Underwater Bubble Blowin' Users Group

[David Nardini]

David: I know this was a reply to GATES' post, but I'm assuming it is also addressed to the design community at large.

I was speaking to Ken Corben today and he asked about the possibility of having the housing long enough to support the longest configurations we might need now with REDDRIVE and REDBRICK capability as the current version does, but then being to able shorten the housing later as CF becomes more feasible. I see no problem with this, estimated cost would be between $500-$1000 conversion fee.

Regarding ports, we have already done this. We are using an off-the-shelf Aquatica port system with an internal diameter just under six inches which allows Ken to use the Canon EF-S 10-22mm, as well as the Arri/Zeiss Ultraprime 8R - which is probably the ultimate underwater lens. This system has 6" and 8" dome ports, flat and macro options, extension rings, etc. - AVAILABLE TODAY - which would let you use pretty much any lens out there - cine or still type. Further, since the frontplate is removable we can pretty easily accommodate other lenses that might exceed that diameter.

The interchangeable frontplate would also allow use of a fully corrected underwater optic such as the one Amphibico used on their F900 housing. (Does anyone know if that is a FATHOM design or one of their own?) That type of optic generally needs to be closely matched to a specific lens so we need to see what shakes out in terms of lens options. With that said, it should also be noted that domes are still the mainstay for professional housings for film and video from Pace, Hydroflex, AquaVideo, etc. - because in fact they work very well.

Mike : Absolutely ... comments are for the community at large !

Thanks for the feedback. I'll PM you later in the week ...

As and when you get nearer your final design (I understand that it WILL evolve over time due to feedback, requirements, etc but it would be good to know what are the 'basic/starting point' controls you will provide (in the PRO and non-PRO housings). Focus control ? Aperture control ? What RED controls ?

RED LCD housing : would the external housing for this be such that it is best to dedicate an LCD unit to live in the housing ? or not ? could you share thoughts on this ?

Without having a RED in my hands at the moment, I'd like to get a sense of likely 'heat' issues. I would be very happy with the non-pro version, but the PRO version (being ALU) should be better for keeping things cooler.
Your tests to date involved CF's, and therefore re-surfacing, opening the housing, etc ... and potentially never reaching a 'heat' issue (camera heat generation + drive heat + battery heat ... cannot imagine this lot being a 'cool pack' as a whole !). Perhaps there is an alternative reason for having a tad bigger housing !

Cheers ...

 

[Michael Hastings]

More info:

More info:

Mike : Absolutely ... comments are for the community at large !

Thanks for the feedback. I'll PM you later in the week ...

As and when you get nearer your final design (I understand that it WILL evolve over time due to feedback, requirements, etc but it would be good to know what are the 'basic/starting point' controls you will provide (in the PRO and non-PRO housings). Focus control ? Aperture control ? What RED controls ?

RED LCD housing : would the external housing for this be such that it is best to dedicate an LCD unit to live in the housing ? or not ? could you share thoughts on this ?

Without having a RED in my hands at the moment, I'd like to get a sense of likely 'heat' issues. I would be very happy with the non-pro version, but the PRO version (being ALU) should be better for keeping things cooler.
Your tests to date involved CF's, and therefore re-surfacing, opening the housing, etc ... and potentially never reaching a 'heat' issue (camera heat generation + drive heat + battery heat ... cannot imagine this lot being a 'cool pack' as a whole !). Perhaps there is an alternative reason for having a tad bigger housing !

Cheers ...

David:

A couple of quick comments. I'm building an LCD housing today as Bill Summers was kind enough to loan me his. I will cover it in a later post and ask for some suggestions about closure preferences, but basically I don't think you will need to dedicate an LCD - it will be a pretty quick changeout.

As far as controls, you will definitely have power, start/stop and the two user buttons up at the front left of the camera. Focus and iris will be through control shafts through the housing attached to the Birger mount controls - we should be able to mount them wherever we want. Zoom will be via a simple friction control until we see what RED and/or Birger come up with for motorized zoom (already in the works).

Beyond that, extensive camera setup control is most likely to end up using some sort of remote CCU type interface as we used to do on Betacams, F900, Varicam, etc. but we don't know exactly when that will be available. We can hit any of the buttons on the rear LCD from the top and side but until we know the real functions of those it is hard to know what makes sense - I have mentioned that we will provide 1-3 controls for those buttons as part of the base cost - we consider frame rates, shutter angle, and possible ISO settings as the most important, so would try to implement those. White balance and other settings are really unnecessary with the RAW paradigm and are probably better left to post.

Right now we are leaving a few inches behind the camera for battery and drive options, however in the future we could make an aluminum backplate with pushbutton controls like some of those on Amphibico or Light and Motion. The way RED made the rear buttons means it will be relatively easy to do but would probably mean shortening the housing - as I mentioned in a previous post shortening the housing is a fairly low cost conversion of the longer housings. Again, you really need to think through the raw paradigm - I think waiting for an electronic remote for those functions - if they are even necessary for most people - will be the way to go.

Heat is probably not a big issue with the aluminum housing.

On the PVC it may be an issue (We have experience with the HVX200/firestore combination - which gets very hot - in a PVC housing and we were able to work for an hour or so in the Florida keys without a shutdown - we don't have experience with longer times. We did feel that if necessary we could have used a chemical coldpack.)

The first solution if the RED builds up to much heat in the PVC housing would be to go to the aluminum frontplate and I think this would provide a large percentage of heat transfer as the full aluminum housing (a couple of fluted heat sinks could give added transfer as well).

As far as gears for focus, iris, zoom controls for cine lenses, that will have to be done on a case by case basis and one should expect it to cost several thousand dollars, unless one of the rumored low cost Russian lenses (or if Jim decides to make a RED 10mm) becomes so much of an answer for us that most users would use it. This is one reason the Birger mount and still lenses is such a big deal for us bubble blowers. However, a final thought. Lenses like the Zeiss 8R (the ultimate underwater lens) could probably be handled very well with low cost friction controls if desired because basically you preset focus and with the RED raw capability you don't have to ride iris as much either.

 

[David Nardini]

RED housing

RED housing

David:

A couple of quick comments. I'm building an LCD housing today as Bill Summers was kind enough to loan me his. I will cover it in a later post and ask for some suggestions about closure preferences, but basically I don't think you will need to dedicate an LCD - it will be a pretty quick changeout.

As far as controls, you will definitely have power, start/stop and the two user buttons up at the front left of the camera. Focus and iris will be through control shafts through the housing attached to the Birger mount controls - we should be able to mount them wherever we want. Zoom will be via a simple friction control until we see what RED and/or Birger come up with for motorized zoom (already in the works).

Beyond that, extensive camera setup control is most likely to end up using some sort of remote CCU type interface as we used to do on Betacams, F900, Varicam, etc. but we don't know exactly when that will be available. We can hit any of the buttons on the rear LCD from the top and side but until we know the real functions of those it is hard to know what makes sense - I have mentioned that we will provide 1-3 controls for those buttons as part of the base cost - we consider frame rates, shutter angle, and possible ISO settings as the most important, so would try to implement those. White balance and other settings are really unnecessary with the RAW paradigm and are probably better left to post.

Right now we are leaving a few inches behind the camera for battery and drive options, however in the future we could make an aluminum backplate with pushbutton controls like some of those on Amphibico or Light and Motion. The way RED made the rear buttons means it will be relatively easy to do but would probably mean shortening the housing - as I mentioned in a previous post shortening the housing is a fairly low cost conversion of the longer housings. Again, you really need to think through the raw paradigm - I think waiting for an electronic remote for those functions - if they are even necessary for most people - will be the way to go.

Heat is probably not a big issue with the aluminum housing.

On the PVC it may be an issue (We have experience with the HVX200/firestore combination - which gets very hot - in a PVC housing and we were able to work for an hour or so in the Florida keys without a shutdown - we don't have experience with longer times. We did feel that if necessary we could have used a chemical coldpack.)

The first solution if the RED builds up to much heat in the PVC housing would be to go to the aluminum frontplate and I think this would provide a large percentage of heat transfer as the full aluminum housing (a couple of fluted heat sinks could give added transfer as well).

As far as gears for focus, iris, zoom controls for cine lenses, that will have to be done on a case by case basis and one should expect it to cost several thousand dollars, unless one of the rumored low cost Russian lenses (or if Jim decides to make a RED 10mm) becomes so much of an answer for us that most users would use it. This is one reason the Birger mount and still lenses is such a big deal for us bubble blowers. However, a final thought. Lenses like the Zeiss 8R (the ultimate underwater lens) could probably be handled very well with low cost friction controls if desired because basically you preset focus and with the RED raw capability you don't have to ride iris as much either.

Mike, thanks for the update.

Good to know the LCD housing is 'taking shape'.

Agreed about the RAW capability, but I would like to think WB can be set (I gather the two buttons at the front are use programmable, so this should not be an issue (I'm thinking of experimenting with colour correcting gels/filters; yes, with 12bit RAW, there's room to play, but helping it along is surely not going to make matters worse).
Once the RED s/w gives us real-time 'levels' feedback (eg : bottom of LCD ?) it should be simpler to keep the data in the correct (sweet) spot !

The Birger mount will make lens control significantly simpler ... so fingers crossed Erik posts some information soon ;-)
With regards to focusing, sure, anything around the 10mm, or less, at 5.6 or above will need little to no focusing.

Agreed about the PVC / ALU options ... let us know turn around time for the ALU housing once you've finalised your design / tooling up. I'm OK with a longer housing to start with pending larger CF or cheaper RAM drives ;-)

I'm looking for a housing for 2 key 2008 trips (Mar/Apr - Thailand) and (Jun/Jul - Red Sea) ...

Cheers,

PS : I'll be sticking to Canon glass ... cannot see the Zeiss output being significantly better (once placed behind a dome, etc) - certainly not a multiple equivalent to it's cost differential !
Oh, and yes, it would be rather nice for RED to give us a 8 or 10mm / 2.8 prime (with close focus ability like the 18-50, which focuses down to a couple of inches ... would be perfect for sticking behind a good quality dome). Then again, maybe the Russians will help us underwater folk out ...

 

[GATES]

Deep Red

Deep Red

Thank for all posted information, it's appreciated. To answer some questions:

* What are your plans for LCD monitoring ? Housing the RED unit ?
-- Either the RED LCD in it's own shell or Gates EM43. Both have tradeoffs (size, resolution, etc).

* Integrated battery ? or using a RED Brick ?
-- RED Brick

* Planning on housing electronics driving the RED (if this is possible) or driving the RED more via more traditional housing controls ?
--It will be classic Gates approach with mechanical controls. There are three important notes on this:
1) Electro-mechanical controls are required when using a Canon/Nikon lens with the Birger Mount. This means a control will penetrate the housing and mate with the Birger encoder.
2) Using mechanical controls one might ask how we can use fully manual lenses (such as various primes). Herein lies one of the powerful features of the design from Gates (and with due credit to Howard Hall). I won't delve into the boring details, but suffice to say that the zoom/iris/focus controls will remain in the same location on the housing -- at your fingertips -- regardless of lens employed. You won't have to re-learn or reach for a control to change these functions with different lenses.
3) Lastly, Howard and I are in agreement that the user will not require access to the rear panel controls inside the housing. At RDCC we learned that functions accessed on the rear are setup in nature and not necessary to change underwater. We're not even convinced you will need to see the display as salient information is provided on the LCD. Of course this could change as RDCC is constantly updating the software, so stay tuned.

Thanks again for the comments. Keep 'em coming.

Cheers,

John

 

[Michael Hastings]

Lens control on UW RED - details

Lens control on UW RED - details

I won't delve into the boring details, but suffice to say that the zoom/iris/focus controls will remain in the same location on the housing -- at your fingertips -- regardless of lens employed. You won't have to re-learn or reach for a control to change these functions with different lenses.

Ditto, on that. Re: the details: Basically - at least for us - this involves use of motor drives for those functions. Since 1985 we've built housings for 14 different broadcast cameras from from DXC3000s and Betacams, to the Varicam and F900 - all using Canon or Fujinon lenses with built-in motors for iris and zoom and stock add-on motors for focus. These are all driven via pots or servo controls via a penetrating shaft. Therefore I have almost no experience with lens gears, hence my comment in the earlier post that mechanical gear drives could be done but expect to pay thousands. Since all of the cine primes have built in gear teeth on the iris and focus, the use of small geared motors to drive these seems like the way to go. (I've always wondered why the housings for 35mm Arris and such had manual gears.) As far as specifics, I'm just waiting a little bit to see what Birger ends up with for their zoom control (which will be this kind of motor with a gear on the lens, which will have some method of physically attaching to the camera. Remember iris and focus are driven by the internal motors, but zoom is physical rotating of the zoom ring). It would be nice to be able to standardize on a physical mount like Birger's so that it could be used on either type of lens. This will be a separate item, so could be used whether or not you get the Birger Canon EOS or Nikon mount. If this doesn't work out we will build our own - we're already doing some preliminaries on this for the Zeiss 8R.

The same control shaft penetrations used for the Birger encoders would be used for the switches/pot/servo for the cine lenses. You would just put in a small box for the drives instead of the birger encoders - hence the same control positions on the housing for either Birger or cines.

Since they are pretty smalll, we should be able to drive these motors through the DC outputs on the camera.

 

[WesG]

I've been digging around the net but without any luck and thought this group might have some tips.

I'm going to be building a housing for my RED but want to be able to feed out vision from the camera as it is shooting underwater (only shallow water use like being in a pool).

I need some sort of HD-SDI or HDMI waterproof plug and cable connector that could be glassed into the housing.
I've seen some Sony Z1 housings that have a waterproof feed point but I'm not sure if it's DV, HDV, SDI or HDMI.

Any advice, experience or tips?????

Thanks!

 

[SalaTar]

BK-7 optical glass.
Subcon power & control.
D.G. Obrien single mode fiber optics.

 

[GATES]

HD-SDI on Red

HD-SDI on Red

I need some sort of HD-SDI or HDMI waterproof plug and cable connector that could be glassed into the housing.
I've seen some Sony Z1 housings that have a waterproof feed point but I'm not sure if it's DV, HDV, SDI or HDMI.

Hello Nalu,

We've successfully transmitted HD-SDI out of the Gates XL H1 housing over 100 feet the surface. This feature is planned for DEEP RED as an option.

J-

 

[WesG]

Hello Nalu,

We've successfully transmitted HD-SDI out of the Gates XL H1 housing over 100 feet the surface. This feature is planned for DEEP RED as an option.

J-

That's great. Could you point me the right direction for parts or you don't sell the parts yourself???

:wink:

 

[GATES]

HD-SDI on Red

HD-SDI on Red

In this case we don't sell the parts individually. It's very much a custom installation for a housing/camera combination. Further, some of the components are made for Gates specifically to support the very high data rates of HD-SDI.

J-

 

[Michael Hastings]

Try ikelite

Try ikelite

That's great. Could you point me the right direction for parts or you don't sell the parts yourself???

:wink:

Gates and I are manufacturers ourselves so a little hesitant to help you design your own housing, but with that said Ikelite makes a compression cable seal for 1/4" cable and, depending on your depth needs and if you are careful to avoid cutting the cable, you can use it with a good quality HD-SDI rated RG-59 type cable.

 

[Michael Hastings]

AquaVideo pricing/description post from other thread.

AquaVideo pricing/description post from other thread.

When Gates posted his info I said that I would copy this post from the Gibby's RED takes a dive...Underwater RED 4K thread after a day or two and then I forgot. If you want to see some pictures, videos, impressions, etc. check out that thread.

http://www.reduser.net/forum/showthread.php?t=4908

Since the original announcement 2 weeks ago we have been moving forward and now have the LCD case proto'd, just waiting for the connector pinout from RED, and the special size aluminum is on its way, and are moving forward on Sharky's and a couple others that have already been ordered, for delivery in about a month.

Here's the announcement post:

Anyway, thanks to Ken who has been making exceptional efforts to get the RED in the water, and to Gibby for trusting us with #8, and everybody else on the team.

I think Ken has been hesitant to detail things before talking to me but I will try to lay things out for you, he has pictures too, which I hereby give my permission to post if he was waiting for that.

First, I am amazed at the images they achieved, but I will also tell you that they are nowhere near what the RED is capable of simply because they were forced by time constraints to use the 18-50 zoom. It is an excellent lens, but nowhere near as wide as we would normally use underwater. The 18 gives a little over 70 degrees angle of view - where we are used to shooting at about 105 degrees with Varicams, F900s, better HDV cameras, etc.. The equivalent lens would be a 10mm on the RED. The main difference is that everything would appear much clearer since they would be shooting at almost half the distance - a huge difference underwater.

The housing was actually setup for the 10-22 Canon EF-S lens on the Birger EOS mount. We had the physical prototype last week but Erik at Birger is doing some final tweaks and is about another week or so from having a working proto for us to use in the housing.

A lot of the questions have to do with ports and lenses and rightly so since it is all about the optics, and it has never been more so given the new paradigm of shooting RAW with RED. With that in mind we are making the housings with total flexibility in terms of ports and lens adaptability. The standard interchangeable port system we are using is very flexible, but the second part is that the front plate is removable so even if someone comes up with a funky lens setup that we can't accommodate with the standard ports we can easily, and fairly inexpensively, make a new front plate to specifically accommodate that lens. For instance, I doubt if there is much point to a Fathom type optic since we have interchangeable lenses and excellent superwide choices from Canon, Zeiss, etc. (the Fathom has been more about converting fixed mount lenses) but it would be pretty easy to accommodate simply with a different front plate.

Initially we are using a series of interchangeable ports from Aquatica. The front plate has the mating part for the Aquatica screw type ports. We are using the screw type because it provides a much larger hole diameter than any of the off the shelf bayonet style. The large diameter is necessary to accommodate the larger cine style lenses and particularly the Zeiss Ultraprime 8R which is probably the ultimate underwater prime lens (price: US$25,000). We think most will use the EF-S 10-22 (or the nikon equivalent) on the Birger mount, but we wanted to make sure that everyone could also easily use the 8R for super high quality projects. Both six inch and 8 inch domes are available as well as extension rings, macro ports, etc. which allow use of virtually any lens I have seen.

Currently the domes are acrylic, and should be excellent for most users. (The vast majority of high end underwater motion pictures and stills that you have ever seen were shot through acrylic domes - including IMAX ( I know because I have built replacement domes for some of them.) However, Aquatica recently introduced a 9" glass dome for their bayonet style which should be adaptable to the screw mount. I also have other sources for 8" and 6" glass domes as well if necessary.

Re: housing pricing and specs. Two things: 1) Depth rating is primarily determined by the front plate/port. Either PVC or Aluminum body can do 300 feet or more. You wackos that are pushing 400 or more need the aluminum. 2) Pricing is sort of a la carte. Partly because many users will be using it in the surf line and/or less than 130 feet and partly because we don't know the exact configuration of things like the RED drive, the cost of LCD and power cables, etc.

The housings, whether aluminum or pvc will be 9" OD and our prototype was about 21.5" long plus port with a displacement of about 50 lbs. (displacement is the amount it has to weigh to be neutral in the water. It is the same for aluminum or pvc but the pvc needs a 12 pound weight to be neutral, the aluminum just 3 or 4. (If you are using lights they would offset the need for weights.)

Given the shape of the camera and the position of the viewfinder connectors there is no size advantage to a rectangular shaped housing. If Ken posts pictures the one straight on seems to show a lot of space around it but it is a lens/perspective thing - it is actually pretty tight (and the pvc has about .20 extra on the ID.

Pricing: Tentatively, with the usual disclaimers: $3999 will get you the PVC body with acrylic front and back plate, Aquatica screw type interchangeable port mount with 6" diameter dome standard. We rate this configuration to 175 feet. Thicker acrylic or aluminum frontplate give optional greater depths (see below for price and rating for other plates.)

Standard controls are Power, trigger, and magic focus (i.e. the two small buttons on the front left of the camera). It will be set up to use the Birger mount with control shafts to the Birger focus and iris controls. We may end up with an option for incorporating separate birger controls/cables into the housing so they are semipermanent and you don't have to use the regular knobs that comes with the control. (est. 300-400 each control for the encoder and cable). We will include a simple friction type, works but not super slick control for zoom on the 10-22 canon. Smoother control will be available once we get to evaluate the Birger and other zoom motors.

We will probably include one or two controls on either housing for the menu buttons on the rear LCD but it has not been determined what of those will be possible/desired. The aluminum housing will have a flat on the top over the LCD control area and allow more functions to be accessed, up to the point where the spaghetti factor takes over. We feel fairly confident that at some not too distant point, RED will make or allow much more sophisticated camera remote control through the USB interfaces.

However, it is important to remember that we will be shooting RAW so most of the adjustment controls you have been used to in the past will not be used - i.e. white balance, gamma/knee, picture profiles and the like. This will be more to allow shutter angle, iso, shutter speed, etc. to be changed underwater.

The aluminum body housing will be $6200 with the same functions and the same acrylic plates depth rated to 175ft. Add $900 for aluminum frontplate which jumps you to 250 foot depth rating, and if you add $150 additional for a 1.25" backplate increase depth rating to 300+ feet (I will work the numbers further - don't know the real rating of the dome ports. With the proper ports and an aluminum backplate, 1000 foot or greater ratings would be possible - again with the same main body housing. With the same endplates the PVC housing would have the same depth ratings, up to 300 feet. Much beyond 300, prudence would dictate using the aluminum body even though the PVC should be able to go much deeper as well.

The camera mounts on a slide in tray - currently either front or back load depending on user preference and lens used (wouldn't want to scratch a $25000 Zeiss 8r sliding it in from the back and hitting a control). We will provide a mount at the rear of the camera tray for the RED LCD. It is so slim that it adds almost no additional length even if you choose to use the LCD externally instead. We will also offer an external LCD housing for the RED LCD priced at approx. $999 (it is being prototyped now) plus connectors/cable (to be determined - could be from $300-700 more.)

In keeping with the RED tradition of naming prototypes after cities we named it - what else? - ATLANTIS. While our normal trademark color is yellow - we have made a special exception and the PVC housing will be RED. The aluminum will be the standard Navy spec Black hardcoat that we do on all of our broadcast housings - with undetermined label colors. We can provide a white or yellow material to put over the black hardcoat for those that feel it necessary to avoid heat buildup.

REDDrives were not available but there is room behind the camera for the drive if we use a separate battery (about $400) along the left side of the housing. If both battery and drive need to go behind the camera it would add about 2" to the length.

I'm sure I've forgotten something important but we can add to it as we go along. Just wanted people to get some ball parks. That is: $3999 for PVC Surf/dive to 175 ft. housing. $7100 for aluminum with aluminum frontplate base rated to 250 with low cost upgrade to over 300 feet.

Add ~$400 for special battery. Add $1500 or so for external LCD (monitor not included - take your red LCD in and out or buy a second - it is the best you can get for anywhere near the size, clarity and money.), and maybe $600-$800 for permanent mounted focus/iris encoders/cables.

 

[WesG]

Gates and I are manufacturers ourselves so a little hesitant to help you design your own housing, but with that said Ikelite makes a compression cable seal for 1/4" cable and, depending on your depth needs and if you are careful to avoid cutting the cable, you can use it with a good quality HD-SDI rated RG-59 type cable.

Totally understand and thanks for the tip. I'll be building a one off shallow water housing and then hoping once the camera is earning some dollars for me I can look to purchase one of your professional rigs.
cheers

 

[chuck colburn]

Although I think one would be crazy to put their RED in a home built housing, (I would do it but then I'm nuts) here are some domes that MIGHT work.

http://www.surplusshed.com/search.cfm

Type in domes in the search box.

There are more under misc.

 

[Mark Thorpe]

Personally I cringe at the thought of a non professional construction for an underwater housing. What would the heartache be if one floods the camera / housing due to construction errors? Let the pro's do their work and let them protect (hopefully) your investment with their knowledge.

Just my 2c
Cheers,
Mark.

 

[WesG]

Fear not CamDiver it's a professional job for my builder who's built housings for me before - he's a gun. Not made rated for 300ft but more for shallow water work.

Thanks for the tip Chuck but I can't find anything like I need there - if you want "1920's Swiss Army Snow Goggles" though it's the place.

 

[Mark Thorpe]

nalu, kudos for going with your builder. Best of luck, be sure to let us know how it goes.

Cheers,
Mark.

 

[Michael Hastings]

Re: Cineport from other thread

Re: Cineport from other thread

Pawel Achtel posted an announcement about his Cineport in another thread called Red is Going Under...

In order to try to keep this thread as complete as possible I have copied this response here - and will also put the in-depth response here later this weekend.

Mike

Pawel:

Welcome to the party. The only thing you said that I totally agree with is that for closeups with longer focal length lenses it is not hard to achieve good results with a flat port.

From your website description it does not appear that you have any sort of aspheric or multi element design - it appears to simply be a slice of a 6" radius/12" diameter dome. That is fine, but be honest about the tradeoffs you are making.

I will post an in depth response and explanation of the issues later this weekend because it will take some time and right now I need to finish a housing for the Harbor Branch Oceanographic Institute.

In the meantime it would useful for you to confirm:

1) the optical principle of your port, 2) the position i.e. distance from the surface of the dome that you will be placing the nodal point of the camera lens, 3) how you will compensate for differing lengths of different lenses and 4) explain how the images obtained by all of the 35mm major motion pictures with underwater shots, Imax movies, thousands of even higher resolution stills (since SLR size is substantially larger and higher resolution than RED 4K) from Stephen Frink, Rick Frehsee (RIP - who was a personal friend who gave me a very rare copy McNeil's book), Chris Newbert and many others - seem to be very sharp, yet all were shot with six and 8" domes, mostly acrylic (and many by Aquatica).
__________________
Red One, EVF, LCD, Basic Prod. Pack, Red CF, 2x Red Drives, Birger Canon mount+ff, Canon 10-22mm, 24-70mm f/2.8 L, 70-200mm f/2.8 L IS. Zeiss PL avail. RED#206 Available for rent in November. Underwater housing available now. SAVE THE SHARKS.

 

[Michael Hastings]

Underwater optics 101 part one - copy from other thread, might find useful.

Underwater optics 101 part one - copy from other thread, might find useful.

First, before getting into the actual optics, this Cineport is based on the Seacam thread/bayonet mount which has an opening of about 4 inches which would be okay for still lenses like the Canon 10-22 but means you simply couldn't use the larger cine lenses, like the Zeiss 8R, which are physically too large to fit into the port. This is why we chose to start with the older screw type aquatica mount as it has almost a 6" hole which lets us use the 8R, most other cine lenses, as well as still lenses like the Canon 10-22. IMHO this size limitation of the cineport would be a major deal killer right from the start.

Second:

Since the claim is for 99.7% it means that both sides of the dome have to have the best of the best in multi coating If coating is required on the outside surface I wonder how it will hold up in salt water.

AR and anti-scratch coatings are available on Aquatica's 9" glass megadome. Chuck if you could also chime in on how much loss -in F stops - would worst case 90% transmission equal, as we generally haven't seen that as a huge problem in the past, even with acrylic ports, and should be even less so with high sensitivity REDs. (One of the big questions we have had is best way to do neutral density, because we know we will need it.)

As a practical point regarding his superduper, only a few places can do it coating, - although I often hear claims of superior this and that - the truth is all underwater manufacturers are microbusinesses and if something is available to one it is probably available to all. There are no Sony's or Jim Jannard billionaires in the dive industry. (Standard joke goes: Q: How do you make a small fortune in the dive industry? A. Start with a big one.)

One of the main things Pawel is talking about is one of the basics of dome ports - they all (including his cineport) create a curved image. In other words the camera sees a curved world kind of like one of those curved trade show displays (or an Imax screen) where the center is at a greater distance than the edges to a plane going through the front of the lens. The center of the trade show display would be at say 10 feet but the edges would be at 8 feet, so if you have a lens that is designed to focus on a flat field - a tele lens (say 135mm) at a large aperture (f2.8) and focused at 10 feet, the edges would be out of focus.

His point is that with the larger radius/smaller curvature of the Cineport this curvature is less. The reason this isn't a big deal is that underwater you are trying to use as wide a lens as possible (and as you will see in a minute the whole reason for dome ports is for wide angle lenses). The depth of field of wide angle lenses is so great that it is all in focus anyway.

You can check depth of field calculations at:
http://www.nikonians.org/html/resources/guides/dof/hyperfocal2.html
I used the calculations for the Nikon D2x because it has the same size sensor, with 12 MP and 5.5micron photosites - as the RED.

You will see that for an 8mm lens, even at f2.8, subjects between 1.8 ft to infinity are all in focus. At f5.6 everything from 11 inches to infinity is in focus, at f8 8" to infinity, and even more for higher f-stops. The numbers are very similar for the 10mm. So you can see that the curved field isn't a problem with the Superwides we are likely to use.

It is true that the curvature is less with the larger radius of the Cineport, which might be more of an issue if we were stuck with say the 18-50 RED zoom as our widest lens, but in fact we have excellent superwides available and at reasonable cost (the Canon 10 is only $800). There are other tradeoffs with the cineport that he isn't telling you about and that is the point of this:

Underwater optics 101

First of all, any diver knows that a flat port (i.e. his dive mask) creates a 25% magnification so with a camera your 10mm becomes a 12.5mm - not the direction we want to go. The first effect of a dome port is it eliminates this magnification. Underwater this is a major deal and one reason why we use domes even on inexpensive SD camcorders where the other optical correction isn't as important.

I have posted an image where you can see the difference in how light rays strike the lens through a flat or dome port. As you can see with the flat port the light rays get refracted and converge (focus) at different places depending on how far off axis the ray comes in. In addition, since different colors get refracted differently you get chromatic aberrations as well. The situation gets worse the wider angle the lens. Hence, for macro shots with longer tele lenses we use a flat port. With wide angles we start to look for corrective solutions - generally the most practical solution is the dome port. (The other big solution is the Ivanoff corrector - subject for U/W optics 201 if we ever get there - suffice to say it is more complex, substantially more expensive and has to be fitted to the particular lens - probably only a practical option for us if RED or the Russians come up with a super high quality, low cost, superwide 10mm or less.)

If you look at the diagram with the dome port you see that all of the light rays come straight through the port (more correctly: perpendicular to the tangent of the dome) so don't get distorted AS LONG AS THE FIRST ELEMENT OF THE LENS IS AT THE CENTER POINT OF THE DOME. This is the part Pawel is not telling you about the Cineport. To illustrate: for the most common 3" radius (6" dia) dome the lens needs to be 3" behind the dome for a 4"R it has to be 4" behind, etc. and to use a full dome the domes get physically huge if you go much larger. (The largest common sizes are 8 or 9"). You can use a slice of the dome but for the 105 or more degree angle we want to use you can't slice it much UNLESS YOU WANT TO CHEAT BY MOVING THE LENS FORWARD INTO THE DOME. This allows you to use a smaller slice of dome, but your light rays no longer converge on the proper point - NOW YOU GET ABERRATIONS AT THE FIRST ELEMENT OF THE LENS and THIS IS THE TRADEOFF IN THE CINEPORT. Without getting into virtual images yet, his spec:

For object distance 5m, the focusing distance is 771mm.

suggests a radius of 10" or more. (I know there are some things that could alter this, but optical calculations give me a headache too, and it is pretty close for illustrative purposes.) This means the lens would have to be 10" behind the port to converge properly. Obviously he is not doing this since, with a 100+ degree lens in an 8" diameter housing, if the lens were inset that much, half of your picture would be the inside of the housing.

The way he deals with this is to use a very small "slice" (see the line across the dome of my second picture) and moves the lens very close - ACCEPTING THE ABERRATIONS THAT CREATES. Why do this? It has to do with the focus shift/virtual image that the dome creates.

The idea of virtual image tends to confuse people but the basic idea is that when the dome contacts water it becomes a negative diopter lens so to the camera it appears that the whole world has been compressed to within 3x the radius of the dome in front of the dome. In other words focus has shifted thus:

an object actually 4 feet in front of the dome appears to the camera to be at:
9" in front of the dome with a 4" radius dome
14" in front of the dome with a 10" radius dome

an object actually at infinity in front of the dome (or call it 100 feet for practical purposes) appears to the camera to be at:

12" in front of the dome with a 4" radius dome
30" in front of the dome with a 10" radius dome

The easiest thing to remember is that infinity is always 3 times the radius in front of the dome and everything else is progressively closer. So if your lens can't focus down to that 3R distance everything would be out of focus.

Remember that you add in the distance the lens is behind the dome to get the focus distance, so for example with our 8"/4" radius dome:

an object at 4' appears to camera 9" in front of the dome + 4 inches = 13" focus setting on lens.

an object at infinity (or 100ft. practical) appears to camera 12" in front of + 4" or 16" focus setting.

with pavels 20"/10" radius dome:

Remember that you add in the distance the lens is behind, but in this case it is less than full radius probably about 4".

an object at 4' appears to camera 14" in front of the dome + 4 inches = 18" focus setting on lens.

an object at infinity (or 100ft. practical) appears to camera 30" in front of dome + 4" or 34" focus setting.

It isn't critical whether I have guessed right as far as Pawel's distance behind his dome because I agree with him that he has achieved focus distances that are "well within most lenses close focus capability".

ADMITTED: The ADVANTAGE OF PAWEL'S LENS IS THAT ALMOST ALL LENSES CAN ACHIEVE ACCEPTABLE FOCUS WITHOUT DIOPTERS OR SHIMS.

The Zeiss 8R (probably the ultimate underwater lens right now) focuses to 12". So wide open at f2.8 it would give focus from 1.9' to infinity; at f8 1' to infinity; at f11 9" to infinity, so still very workable with an 8" dome even without diopter or shim.

But here is my problem with it, THERE IS NO WAY TO GET RID OF THE ABERRATIONS FROM MOVING THE LENS FORWARD IN THE DOME, yet with the smaller dome we can maintain the proper placement AND WE CAN PERFECTLY COMPENSATE FOR REFOCUSING VIA A SIMPLE SHIM of a known value.

This is why we have elected to use a full dome WITH PROPER PLACEMENT. (For now - nothing stops us from using a larger radius slice in the future - lower curves are actually easier to get made and it would simply be a change in port - you could go from one to the other. IMHO the best use of this type of large radius slice would be for split water shots - but that's another story.)

10K character limit so - THE END

 

[Ken Corben]

Mike,

The previous post took at least 45 minutes away from finishing the RED pro housing - priorities buddy

Well done - thanks for the optics 101 post. I received a bit of email traffic query on this one.

Sharky