Welcome to our community

Be a part of something great, join today!

Register Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

The Gentleman's Underwater Bubble Blower Thread

I once shot a Jackie Chan movie and the camera assistant handed me the Arri 3 in housing and I submerged and the housing instantly flooded as the camera assistant closed the housing on a cable.....300 grand down the drain
 
We're all guilty your honor. Ironically, the older and more experienced I've gotten, I don't see a decrease in these incidents, must be the epidemic of ADD affecting everybody nowadays....

So Pawel, your solution to the Epic problems is to use a whole different sensor altogether, not making it a solution at all? humm, yeah, maybe Jim will invest in your project. FWIW, I have a feeling that Pawel's sensor design will be something completely new nobody else has ever thought of, but CCDs were always much better at color accuracy than CMOS. By a lot. And there are 4K CCDs out there beeing produced, this one is based on the original Kodak deisng which is actually 4K x 4K, anamorphic heaven unleashed: http://www.fairchildimaging.com/files/ccd485datasheetrevc.pdf and then, I think Dalsa still makes them as well. The F23 and the Genesis, both based on the same CCD, made incredible underwater images, even if only 1080p. I'll see if I can find some tests I shot with the F35 and share them with you all, incredible colors in deep water, without lights. At any rate, it seems that two of the most important things for RED are resolution and frame rates, and the only way to achieve high count of both will be with a CMOS, but while CMOS make more than adequate colors on land, I am missing the older technologies and their superior colors underwater...
 
Rudi, Different sensor design is the best, but not only, solution to the problem. It can be applied for either CMOS or CCD sensors and has very generic benefits of much cleaner, lower noise in blue and green channels, which is a weakness of all silicon sensors, both CMOS and CCD. The solution I have at present, while still complex, can be applied to any sensor, including existing MX sensor.

The 2/3" 3-CCD cameras made nice underwater pictures for many reasons, the main was that the image plane curvature was significantly lower behind a dome and therefore they produced much higher contrast and detailed images than larger 4k sensors behind a similarly sized dome port. The colour science can be overcome, but image plane curvature can not (unles you ditch the port).

I'm in the process of testing and refining our solution. I will post images when they beocome available.
 
Hugh Miller is a member of the forum here and he is listed as part of the crew, so he must know and can tell you if he ever gets the chance to pop by...
 
Hugh is a very experienced u/w shooter in extreme conditions, I'm sure his team had all their ducks correctly lined up. Shit happens.
 
Don't often agree with Pawel ;-) but he's probably right - the glass front was probably a large flat plate and with glass this can be very dangerous as you have to be sure that there are absolutely no shear forces as the pressure increases as while glass is very strong under compression it doesn't handle shear well at all. This is a lot harder than it seems to achieve because you need to have some sort of pressure plate to put pressure on the glass up against the o-ring seal to make the initial seal. It is very difficult to do this perfectly evenly - the typical method is to have screws every couple inches through a metal flange but even there it isn't hard to have some uneven pressure. Any minor flex can cause a crack/catastrophic failure. In this case plastic is much more forgiving because it has some flex.

Just one of the bugaboos of shooting 3D underwater - but it was easy everybody would be doing it.
 
Well, there's a lot of speculation and guessing going on here and none would apply to the actual incident. I think everyone is jumping the gun here. Pawel you are making a guess first of all. But as far as i know...in the previous shoot for Galapagos, they never shot 3d underwater, but only 2d converted because of the cost and complexities. So since we are all wildly speculating and if you read the article and pay attention to wording there is no reference to cameras with an "s", but only camera. So it just might be that what actually happened is as was written....and they had a custom housing that imploded and the camera flooded....simple as that.

I could be wrong, but have a feeling they may not have shot a beamsplitter underwater for the same reasons they did not on the last shoot....which were complexity & cost.
 
Michael Hastings said:
Just one of the bugaboos of shooting 3D underwater - but it was easy everybody would be doing it.
Click to expand...

I fully appreciate the difficulties, Its like we chose to go to the moon kinda thing.. Insane, Bold, Awesome and just cool :)

Just was saying 10m is only 14.7psi thats pretty damn iffy to have something blow like that. However as Tom said shit happens.
 
Apparently they had custom made housing, which

1. They have never calculate the stress or strength of the housing
2. They have never tested it without the camera
3. They took it to Antarctica without any prior testing
4. The housing weight was over 500kg for Neutron/Quasar 3D rig and 2xR1!

and

5. Johnny, it is very educated guess based on specific information that I have :)
6. I agree with Michael (but, only on this occasion!) LOL
7. Tom and Greg, I believe that, in this case, the shit happened simply because they had a bunch of drongos ligned up correctly
 
Pawel Achtel said:
Apparently they had custom made housing, which

1. They have never calculate the stress or strength of the housing
2. They have never tested it without the camera
3. They took it to Antarctica without any prior testing
4. The housing weight was over 500kg for Neutron/Quasar 3D rig and 2xR1!

and

5. Johnny, it is very educated guess based on specific information that I have :)
6. I agree with Michael (but, only on this occasion!) LOL
7. Tom and Greg, I believe that, in this case, the shit happened simply because they had a bunch of drongos ligned up correctly
Click to expand...

I must say I have thought twice about replying to this thread. In the past I was so appalled at what occasionally passed for debate here that I deleted most of my posts and moved on. However, given the speculation that is going on I feel I need to respond. Firstly I should say that that shoot ended over two years ago and a lot has changed in that time. Epic for a start, and 2D and 3D housings have progressed too, as have the rigs that support the cameras. I did a shoot last year where I logged over 120hrs in the water with the Gates Deep Atom and I will say its everything I wanted in a 3D system. Most of those 'wants' were formulated on that penguin shoot.

EDIT: The above paragraph is not clear enough in stating that there is NO CONNECTION IN ANYWAY (and that includes me, I had nothing to do with the design of either housing, I just happened to have been an operator on both) between the 3D housing used on the Penguin shoot (a custom rapidly produced housing made in the UK) and any Gates product (such as the Gates Deep Atom). I mentioned the Gates Deep Atom in the above paragraph only as a counter point to my prior experiences with 3D housings!
End of Edit

Firstly to respond to Mr. Achtel's post
1) Not true. The ports were strong enough as was the housing (would it be a huge surprise to you that these things had been calculated?) It was the smaller port that blew from a failure in machining, that small concussion then blew the larger port.
2) There were no cameras in the housing at the time. The details in the article are not entirely accurate.
3) There was testing before it went south, not much, given the time frame, but there was testing. It was a calculated risk, in the 'either go and try and maybe fail or dont do it at all' category. SG is sub-antarctic, to be strictly accurate.
4) I dont know how you come up with this figure. Firstly the Quasar is about three times the size of the Neutron so they cant really be put in the same weight category. Secondly although R1's were the primary topside camera (on a Quasar), in the underwater housing SI2K's were used on a Neutron beam splitter with a Cinedeck. A Red One one in a Gates was also used in the latter stages, alongside the 3D housing.

5) It may be educated but clearly not well informed and not from anyone who had any first hand knowledge of the shoot.
6) About time.
7) Well, I've likely been called worse by people who are far more important to me than you. I will say this though, its a small world, especially in underwater film making, and making derogatory statements like that on a public forum are silly at best and hardly ingratiates you with working professionals (who are potential customers, right?).

Would I go and repeat that shoot now with that equipment? No.
Within months almost all of that equipment, bar the lenses (which included zooms), was obsolete. If that shoot was repeated almost all the problems would have been mitigated.



In more general terms I would say with the greatest respect to everyone involved in that article, that its a piece of publicity (albeit industry publicity) and publicity pieces should perhaps not be relied upon for technical detail. What makes a good story two years after the event isn't necessarily exactly how it happened. I'm used to that. 'BTS' and 'Making Of's' are all interpretations put together many months later (often with the involvement of people who have a good eye for a story but were not present in the field) and again are not generally blow by blow accounts, its the same kind of thing.

The fact that any kind of housing existed given the time frame within which it had to be built was a miracle. I was the operator, I wasn't part of the build but I will defend those who were. It had many flaws but it did gather material in the end. Given more time those flaws would have been worked out. Now with Deep Atom its no longer necessary.

It was repaired back in the UK during a planned shooting break. No, it wasnt dropped off by a passing ship, that was something else, water filters or something.

For me it was just one shoot among many successful ones (from Planet Earth, Life, Frozen Planet, Africa, Wild Arabia through to current projects like Survival) but I admit it was one of the hardest in terms of dealing with the high failure rates. The shoot was very long and very remote, so its a challenge and you have to be resourceful and roll with the punches. Failure sucks, it really hurts, but it is part of the deal of trying new things.

[As an aside the top-side crew had many of their own battles to deal with. I struggle to remember a piece of kit that did not fail or had to be taken apart and dried carefully over a oil heater. SG is sub-antarctic, when its not snowing is very wet, very muddy. The only things that did not fail were the Red One's.]

I will always respect people who try something new, whether it works or not, they tried. Its easy to criticise with all the benefits of distance, time, hindsight, new tech etc but unless you were there, its difficult to do so constructively or with much authority.
If something fails you change what your doing and try again. If you keep doing the same thing expecting different results then you can be legitimately called a drongo (to paraphrase), but not before.
Don't forget that things change very quickly. Back in 2009 I filmed sailfish for 'Life' with a high-speed housing, again custom built housing for the shoot and very hard to use, 3x two second shots, 45min downloads and an external battery pack mounted on my back, but it worked and we got shots that revealed (or at least confirmed) new behaviour and went on to be opening shots in the series. I could do the same thing now with my Deep Epic with all the shot and battery time I could want, the point is, you use what you have available and if its not available then build it, be inventive. This is a creative industry after all, if it works great and if it doesnt then adapt and change, but at the very least try.

Hugh
 
You must log in or register to reply here.
Back
Top