where you still have a lot of ambient light and colour info making it to the sensor, but the spectrum is quite alterated (typically 10-20m in strong ambient light areas - ymmv...)

back to the future - using filters in an unfiltered world

 

Personally I’ve been using the UR-Pro cyan filter since the GoPro4 – it’s a proven, solid filter, that works very well for this purpose (shooting in flat, white-balancing in post)
I have 55mm SRP-UR-Pro acrylic filters, and also a glass UR-Pro filter.

I'll post some ambient light examples, shot in different locales/conditions, below.

Alas, UR-Pro doesn’t exist anymore, and it’s also very difficult to find 55mm / 52mm filter mounting options for recent cameras.

And this is also complicated by the new wide-angle lens accessory options, which are seemingly finally improving quality and reducing wobbly corner / angle distortion which have plagued GoPros since stabilisation was introduced and the field of views changed (GoPro4 > GoPro5, corners, argh...)

 

Also shooting ambient light with a compact camera I had, however, also experimented with filter gels, which are the way to go with fixed-lenses (taped to the lens, not super sexy but hey...), notably the Keldan Spectrum filter (SF -1.5 gel) and the original Magic Filter gels.

I tested the Keldan gel on the GoPro when I was working in Palau and also using the UR-Pro daily.
Well, while the Keldan's high quality filter turned out to be my favorite filter for ambient light on my Lumix compact, it was actually a let down on the GoPro...

 

However, after a long hiatus, I finally got the motivation to try something I hadn't done yet, which is use an original Magic Filter in the GoPro7 housing.
I tested this yesterday morning, and results were quite good, making it a good option for use with newer camera models and accessory lenses, no mounts required.

This is something I would be pursuing in the future if I finally upgrade my veteran GoPro7 to either a newer model or a DJI or Insta360...

More testing would be needed, but seems like a good solution for using filters on recent action cams.
Sure, you would lose the option to shoot with the camera (in the housing) topside and on the surface, but it's always a compromise, and there's always the solution to remove use it out of the housing.

 

ambient - and beyond?

Of course, all this applies primarily to ambient light / dominant ambient light shooting.

 

If artificial light is your dominant light source, auto WB or a kelvin setting is the way to go - unless.... you use ambient light filters on the lights, ie a mixed lighting scenario.

While this has been mostly researched on cameras capable of manually white balancing at depth, this his would be something interesting to play with with action cams shooting in flat profiles - a good base to start would be the ambient filter tests done by Interceptor121 and Dreifish previously linked above, matched to good quality, well-designed camera filters (the usual suspects being the late UR-Pro, Magic Filter and Keldan Spectrum filters).
 

If the filtered artificial light spectrum is very close to matching the ambient light spectrum, given the results given by white-balancing flat profile in post, there's no real reason it shouldn't work like it does on manual white balance capable cameras....

 

To conclude, and go beyond filter use,  I'd love to see examples of ambient-light footage shot in flat and rebalanced in post ( especially with a good filters) on recent action cams, and seeing how these results compare to in-camera adjustements, and recent AI-enabled depth-correction bells and whistles.

cheers
ben

 

 

 

Edited June 14, 20241 yr by bghazzal

Some examples, in different locales/conditions/depths, of GoPro 4 to 7 footage shot exclusively in ambient light only with UR-Pro filter, flat profile white-balanced in post.

GoPro4 ambient light + UR Pro

(no stabilisation, not waterproof, colours were not great, but loved the sharpness and corners of medium fov, and miss it to this day...)

 

- GP4 Raja Ampat - 15m

This is one of the first clips I shot to test the GoPro4 + UrPro combo in 2016 or 2017 - not sure what I graded it on...  not interesting but illustrates GoPro4 + UrPro filter colours.
 

- GP4 Andaman Sea, 10-20m depth

 

 

- GP4 LaPaz Mexico (greener water) 18m depth (it's on the sand next to a wreck)

 

- GP4 LaPaz 15-20m (the mantas returning after 20 years were big news in 2018) - average viz not much colour there, shows the limits of the GP4_filter at depth.
 

- GP4 LaPaz 10m depth
 

 

GoPro 4 to 7 + UR Pro

Andaman Sea, various depths
 

 

Edited June 14, 20241 yr by bghazzal

GoPro 6 + UR Pro

 

I hated the clunky 6 and only used it a few times - let down coming from the 4, even though the newly introduced stabilisation was nice and waterproofing offered peace of mind.

 

GP6 Andaman sea, 12m-15m depth

 

 

Edited June 14, 20241 yr by bghazzal

GoPro7 + UR Pro

Coming from the GoPro4, corner distortion is ugly (this is all in linear, argh), but there's colour data more to work with at depth - hope the new accessory wide angle lenses + wide will help!

I was shooting in linear 2.7K as it's a hassle for my aging laptop to handle GoPro's HEVC265 4K codec (the joys of nomadism...)
 

- GP7 Komodo  - 10-20m depth
 

 

- GP7 Komodo  - various depths, mostly shallow 10m

 

 

- GP7 Japan (quite blue water, 10m depth)

 

 

- GP7 Japan 10m, greenish water,

 

 

- GP7 Maldives  - Kudara Thila roughly 10-20m, early in the season so greenish/bad viz that year...
 

- GP7 - Palau / Blue Corner, roughly 15m
 

 

 

- GP7 - Palau 18m - average viz, but the white sand acted as a reflector
 

- Finally, here is yesterday morning's test of the GoPro 7 with an original Magic Filer
 

GoPro7 4K60fps wide (yes, not 2.7K this time), flat profile, with an Original Magic Filter gel inside the housing.
 

First clip is at 18m and second clip at 9m, shot in tropical (east Bali) mid-morning ambient light only in the usual slightly murky, particle-rich water.
You can see how strongly the Magic filter is kicking in when I film the depth on backlit dive computer.
 

cheers

Edited June 14, 20241 yr by bghazzal

Those videos make me sad because it's more than 2 months until we can step in the water again...

 

interesting though the discussion about filters.

This weekend I'll try to get some test footage and post it raw and color graded without filters, but with the wide angle lens (GoPro 11)
I am really interested in this topic, and I'd really like of what people prefer between red filters/no filter, the time where I used them, I mostly screwed up with a bit too unnatural color

9 minutes ago, Elvandar said:

This weekend I'll try to get some test footage and post it raw and color graded without filters, but with the wide angle lens (GoPro 11)
I am really interested in this topic, and I'd really like of what people prefer between red filters/no filter, the time where I used them, I mostly screwed up with a bit too unnatural color

Great thanks - not sure how easy it is to change the settings UW, but if you could film a short clip, like 20 seconds of a good colour reference at depth in a flat profile (before it was ProTune on, white balance native, gopro colour, not sure on the 11) it would be great. if your editing software can't white balance I can try to do it for you on FCPX - really interested in what the GP11 is capable of in flat.
I've also just bought the wide angle lens (an AOI UWL-03), should be arriving in a few days.

I am trying also the QR code labs feature, I am printing the code for fast swap of the settings

16 hours ago, fruehaufsteher2 said:

Those videos make me sad because it's more than 2 months until we can step in the water again...

 

interesting though the discussion about filters.

Aah, winter... hang in there!

Interestingly, looking at this small sample of an 8 year GoPro footage history  despite clearly capturing less colour info, I have to say I like the definition of the GoPro4 best - In good conditions like the first tentative Raja clip, I really like the ambient light results - but... it the GoPro4 was not waterproof  (stressful as...), had no stabilisation and no 4K...

When I started working in the Andaman sea, and in the Sea of Cortez and still used the GoPro4, dives were generally deeper and with less light - and you can see on the same GoPro4 setup clips that  the loss of colour is more marked than with GoPro6 and 7 - the UR-Pro filter helps balance the footage out, but there's just less colour info - Shallows are good, but deeper tends to be more monochrome.

Recent models like the GoPro 6 (the Koh Tachai clip was quite deep - 12m at the shallowest, and there's just much more colour info, it's not monochrome anymore) and then the GoPro7 were an improvement in this aspect, but the rest, not so sure.
It's just less sharp in general (all footage was shot with the sharpen setting on "medium"), more distorted.

A lot of things changed when the GoPro5 was introduced - I'm guessing this is linked to stabilisation and the change of fov. I'm guessing that beyond the field of view change, the processing power need to handle stabilisation is what affected overall image quality - the conceptors seemed to have focused on improving colour capture more than anything else.


Now that the corner issue can be dealt with by accessory lenses, it would be great if newer action cam models (GoPro or other) allowed us to reclaim the overall image quality models like the GP4 had, with improved colour capture and dynamic range.
The Magic Filter looks promising - I quite like the skin tones on the 9m clip for instance, which is a good sign.

I would love to try playing around with a short ambient light clip shot in a flat profile at depth (ideally with a good filter on the lens!), so I could see what's under the hood now.

I find it very difficult to make sense of action cam comparison vids, because they're systematically shot in auto WB/colours, often showing only non-graded, out-of-the-box  footage, straight off the double-mounted tray.
I understand the logic, since this is how most users use/wish to use action cams in our days of instant-sharing and interaction , but it's also nice to see what graded, end-result ambient-light footage looks like.
 

Edited June 15, 20241 yr by bghazzal

On 6/14/2024 at 1:45 PM, bghazzal said:

Great thanks - not sure how easy it is to change the settings UW, but if you could film a short clip, like 20 seconds of a good colour reference at depth in a flat profile (before it was ProTune on, white balance native, gopro colour, not sure on the 11) it would be great. if your editing software can't white balance I can try to do it for you on FCPX - really interested in what the GP11 is capable of in flat.
I've also just bought the wide angle lens (an AOI UWL-03), should be arriving in a few days.

I am slowly editing the video I made, I had a really short time to film them, with a bit of a poor result, because it was the last dive of the day, and the guide was going pretty fast 😅
In the next days I'll put the wide clip result, I started with this since I wanted to check the result with the Inon g55
 

 

 

PS, I am not convinced that i made a good job on the color correction, any suggestion on that?
I am still learning how to do it 

Hi and thanks.

The first thing that strikes me is the pink / magenta colour cast which shows up in the white of the sand.

 

On one hand, the colour correction works because you got rid of the excess green, restored the colour of the hermit crab and made it more visible. But the pink colour cast is odd.

Are you shooting with an artificial light and a filter?

If you can white balance manually in your editing software, did you try white balancing on an area of the sand which should be white (eye-dropper, select an area or the sand which should be white)? You could try different areas, see what works best at restoring some colour balance.

 

For the pink/magenta hue my editing software (FCPX) I would try to use the hue vs. luma tool and the colour curves to dial down the cast, but it would better not to have it in the first place, as heavy colour adjustments degrade image quality as well.

 

 

 

Edited June 21, 20241 yr by bghazzal

4 hours ago, bghazzal said:

Hi and thanks.

The first thing that strikes me is the pink / magenta colour cast which shows up in the white of the sand.

 

On one hand, the colour correction works because you got rid of the excess green, restoresd the colour of the hermit crab and made it more visible. But the pink colour cast is odd.

Are you shooting with an artificial light and a filter?

If you can white balance manually in your editing software, did you try white balancing on an area of the sand which should be white (eye-dropper, select an area or the sand which should be white)? You could try different areas, see what works best at restoring some colour balance.

 

For the pink/magenta hue my editing software (FCPX) I would try to use the hue vs. luma tool and the colour curves to dial down the cast, but it would better not to have it in the first place, as heavy colour adjustments degrades image quality as well.

 

 

 

I shoot without red filter and with light, but I used my spot lights, because I did not have time to adjust my two video lights, overexposing a lot in certain areas of the image, mostly on the sand 🙃

 

I am using Da Vinci Resolve, but I moved way too much settings I think, I have to try again the colour balancing starting from the base video 😁

The clip is so badly damaged that it is difficult to save it. The pink cast is terrible.

As Ben wrote, you should start over and start on Resolve with a WB on a spot that is presumably white. Try sand.

Do several trials on different spots and once you find an acceptable compromise, start with the adjustments.

Take two on the clip, changed way less value, and the result looks better to me, i also cut the last part where colour went crazy, and slowed down the clip
 

 

 

The next video I'll put the wide angle shot, I promise to stop fiddling with the hermit crab 😁

Thank you for kicking off the discussion, @bghazzal. Anticipating that Wetpixel may not be around forever, I archived the findings from my 2022 tests of the red filters and blue (ambient light) filters here. 

3 hours ago, DreiFish said:

Thank you for kicking off the discussion, @bghazzal. Anticipating that Wetpixel may not be around forever, I archived the findings from my 2022 tests of the red filters and blue (ambient light) filters here. 

Awesome, thanks a bunch for this!

cheers

ben

As a follow-up to this discussion on shooting action cams in flat  profile with filters, I ran a little twin test the other day, to test differences between:

- shooting  GoPro7s in a flat profile in ambient light with a UR-Pro Cyan filter and with the original Magic Filter

- while also testing between results of using GoPros with an accessory wet wide-lense (the AOI UWL-03 that I had just received) and without one.

 

Test parameters were as follows:

Both camera gootage is shot on two GoPro7 black cameras using the same settings, ie. shooting in 4K 60fps, in a flat profile (white balance native, go pro colours), wide fov, sharpness high, ambient light only.

 

• GoPro 1, on the left side of the screen, is mounted with the AOI UWL-03 wide converter lens and an original Magic Filter gel placed inside the supersuit housing.
   
• GoPro 2, on the right side of the screen, is using only a UR-Pro Cyan filter (on a 55mm SRP-Blurfix adapter) mounted on the supersuit housing.

 

- Footage is taken in the afternoon, tropical ambient light only, in slightly murky, greenish/blue water, between the surface and depths of roughly 10m / 15m at the deepest.
Not perfect conditions, with generally brownish - and sometimes bleached out - impacted coral reef / bommies.

- Both cameras are handheld, each in one arm/hand, and the footage is shot in motion while swimming on breath-hold (but without weights…),  so not the most stable or consistent with angles, but still gives a good impression.

 - Footage is left unstabilised in post to highlight differences in the lenses' shooting angles.

 
- The test clip was also quickly graded in FinalCut ProX (shooting / testing conditions were not fantastic, and I didn’t spend hours on it - this was mostly to highlight the differences in filter response)

 

 

Here is the ungraded comparison test clip, flat profile out of camera:

 

 

Here is the same comparison test clip quickly graded in FinalCut ProX:

 

 

And lastly here is the GoPro7 with AOI UWL-03 + Magic Filter, quickly graded footage only:
 

***

 

First observations

 

On the UWL-03 wide lens:

 

- The wide-lens is lovely – it’s actually quite easy to handle underwater, feels about twice the weight of a standard GoPro.
It was possible to hold one in each hand to shoot the test – not ideal of course, and this snorkeling/freediving footage isn’t the most stable, but not too bad given the circumstances…
 

- The lens widens the basic GoPro fov as expected, reduces working/focusing distance (see closeups), and basic splitshots - within a very limited range - are somewhat possible with the lens (as in the final shot)
 

- It also does seem to improve corner warping / deformation underwater (not so much on land), which has plagued GoPros since the GP4 to GP5 design change.
 

- This would need to be checked, but the AOI lens didn’t seem to require burping out air bubbles (whereas the UR-Pro / Blurfix certainly does (since I kept descending and ascending while freediving a couple of bubbles made their way into the UR-Pro filter on some of the first clips...
 

 - There is no vignetting in Wide fov on the GoPro7. The black corner you see on some of the footage comes from me holding the two cams together in each hand to try to align the cam’s shooting angle.

 

On the filter/ footage grading:

 

- The conditions were really not great, a bit murky as usual, and water a little greenish.
 

- The footage itself isn’t super well exposed – light was harsh in the afternoon ( overexposed on skin closeups) and I was filmin on breathold and in motion, so limited time at depth, and paying less attention to the angle of the light than I would diving.
 

 - I usually shoot in medium sharpness, I boosted it to high to see how this would work with the contour damping effect of the filters, but I think I’ll probably go back to medium as it does look a little too crisp/artificial for my tastes.

 

Both filters work for ambient tropical light, but with differences:

 

- The Magic Filter gel works with the GoPro, but the UR-Pro generally looks better ungraded, and importantly requires less adjustments in post, as it’s more uniform.
The UR-Pro often gives deeper, somewhat more pleasant blues (even if the water was a little greenish), as a base.

- The Magic Filter has a milder effect, but really kicks in close-up, sometimes a little too much for my taste in the reds/oranges.
This means that for wider shots, red spectrum / colour rendition is more pronounced for objects further in the background with the UR-Pro, whereas they will stay green/blue with the Magic. The colour balancing effect is generally stronger with the UR-Pro - this is very clear at 1'50" of the clip for instance

- However, closeup at depth, the Magic Filter does seem to really boost the oranges, which can sometimes be a little problematic to deal in post (10m depth seafan / coral bommie shots, for instance like this sequence at 30" in  or 1'40" in), whereas the UR-Pro is more consistent and uniform.
 

- That said, in certain conditions, colour rendition actually does actually seem more accurate / lifelike with the Magic Filter than with the UR-Pro, which is generally more “baked-in”, with brownish hues (something I'd noticed in different conditions as well).

 

- The filters were never removed from the camera, showing that both filters are useable on the surface in a flat profile, and don’t give crazy reddish hues, unremoveable in post -  remember these tests are  shot freediving, so with a lot of quick depth variations - scuba footage doesn't have the same variations, so much easier to grade consistently.

 

- I started by grading the Magic-Filter + lens footage, then worked on the UR-Pro one, and found myself adjusting the Magic-Filter footage again to try to match the UR-Pro results in quite a few clips…

 

Overall, since the AOI Wide lens performs nicely, it would be great to be able to continue use the UR-Pro Cyan filter for this shooting style, as it’s more consistent overall on the GoPro, and this consistency of the UR-Pro filter really does make life easier for grading, as there are less surprises.

cheers

 

ben

Edited June 22, 20241 yr by bghazzal

Log story shorth: the UR-Pro footage SOOC seems better but I like your quick grading of the Magic filter.

 

Searching for UR-Pro on the Web I see a lot of results. Are all inventory stocks?

Remain the fact that for the AOI lens you would need a gel version and I've never seen it. 

 

During my previous testing, I found that the UR Pro filter is more effective at filtering out green light than the Magic filters, which is exactly what I think your tests demonstrate, @bghazzal. In fact, the Magic filters seem to do very little to filter out greenish cast in the ambient light.

 

• No blue filter, no orange filter = 4700K, +20M
• Lee 353 + Rosco 4330 = 23000k + 121M
• Lee 353 + Rosco 4330 + Original Magic Filter = 8600k + 150M, -1.3 stops of light
  • (Worst of the bunch.. looks more yellow than orange. introduced MORE green into the image, to the point where even after doing WB in post, there was a greenish tinge. Did a decent job of countering the blue, but would not recommend using it in anything but the most pristine blue waters) 
• Lee 353 + Rosco 4330 + Auto-Magic Filter = 10000k + 121M, -1.6 stops of light
  • (Decent. Knocked out most of the blue cast, but did little for the green. Again, use only in pristine blue water)
• Lee 353 + Rosco 4330 + UR/PRO CY Filter = 7700k + 77M, -2.3 stops of light
  • (Most effective. Knocked out most of the blue AND green cast. Would be a good all-around choice)
• Lee 353 + Rosco 4330 + Keldan SF-2 Filter = 12000k + 114M, -2.5 stops light 
  • (Ok, but a bit weak. This is designed to compensate only for 4 meters of water depth, and reduces both the blue and green tint. However, in this case, the lights with the Lee 353 and Rosco 4330 filter appear to be better matched to 12 meters of depth..)

 

 

It would be great if a UR Pro gel were still available, but to my knowledge it never was. I think your grading results look great though, so there's enough color information in the flat footage to be able to color correct the green bias in post.

 

Another option might be to supplement the Magic Filter with another gel filter that predominantly filters out green -- some sort of magenta filter I guess. Perhaps there's something like that in Rosco or Lee's range of filters?

You could try supplementing the Magic Filter with something like this that filters out green preferentially:

https://us.rosco.com/en/products/filters/e279-eighth-minus-green

https://us.rosco.com/en/products/filters/r3314-tough-14-minusgreen

https://leefilters.com/colour/249-quarter-minus-green/

https://leefilters.com/colour/249-quarter-minus-green/

https://leefilters.com/colour/248-half-minus-green/

19 hours ago, Davide DB said:

Log story shorth: the UR-Pro footage SOOC seems better but I like your quick grading of the Magic filter.

 

Searching for UR-Pro on the Web I see a lot of results. Are all inventory stocks?

Remain the fact that for the AOI lens you would need a gel version and I've never seen it. 

 

That’s pretty much it.
The Magic Filter footage is definitely useable/workable, but more fiddly, and also I’m a little worried about the red/orange/magenta boost at depth.
 

It’s still possible to find UR-Pros online, especially Ikelite models, but they’re generally in odd sizes.
I wonder if it would be possible to DIY it, hole-punch / cut up one of the UR-Pro acrylic filters to make a piece which could slide in the supersuit, or another system which would allow the use of an UR-Pro filter with the lens (as mentioned earlier, I'm not interested in Backscatter's proprietary filter solution).

if going the filter + screw on lens (with the 55m adapter for instance, like Inon's), I'm afraid it will induce vignetting / distortion on the UWL-03... Having a slip-in filter in the housing is clearly the best solution in this case...

Edited June 23, 20241 yr by bghazzal

15 hours ago, DreiFish said:

During my previous testing, I found that the UR Pro filter is more effective at filtering out green light than the Magic filters, which is exactly what I think your tests demonstrate, @bghazzal. In fact, the Magic filters seem to do very little to filter out greenish cast in the ambient light.

 

• No blue filter, no orange filter = 4700K, +20M
• Lee 353 + Rosco 4330 = 23000k + 121M
• Lee 353 + Rosco 4330 + Original Magic Filter = 8600k + 150M, -1.3 stops of light
  • (Worst of the bunch.. looks more yellow than orange. introduced MORE green into the image, to the point where even after doing WB in post, there was a greenish tinge. Did a decent job of countering the blue, but would not recommend using it in anything but the most pristine blue waters) 
• Lee 353 + Rosco 4330 + Auto-Magic Filter = 10000k + 121M, -1.6 stops of light
  • (Decent. Knocked out most of the blue cast, but did little for the green. Again, use only in pristine blue water)
• Lee 353 + Rosco 4330 + UR/PRO CY Filter = 7700k + 77M, -2.3 stops of light
  • (Most effective. Knocked out most of the blue AND green cast. Would be a good all-around choice)
• Lee 353 + Rosco 4330 + Keldan SF-2 Filter = 12000k + 114M, -2.5 stops light 
  • (Ok, but a bit weak. This is designed to compensate only for 4 meters of water depth, and reduces both the blue and green tint. However, in this case, the lights with the Lee 353 and Rosco 4330 filter appear to be better matched to 12 meters of depth..)

 

 

It would be great if a UR Pro gel were still available, but to my knowledge it never was. I think your grading results look great though, so there's enough color information in the flat footage to be able to color correct the green bias in post.

 

Another option might be to supplement the Magic Filter with another gel filter that predominantly filters out green -- some sort of magenta filter I guess. Perhaps there's something like that in Rosco or Lee's range of filters?

 

Yes, that’s exactly it - clearly the same issue, and as you pinpointed it, it's clearly a weaker filtering of the greenish cast in the ambient spectrum...

This is actually also an issue I had why I tried the Keldan Spectrum on the GoPro – it was way too weak, even in Palau’s generally blue water.  And to add to this, the Keldan Spectrum gels are also slightly weaker than the SF2, as they’re sold as SF -1.5... 
 

Interestingly, they worked fine when manually white balancing the Lumix to ambient light at depth, the UR-Pro performing the worst of the 3 (Keldan SF-1.5 gel > Original Magic > Ur-Pro CY), and it’s exactly the opposite ranking on the GoPro (Ur-Pro CY > Original Magic  > Keldan SF-1.5 gel ), which clearly needs a more powerful filter.
 

All this highlights the importance of having a good filter to work with, as I was mentioning above – these 3 filters are well-designed specialist products, and yet they behave quite differently, which becomes even clearer when working on the footage in post...
 

While the Magic is useable, I’m also a little worried about the red/magenta/orange boost I saw a little deeper – I’m guessing it’s also an effect of the white balance being thrown further off by blues/greens, leading to overcompensation.
This is not good, as needs more work in post and this further degrades image quality…
 

Thanks for the links to gels, it would be great to try to reverse-engineer the UR-PRO SWCY formula – they really had something going with it, it's really a shame to lose such a tool.

 I have no idea how manufacturers work on designing custom filters – do you think the UR-PRO SWCY was actually a proprietary design (like the Keldans), or based on some pre-existent tone (which could then be source / purchased?).
 

Unfortunately, the Rosco/Lee gels are near impossible to source in Indonesia – there is a distributor in Singapore, but the sample I ordered didn’t make it over the first time, and with custom duties it’s a bit of a nightmare.
I think it’ll have to wait a few months, when I’ll be in a place where gels are not accessible...

cheers

 

 

Edited June 23, 20241 yr by bghazzal

2 hours ago, bghazzal said:

I'm not interested in Backscatter's proprietary filter solution).

 

Actually I think it would be very easy to cut a rectangular slice from an existing circular filter. IIRC the Backscatter filters are just the central sections of circular filters...

30 minutes ago, Davide DB said:

 

Actually I think it would be very easy to cut a rectangular slice from an existing circular filter. IIRC the Backscatter filters are just the central sections of circular filters...

I don't really understand how the Backscatter mount modification works. They use an AOI mount, but with a slot on the side to slide the filter in.

But looking at my AOI mount and lens, don't really see in what space the filter would fit in, since as far as I can see it sits snug next to the port. So was it moved further away from the port?

 

The other issue is that my acrylic UR-Pro are not flat, but slightly rounded, as per the SRP design...

This is the page for their rebranded AOI lens kit. Down to the page you can see the slide-in filter.

Maybe you can understand the differencies between you kit and that.

Ad you can see that filter is just a cutout of a circular filter.

 

https://www.backscatter.com/Backscatter-GoPro-Action-Cam-Sharp-Wide-Lens-Pro