Underwater Bubble Blowers User Group Thread NEW

[Frazier Nivens]

Very pretty shot Pawel, worthy of David Doubilet. Significant softness on the edges though, is that the aperture you were using?

Depth of field in Pawel's image Tom, congrats Pawel, the stills look great, will love to see some motion imaging.

Best,
Frazier

 

[Rudi Herbert]

I have seen some amazing things of late...

For the past 8-9 months I've been involved in some private tests regarding an optical system for an upcoming theatrical feature that will have a substantial number of underwater scenes. My part of the tests consisted of shooting a large number of set ups using both the Nikon 14-24 and Tokina 10-17 lenses, as well as some custom filters. The idea of the test was to determine a lens upon which to build a complete optical ecosystem if you will, including ports, filters and focus drivers. Among the lenses tested were everything from the new Leicas to the Zeiss MP 14 mm and UP 8mm, to a few 8 mm super fisheyes, to some submersible lenses. Interestingly, the Nikonos 15 mm thata has received so much notice of late was discarded early on, its optics deemed antiquated and its sharpness problematic at anything under f8, while the Sea & Sea 12 mm superwide was found to use substandard glass in its construction. The nikonos RS 13 mm lasted a little longer but its lack of modern features such as anti glare and low dispersion coatings did it in, nobody able to defend the glares it produced under certain conditions, though its sharpness and ruggedness was loved by all. The final winner has been, no surprise to me, the Nikon 14-24.

This lens outresolved anything else it was compared to, but at the same time, presented some problems with sharpness when used behind traditional domes. The same company that built the color filters we're using, then went to work to build a special dome port to precisely match the characteristics of the lens. I have just seen the first prototype and saw some images shot by another person with this very dome and I was left speechless. The images did the lens justice for the first time, matching the corner to corner tack sharpness the lens is capable of, something that has not happened until now when the lens was used behind conventional 8'domes. what I can say is that this dome is around 9.3/4" wide, and its shape is PERFECTLY spherical, in keeping with he old precept that a lens' nodal point MUST be equidistant from both the center of the dome and its sides, something ignored by most manufacturers of large domes. The dome also boasts thicker glass on the peripheral areas which then perfectly tapers in towards the center, where the glass is actually paper thin. I will shoot some tests for the next 3-4 weeks with this dome, at which point I will receive version 2.0 of it, which will actually be COLORED! The color correction coating will be integrated onto the dome itself, and for what I hear, there will be two different color coatings, on both the outer and inner surface of the dome. One of the coatings will be lighter, towards golden yellow, the other one somewhat darker, towards pink/violet. They both will be low saturation, yielding about one stop of light loss altogether, and I don't know color goes on what surface.

I cannot share any images from these tests but have been allowed to share some of these technical specs, but I can tell you, this system will blow some socks way, way, way off. In talking with one of the engineers that tackled the construction of the dome, the most important thing I learned, was that although dome optics are very well known, very few manufacturers truly respect them. He told me that after looking at 18 different dome designs, only 2 of them somewhat allowed one of the specified lenses to perform as intended, all others featuring dissimilar distances from center and sides to lenses' nodal points. He also pointed that a dome truly will only match one or two lenses, not a parade of them, as most people try to do nowadays, and that spacers are not the solution to the problem. Lastly, he added that varying thickness in the glass of the lens will help the lens correct many of the problems encountered underwater, but that again, this solution is rarely, if ever, implemented, even though its cost is only fractional compared to domes of even thickness. Anyway, I thought I'd share some of these news, which to me are nothing but exciting.

 

[paul engstrom]

Hey Rudi,

Question for you: "paper thin" near the center implies, to me at least, easily broken... Is there any concern that this port would be particularly vulnerable to breakage or am I misinterpreting things here?

Other than that, this sounds very good!

 

[Rudi Herbert]

Paul,

Well, paper thin might be an exaggeration, I'd dare say the thickness is a hair under 1 mm, but as compared to the edges, yes, it is much thinner. I don't think these people are looking to commercialize this system, in fact, I know they are not, so they don't need to build something that will necessarily stand to the rigors of air travel and getting on and off boats, etc. However, the dome is QUITE strong as is, and I've been assured that an accidental tough here and there won't even scratch the surface, much less break it. The engineers did tell me, knowing my propensity for deep diving, that due to the dissimilar thickness, they wouldn't certify it deeper than 7 atm, or 60 meters/200 feet, but then again, that's a depth most people can happily live with, including me, and also, the intended final use of this dome will be in dive tanks no deeper than 12 meters/40 ft.

 

[Johnny Friday]

Rudi,
is this made for a specific housing? a custom housing being built as well or are there plans to adapt for various housings? I had wondered who would be first to come out with something like this---sort of like the Montreal based housing maker's port that was somewhat rectangular....Yep, 14-24 is exceptional above water...but underwater i'd say it's ok.....it needs help.

 

[Rudi Herbert]

Johnny,

These guys make their own housings, but if ever they decide to make the dome design available, which for legal reasons would be always after the release of the film, I hope that by that point other manufacturers should have experimented with the same idea. Another thing they are working on is an iris based constant (auto?) focus system which is pretty cool. They start with a fixed focus formula based on hyperfocal distance, which with this dome is quite generous, a good 18" or so, and then they are developing auto focus patterns where for each movement, in either direction, of the iris, the module tells the autofocus system exactly what to do and when to stop, thus preventing tracking and other common autofocus errors, so that the focus always remains fixed at the image plane the dome projects. Most autofocus systems can already do this, but underwater, specially with particulate present, the system gets thrown off and focus tracking is quite common. This little module is basically aimed at preventing this sophistication from messing with your focus, effectively eliminating the need to do any manual focus. So far, the idea is that this module will fixed around the lens mount, housing its own motors and boards, so whatever image comes to the camera will always be focused. For what I've heard, it works at 99% accuracy at no less than F4, but then again, through our test we sort came to the conclusion that to really get the best from this lens, you should not be any more open than that. Also, because a lot of the set ups will include plenty of green screen, they want deep focus, so the plan is to actually stay at f5.6-f8 all the time. Again, this is all calculated with the knowledge that they will shoot in highly controlled conditions where, if they need more light, they will bring it down there.

 

[Jon Shaw]

Rudi,

Whats's the advantage of making a coloured dome as opposed to filtration on the lens? Sounds like an amazing piece of kit. I would be interested to hear how you rated each lens underwater

 

[Pawel Achtel]

...They start with a fixed focus formula based on hyperfocal distance, which with this dome is quite generous, a good 18" or so ...

At what focal length and COC would that be and why would it depend on the type of the dome?

 

[Pawel Achtel]

...This lens outresolved anything else it was compared to, but at the same time, presented some problems with sharpness when used behind traditional domes. The same company that built the color filters we're using, then went to work to build a special dome port to precisely match the characteristics of the lens. I have just seen the first prototype and saw some images shot by another person with this very dome and I was left speechless. The images did the lens justice for the first time, matching the corner to corner tack sharpness the lens is capable of, something that has not happened until now when the lens was used behind conventional 8'domes...

So what is the corner sharpness? "Tack sharp" means different things to different people. Can you show the MTF graph for the corners and how was it obtained?

As you would appreciate, it is very hard to compare "tack sharp" with "pin sharp" or "very sharp".... it is just completely meaningless.

How would such a dome overcome image plane curvature and poor short focus performance (geometric distortions and low MTF caused by short lens to virtual image distance) and how can it be any better from the 12" CinePort matched to a Master Prime 14mm (that has the entry pupil point aligned up to fraction of a millimeter with the optical centre of the dome) and which I have done over 4 years ago?

 

[Rudi Herbert]

Pawel,

I think I may have mentioned this before, but these people do NOT much care for shooting charts. They sort through all the footage, pick whatever shots they find most relevant, and then convert them to several formats and watch it all on a 40 ft screen. If they like what they see and it looks acceptably sharp to their eyes, sharpness being just one of many things they look at, then they move on. One of them told me when I asked about charts that many of cinema's and cinematographers' most beloved lenses ever would not pass muster if tested on charts, and in fact, even lenses used for IMAX acquisition do not do so well on charts, so they did away with measurable tests long ago.

As for how I, a non-engineer with an eye for good looking images nonetheless, conclude things to be sharp is by blowing images up to 2000% and inspect the areas within the frame, center, corners, etc. I sometimes measure the jagged edges that make up lines at both center and corners and compare them, in terms of pixel size, and if it looks very close from corners to center, then I conclude the sharpness to be very close to, or equal. That's good enough for me and undoubtedly nowhere near good enough for you, but I'm not alone in this way to ascertaining image quality I can tell you.

I won't argue against your numbers, but I am very curious about a couple of things. First, have you tested your highly rated cineport with a precisely aligned lens such as your 14 mm MP and see what kind of sharpness you get? Is it as bad as your numbers suggest? Or are you testing only combinations that will knowingly yield a subpar result? The other thing that intrigues me is how you have flat out refused the notion that your touted Nikonos 15 is NOT a land lens housed behind a cleverly constructed dome, tailored to that particular lens? You keep saying this isn't the case, even though everybody keeps telling you that this has been WIDELY accepted by Nikon itself and is PERFECTLY documented. So then in your opinion, how did nikon manage to pull this rabbit out of the hat, what is the principle behind that 15 mm lens that makes it so unique? If it is not a self contained system built around a lens with a dome on top, a dome that effectively eliminates aberrations, blurriness and degradation, what is then? I am not being ironic now, I would love to know in your undeniably knowledgeable opinion what is the working principle behind this lens and how you think that such a pattern cannot be replicated EVER AGAIN? Look at the lens in front of your housing and tell me that its surface is not a dome. And when you accept that, then why can't the same thing not be achieved again, especially if we start with a vastly superior lens, such as the Nikon 14-24 (which by the way, outperforms the 14 MP when projected) at its foundation, a lens light years ahead of the old 20 mm lens Nikon used to build your beloved 15 mm Nikonos? Surely you must have been baffled by how something so antiquated and simple has managed to outperform even the efforts of better equipped engineers such as yourself when you built your Cineport, which prior to your Deep X concept, you touted as the sharpest, best dome ever built.

Once again, the images I'm seeing look gorgeous and yes, to my eyes and my non mathematically obsessed brain, perfectly sharp from corner to corner.

Cheers,

 

[Johnny Friday]

Rudi,
have you been to an optometrist lately? They can run a TEST on you and your eyes.

 

[Tom Hamilton]

This from Wiki;

The nodal points are widely misunderstood in photography, where it is commonly asserted that the light rays "intersect" at "the nodal point", that the iris diaphragm of the lens is located there, and that this is the correct pivot point for panoramic photography, so as to avoid parallax error.[SUP][4][/SUP][SUP][5][/SUP][SUP][6][/SUP] These claims generally arise from confusion about the optics of camera lenses, as well as confusion between the nodal points and the other cardinal points of the system. (A better choice of the point about which to pivot a camera for panoramic photography can be shown to be the centre of the system's entrance pupil.[SUP][4][/SUP][SUP][5][/SUP][SUP][6][/SUP] On the other hand, swing-lens cameras with fixed film position rotate the lens about the rear nodal point to stabilize the image on the film.

Rudi, how do you determine the nodal points for the exact lens/camera combination you happen to be using, and the precise positioning in relation to the dome, in order to extract maximum performance from the lens?

 

[paul engstrom]

Paul,

Well, paper thin might be an exaggeration, I'd dare say the thickness is a hair under 1 mm, but as compared to the edges, yes, it is much thinner. I don't think these people are looking to commercialize this system, in fact, I know they are not, so they don't need to build something that will necessarily stand to the rigors of air travel and getting on and off boats, etc. However, the dome is QUITE strong as is, and I've been assured that an accidental tough here and there won't even scratch the surface, much less break it. The engineers did tell me, knowing my propensity for deep diving, that due to the dissimilar thickness, they wouldn't certify it deeper than 7 atm, or 60 meters/200 feet, but then again, that's a depth most people can happily live with, including me, and also, the intended final use of this dome will be in dive tanks no deeper than 12 meters/40 ft.

Thanks Rudi

 

[Tom Hamilton]

Depth of field in Pawel's image Tom, congrats Pawel, the stills look great, will love to see some motion imaging.

Thanks Frazier, hence my question regarding aperture, although Pawel indicated it might be something else.

 

[Pawel Achtel]

...As for how I, a non-engineer with an eye for good looking images nonetheless, conclude things to be sharp is by blowing images up to 2000% and inspect the areas within the frame, center, corners, etc. I sometimes measure the jagged edges that make up lines at both center and corners and compare them, in terms of pixel size, and if it looks very close from corners to center, then I conclude the sharpness to be very close to, or equal. That's good enough for me and undoubtedly nowhere near good enough for you, but I'm not alone in this way to ascertaining image quality I can tell you.

You can tell me all you want, not that I need to believe any of it.:laugh:

You made some claims and comparisons of sharpness of various optical systems. When I questioned the basis of your comparison, you back off and say the sharpness is not relevant and further say that you did not have quantitative method of comparing them. Why do you bother writing about sharpness and comparing one optical system to another, then?
The main reason I use Red Epic is because I do care about image quality, sharpness and distortions and I want my lenses to resolve full image sensor and not introduce image distortions, aberrations or fuzziness.

You are incorrect about people not caring about lens sharpness. Lens manufacturers as well as most professional operators shoot charts and make measurements for very good reasons. If we can't measure it, we can't compare it or improve upon it.

...So then in your opinion, how did nikon manage to pull this rabbit out of the hat, what is the principle behind that 15.

The answer is very simple: They designed it from ground up as an underwater lens and did not try to fit a square peg into round hole (as in adapt a land lens to underwater).

......especially if we start with a vastly superior lens, such as the Nikon 14-24 (which by the way, outperforms the 14 MP when projected) at its foundation...

Again: How do you know this? Have you measured it? You seem to be making many claims that I struggle to understand how you arrived at.
It happens that I projected both and this is not the conclusion that I would make.

Both lenses comfortably out-resolve the Epic sensor on land at f/4.0. Nikon is obviously inferior in that it can not be opened to f/1.2, like Master Prime can. The Master Prime is optimised for maximum sharpness fully opened and outresolves the Epic sensor even at f/1.2 corner to corner and therefore I would conclude that as a land lens it is far superior. However, when placed behind a flat or dome port, things change dramatically. Neither 14-24 nor MP 14 can resolve the Epic sensor across the entire frame. The dome creates curved virtual image inches in front of the lens, which degrades performance to unacceptable (as in limiting) levels.

Nikonos 15 does not suffer from those limitations: it does not suffer from image plane curvature and doesn't distort or produce fuzzy edges at close focus range. It has been specifically designed to achieve maximum performance in water and at close working distances. It is the right tool for the job.

....Surely you must have been baffled by how something so antiquated and simple has managed to outperform even the efforts of better equipped engineers such as yourself when you built your Cineport, which prior to your Deep X concept, you touted as the sharpest, best dome ever built...

Again, you are incorrect. Nikonos 15mm is neither "antiquated" nor "simple" (12 element retrofocus design). How did you make these conclusions? The Nikonos 15mm is superbly engineered and very modern optical design that produces amazingly even, flat, sharp and undistorted images from corner to corner. It out-resolves the Epic 5k sensor in every point of the frame. The short focus performance of the Nikonos 15 in water is absolutely amazing and outperforms many (or even most) land lenses on land. The CA correction is amazing and measured 0.3 of a pixel in the corner.

However, what I do find "antiquated" is putting a land lens behind an underwater port and expecting 4k or 5k performance. It just doesn't happen. You can't even achieve SD quality across the entire image frame and claiming it to be "tack-sharp" seems to me greatly exaggerated if not completely incorrect.

 

[Will Keir]

I'm looking for a dive mask for underwater work, wondering what you guys are using?

I have been using the "Aqualung Maui Dive Mask (Black)" last summer and today; the mask just seems to hold up. It keeps a clear image, has a decent FOV and tends to fog less than my more expensive Atomic Aquatics Frameless. I bought this single lens frameless mask (double the price) because I thought I needed to see everything and in clear (not black) to add extra peripheral view but the mask is always fogging on me and the nose fills with water too often. I also bought it because it's a super light weight mask, which is nice.

Do any pros dive with clear masks? I thought this would be the way to go, but the dark frame seems to direct my focus forward where the actions is.

I must have a weird upper lip or something because most masks leak at the nose on me, either that or do I need to shave prior to every dive? I could install a nose purge valve but shouldn't I look for a mask that just fits right in the first place?

Most of my work is free diving in rivers, steams. Any recommendations? Thanks

 

[Ross Isaacs]

I use the Drager full face mask on my megalodon rebreather and it is amazing very comfortable.

 

[Rudi Herbert]

Pawel,

I should have know this argument would ensue, but first, let me apologize if I misled you to think that I need you to "believe" anything I say. I don't. A few answers for you.

Shooting charts and putting lenses under demonstrable and repeatable tests is a proven way to measure lenses against one another based on their exhibited sharpness, etc. Nothing wrong with that, it is the way in which many people choose their lenses, especially when testing several of them for a particular use that requires a particular feature of the lens to perform at a certain level. Great. But, there are many other people that do not bother doing this, who will trust the images the lenses produce as all the proof they need. This includes highly successful people in Hollywood who shoot with multi million dollar budgets and whose films gross even more than their budgets.

I arrived at my conclusions by looking at the projected images and then saying: "Oh wow, that looks tack sharp from corner to corner". That's it. Not scientific enough for you? Too bad. I'm entitled to my opinion, which is all I've posted here.

I've seen the Nikon 14-24 projected on a huge screen against the MP 14, two projectors at the same time, same image side by side, in one of the biggest rental houses in Europe, which has those two projectors set up for such tests. To my eyes, the Nikon looked noticeably sharper, especially at the corners, than the MP, from f4 and upwards. It was also the consensus among the rest of the people attending the demonstration. The fact that the Nikon does not open down to 1.2 does not mean that at the same aperture values it could not be sharper than the MP. It appeared sharper to all of us.

The Nikonos 15 is DEFINITELY an antiquated lens, regardless of how good, and I agree it produces outstanding images, it is even by today's standards. But, it was built using technology, methodology and materials that have been since, far surpassed in quality. Several generations of lens making have come and gone since. That is definition of antiquated. Just as a sports car built in the 1970's is still fun to drive and fast, when you compared it to the technologies available today, you realize what antiquated and modern means. Thus, the quality, precision and coatings of the glass used on the Nikon 14-24 is far superior to those employed in the Nikonos 15, as are the precision of the methods used for its fabrication. Fact.

But, how then does the Nikonos lens still look much better than most, if not all, water corrected lenses behind a dome? The simple answer as you pointed out is that it is a system built from the ground up to compensate for, or eliminate, the problems the underwater medium presents. But in so building this system, Nikon simply took a 20 mm air corrected lens and built around it to eliminate the aberrations that water introduces. The solutions to this problem, though ingenious, are simple:
- use a dome that will precisely align with the lens' entrance pupil, so that nodal point is equidistant from dome center and sides. This will produce a a properly aligned image, though with softer edges since the image projected by the dome is still curved.
- Build the dome with variable thickness glass, such that thicker edges provide a magnifying effect and help render the softer corners sharper so they match the center. To that effect, the center of the dome is much thinner.
- Introduce corrective elements in the lens design that will further correct the aberrations and curvature of image so that final result is rectilinear and sharp from edge to edge.

That's it, is not rocket science. I believe, after carefully listening to the engineers working on this project that the most important element of all is the variable thickness dome. Again, as I once told you, open up one of those 15 mm lenses and measure the thickness of the outermost element, a dome, and you will see that the edges are much thicker than the center. That is 90% of the Nikonos' magic. I have done this when I accidentally cracked one of my 15 mm against a reef and took a peek inside. In fact, I have also with me the front element of a Fathom SWP 44C, the one made for the Sony EX-1 that yields about 115 degrees FOV sharp from edge to edge, and I can see the same thing. I don't have calipers with me right now, but the edges of that element have to be at least 15-17 mm thick, while the center cannot be more than 3 mm. I keep that lens around to show customers why we don't rent expensive optics unless one of our operators goes with it (it was, again, cracked by an enthusiastic rental shooter on a wreck dive) but I will be happy to send it to you for perusal if you so desire.

I don't know what else I can tell you, I'm sure you'll have retort, but I won't. I was merely trying to share my excitement for something I've seen that seems to solve the main problem underwater encountered by a lens, the 14-24, that on land is possibly the very best rectilinear wide angle ever made. I know I am enthused with the images I'm seeing, as would be many other people on this forum that, like me, go more on feeling than on numbers.

Cheers,

 

[Tom Hamilton]

I want one of those domes :conehead:...

 

[Dave Blackham]

I'm looking for a dive mask for underwater work, wondering what you guys are using?

I have been using the "Aqualung Maui Dive Mask (Black)" last summer and today; the mask just seems to hold up. It keeps a clear image, has a decent FOV and tends to fog less than my more expensive Atomic Aquatics Frameless. I bought this single lens frameless mask (double the price) because I thought I needed to see everything and in clear (not black) to add extra peripheral view but the mask is always fogging on me and the nose fills with water too often. I also bought it because it's a super light weight mask, which is nice.

Do any pros dive with clear masks? I thought this would be the way to go, but the dark frame seems to direct my focus forward where the actions is.

I must have a weird upper lip or something because most masks leak at the nose on me, either that or do I need to shave prior to every dive? I could install a nose purge valve but shouldn't I look for a mask that just fits right in the first place?

Most of my work is free diving in rivers, steams. Any recommendations? Thanks

Diveways make a twin lens mask and you can specify the distance between lenses, its very low volume. Its a nice mask for camerawork. The main considerations are low volume and a mask that fits your face.