Intro to underwater photography

Hi, this is George. I run Seapix.com, a site devoted to all aspects of underwater photography. I've just completed an overview of digital photography, especially with respect to underwater photography. Here's the link: http://www.seapix.com/digoverview1.htm. If you're interested, take a peek.

I've been taking pictures for more than 20 years, and have never lost my love of underwater photography. I'm also a contributing writer/photographer for www.ocean.com. It's wonderful to see the advances in underwater photography. Within the last two years the digital revolution has made underwater photography accessible to most divers. And, naturally, kudo's to wetpixel. It's nice to read actual reports of equipment from USERS, rather than the manufacturer's PR department.

Safe diving to all...

Hi George,

I enjoyed reading your intro to digital.

One question (I'm sure I'll have more soon): In your film-to-digital comparison, did you compare a film scan to a shot from your Sony F707?

If so, your statement would apply:

Their increased sensitivity not only limits their ability to reproduce subtle color variations, but it virtually guarantees "blowout" when shooting towards a bright light source.

However, this is incorrect for digital SLR cameras which have much larger photosites and therefore a much higher dynamic range. Meaning, they can capture high contrast scenes - and can actually capture MORE detail in shadow areas when shooting RAW and doing a 1 or 2 stop bracket in software.

Moving on to the next page I especially took exception to your film-to-digital comparison shots. While it's entirely possible to get the "blown out" shot on the right - I also find it entirely possible (using my S2) to get the pleasing "benchmark" film shot on the left.

Cheers

James

In digital vs. film, I take exception to points 3, 4, 5 and 6.

3. Compact. If you compare digital P&S cameras to film SLR's, then sure, but digital P&S's are similar in size to film, esp. APS, P&S cameras. Film cameras can, and do, have integrated zoom lenses. If the discussion is underwater, how many housed digital cameras are smaller than a Nikonos V? Are you arguing that their lens solution is superior to the Nikonos V?

4. Extreme light sensitivity. Why don't you put high speed film in your film camera? Are you saying that an ISO 100 equivalent digital camera is a lot faster than an ISO 100 speed film? Then put in ISO 400 film. The 5050 is ISO 64---not fast at all. Just because the highest numerical aperture is f/8 doesn't mean the camera is fast. It means you can't stop it down. That a disadvantage, not an advantage.

5. Enormous Depth of Field. This is a myth. You can't compare a digital camera and its smallest aperture of f/8 to film by restricting the film camera to f/8. When operated in an equivalent manner in macro, digital P&S are the same or worse than film. In wide angle it's much better, but who cares? Digital SLR's are much more like film that P&S.

6. High-speed strobe syncing. Do you honestly have trouble freezing the action underwater at 1/125? I suspect there are film cameras with totally adequate strobe sync.

This paragraph is also full of inaccuracies:

As light streams into the lens, the image data is interpolated  (a technical term for "guesstimating.")  During the image capture process, algorithms actually create some image data that in reality is not there.  The data is an estimation based on a best-guess scenario of the surrounding pixels.   In essence, nearby pixels are sampled, averaged, and then incorporating as new pixels in the overall image.

This allows the camera to gather more digital information than is actually there. Hopefully this translates into better resolution.

Interpolation is not guesstimating, and interpolation is itself a poor description of what goes on. Luminance, for example is not guesstimated at all, and at lower than native resolutions, no guesstimated is needed at all. How can you "gather" more than is actually there, and how can guessing increase resolution?

Regarding JPEG being lossy, it should be pointed out that all means of capturing images are lossy. The digitizing process is lossy, the storing of the image is (potentially) lossy, recording an image on film is lossy. One form of lossiness does not make a format inferior. Look at DVD. Prior to it's introduction, some predicted it would never be as good as Laserdisc. They ignored the fact that Laserdisc was a perfect copy of the world's most lossy video format while the DVD is sourced from potentially the highest quality master possible. Now such a suggestion is laughable.

I disagree also in whole or in part with all four claimed shortcomings of digital. The author seems to be unaware that digital SLR's exist when he states that lens choice is poor. Digital cameras have trouble dynamic range, but making blanket statements like digital cameras can't produce saturated color is absurd. What film are you comparing it to? If Velvia is the benchmark, well Velvia lies. That's why people like it so much. Digital cameras may be susceptable to blowout, but so is film, and what is the deal with shutter lag? When you shoot film, don't you prefocus? If so, then why don't you with digital? If you prefocus most digital cameras, shutter lag is not much of an issue. I've owned 7 digital cameras starting with the Nikon 950 and I can honestly say I've never ever noticed shutter lag. Of course I always prefocus.

It is true that digital struggles with highlights and is not as forgiving as film in certain ways. It is also true that digital's great strengths are convenience and instant review. The rest is generalization and myth that may or may not hold up under scrutiny.

Wow. Either I'm far more ignorant about digital cameras than I thought (a distinct possibility) or there are some confusing discussions.

The whole section on pixels interpolating confused me. Maybe I don't get the exact point. Pixels, i.e. photosites gather the light, period. Maybe you are talking about the Bayer Interpolation process.

There is confusion over light sensitivity vs. dynamic range, as James and Craig pointed out and also saturation.

The issue is not light sensitivity, but dynamic range, and this is greater for consumer cameras vs. DSLRs. A properly exposed scene with a wide dynamic range will have blowouts, true. But the reason is the limit of range. You can get rid of the blow out by reducing the exposure value. The blow out or highligts will be the same as film, but due to the limited dynamic range, the rest of the scene will appear underexposed.

As far as starbursts go, a key factor in digicams is the inability to stop down past about f/8. Even with a film camera, the bursts require stopping down, underwater or on land. You can try this on land quite easily.

Color rendition and saturation. "In spite of the increasing sophistication of digital cameras, they still cannot compete with film cameras for rich color rendition. Though colors may be vivid, they are noticeably less saturated; subtle gradations in shadow detail and color are lost. The images below demonstrates this. Both shots were taken in bright sunlight using a circular polarizer."

Saturation and shadow detail are a function of the film, not the camera. Saturation and shadow detail vary significantly from one film to another, so you can't make a global statement. Each film has its own response curve, emphasis on curve. Velvia provides enhanced saturation, but lesser shadow detail than other films. That's part of the curve. It is a terrible film for dark scenes, but it does provide "rich" blacks. As I said, all films are different. Digital cameras have no inherent lack of saturation. The sensor just picks up electrical charges and records them as 0 and 1's. From there the data is intrepeted into color. (Bayer interpolation, digital to analog conversion). The saturation and color rendering is controlled by the firmware, with some control often placed with the user. The data can be interpreted any way you like. The reponse "curve" is a bit more linear than most films. That's why we use Photoshop curves or some people use custom tone curves for their cameras.

I do agree (sorry Craig), that digicams have a real negative when it comes to shutter lag. Regardless of pre-focus, the nature of the CCD requires it to do more before the exposure, therefore creating some amount of lag.

Something that really bugs me is your two sample shots showing "blowout". They are identical, except for the exposure. Are you conning us? Same exact shot? Did you carry two cameras? Notice the exhaust bubbles are identical? What are the odds. Sure the second one marked digital shows more blowout, but the exposure value is greater. Just compare the blue water. Truth time.

At first I was ready to take issue with Craig about the DOF increase being a myth. But I concede it is a bit of a myth. General underwater photography with a digicam happens to fit the parameters of greater DOF vs. film. Smaller CCD/frame will create greater DOF to a point, all things being equal. But the truth is that you can get as good a depth a field with a larger sensor/film frame under certain circumstances (all things do not have to be equal) whereas it is equally difficult, sometimes impossible, to reduce your DOF when you want to with a digicam. Otherwise, you would never see a large format photo with any DOF. (hint, stop down to F/32 with no diffraction, something you can't do with a digicam).

Hi again George,

I hope you don't read these comments as attacks. I really enjoyed your article and I'm sure you put a lot of work into it. Craig, myself, and the other George all have experience with high-end digital SLR cameras and they behave a lot differently than the small sensor point and shoot digicams.

To make an apples to apples comparison, it would be fair to compare an F100 to a Nikon D100 or Fuji S2.

It would also be fair to compare an Ikelite Auto35 or MMII to a point and shoot Digicam.

It's not fair, IMO to compare a Nikon F100 with a 105mm f2.8 to a point and shoot digicam and declare the F100 the winner.

Cheers

James

Part of the shutter lag problem is determining exposure (and strobe). If you rely on the camera to do everything after you press the shutter, then lag can be a problem. If you shoot all-manual, it's similar to film. Digital takes longer after the shot, of course, so it's frame rates are harder to make fast. It's all in how you use it, and perceived differences come from the cameras being used differently. Not denying that shutter lag exists, just saying that you need to keep in mind how they are used. A big difference to me with digital P&S's underwater is that, as a group, they don't do manual focus well. Since autofocus can be slow, there can be a big lag. I sure noticed that with the 5050. This is obviously not characteristic of digital and my D100 does not do this. Interestingly, only in digital P&S's is the shutter button defined to mean "take the picture right no matter how long it takes". It simply controls shutter in other cameras.

DOF is also a perception problem. DOF is controlled by the physical aperture, whereas we think of the f-number as being aperture. F-numbers are normalized with respect to focal length, a good thing because it lets us calculate exposures readily, but normalized f-numbers are NOT what you need to discuss DOF. A P&S f/8 has a physical aperture identical to f/32 on film. When shooting wide angle in ambient light you end up comparing like f-numbers (so that exposure is correct) and the digital ends up with more DOF. Why? Because its physical aperture is much smaller. More often than not, as scorpio_fish said, the increased DOF is not an advantage.

The sample shots bugged me, too. The shot taken was out of the film gallery. The digital version was a deliberately corrupted version of the film shot. The point was to demonstrate a problem with digital, but the sample was faked. I think it's true that you want to avoid blowing out highlights in an image. Other than that, was the sample teaching us anything?

Good points on film. Not all films are made to be accurate! Digital sensors, on the other hand, need to be. You simulate the effect of Velvia (if you want) by applying curves after the shot. Fred Miranda will even sell you action for that purpose.

Yipee! another film vs digi debate!

Seriously, I have to chime in on this one as well. I think George wrote an excellent detailed site on UW photog, which is very relevant to new or novice u/w photographers My congrats to you!

That said, I agree that it is unfair to compare an f-100 in a blanket statement to a digicam. Digital SLR's with the exception of the handling of highlights and wide angles, are IMO every bit as good as, and much easier than film set-ups today. (And George, i recently switched back to film). The advantages of having 100-300 images per dive and instant review are huge.

Others in here are much more tech-savvy about digital than I, so I leave the merits of those arguments to them.

In truth, they both are excellent in their own right, they are just different. Either system is capable of getting stunning results, put in the hands of a photographer who knows what he is doing. As digital reaches full frame technology at a reasonable price, I will no doubt switch back someday. The beauty of a histogram and preview underwater is a huge advantage for the non-professional hobbiest. Lets face it, for the average UW diver who dives 1-2 times a year, spending 5k or more to get a good digital slr set up is a difficult pill to swallow.

Until then.....LONG LIVE VELVIA!

Psst, Bob, need some Velvia? I got a bunch in the fridge.

year, spending 5k or more to get a good digital slr set up is a difficult pill to swallow.

The sheer dollar amount certainly is, but if you look at the cost / good image you come back from those trips with it changes things a bit. I'd do the math but I'm just too tired right this moment (mental note, head to Dunkin Donuts), but I suspect that the cost / image would be lower with digital, and that you'd come home with a higher percentage of your shots as keepers.

Just my twisted way of rationalizing how much money I spent on my D60 setup...

Darn!!!! I just bought all my film......its the only piece of equipment I didn't advertise on line for her.

Next time! Thanks

and I am still laughing about the rationalization statement ts. I tried that on my spouse, it did not work.

This is getting quite tiresome - film photographers writing columns with the pretense of being objective, but without the information or experience shooting digital to make an objective comparison. Here are my major issues with this column(in addition to the ones already mentioned):

1) The writer discusses "RAW" files as uncompressed files by definition - I think he means TIFF files, because he says the RAW file from his F717 is 16MB - I don't think so... Such a blatant mistake tells me that the author has marginal experience with the ins and outs of digital photography.

2) On the one hand, he uses identical images to show the blowout highlights, then he photographs totally different views (and possibly different days) to demonstrate the loss of saturation in the rock/water/skyline photos. Hardly objective.

Here lies the problem with the internet, where anyone can pretend to be an authority in any field. I appreciate the differences between film and digital, and we all know the drawbacks of our cameras. I would love to see a truly objective article with side by side comparisons between digicams, dSLRs, and SLRs - same scene and angle of view, and written by someone with a thorough understanding of all three.

I think the Backscatter series of articles has come closest to this.

Sorry if this is a little brutal, but I call it like I see it :ph34r:

Kasey

Not trying to bash, just confused as to what digital equipment geodiver is using. On the seapix website, George talks about his Sony F717 and F707 cameras. In his equipment reviews, he show the Amphibico Surveyor for the F717 "which he has used extensively". OK, here's where I'm confused. The Surveyor for the F707 was designed for the F707 (manual buttons on lens have diff config) and per Joe (Amphibico) who I talked with last week, the Surveyor for the F717 will "not" be available for another 2 weeks. Now I guess one can throw the F717 into the SurveyorF707...I can't say from experience if it would fit.

So my question to geodiver is "what equipment was used for the digital pictures?"

Wow, I didn't think this would generate so many comments. As for the comment that "this is becoming tiresome" I couldn't agree more. My article was a simple overview of digital vs film. It wasn't meant to be the bible of digital photography.

It certainly wasn't meant to be dissected mercilessly by technocrats. Many would-be photographers spend their time discussing the technical details instead of doing what a photographer is supposed to do: get the image! For those of you who obsess over details, and couldn't take a picture without first doing a detailed histographic analysis, you have a different approach to photography than I do.

I don't profess to be an authority on photography, though my work has been published in several national magazines, and my image galleries are featured at Rodale's Dive Magazine and Ocean.Com. I do claim to know something about underwater photography, so to answer some of your questions...

1. I know the difference between a tiff and raw image file. The point (which I mistakenly thought was obvious) was that raw images are larger than compressed images (and, yes, Kasie, I do know the difference between jpg, tif, and raw). And, no, Kasie, the Sony doesn't take raw images. I merely used uncompressed size to demonstrate how large a raw file would be.

2. Yes, my Sony is compact. I used to shoot with a Nikon F4S, two wide angles, a 90mm macro, and telephoto. My equipment bag was large and HEAVY. Now I tote around the Sony, extra memory sticks and battery in a small carry bag. So to restate the obvious: yes, it is compact. And to forestall the obvious comeback: No, it doesn't do everything my old system did. There's a trade-off. That was the point of paragraph (and included picture, I might add)

3. Yes, the Sony is more light sensitive than film cameras. Again to forestall the obvious: you could use high-speed film, and even push process it, but the "gritty high-contrast" school of photography doesn't appeal to me --I'll leave that people who photograph addicts in alleyways at night and spend the rest of their time basking in self glory for revealing the seedy underbelly of life. Personally, I prefer Nature's beauty as a subject matter. But again, I was just demonstrating a general principle: I can shoot easily shoot in low-light situations with Sony without having to resort to tripods or high-speed films.

4. Which equipment used for underwater photography --my apologies if there's any confusion...

I used the Amphibico Surveyor housing and a Sony F707. Since then, I have switched to a Sony F717. There were a few problems with the previous housing design, plus there were incompatiblity issues with several common strobes, so I'm holding off until I receive the new replacement housing before writing a review. I felt I should wait for the new housing instead of writing a review for housing that has already been replaced. As far as I know, both the F707 and F717 will work in the new housing. The controls are accessed via a "LANC cable" --a type of miniplug that mates with the remote input jack of the Sony. Hopefully, that should answer your question, Mellingsong

Last, thanks for your comments, good and bad. In the future I'll be more careful about what I post here. And don't throw your film away. I like Fuji film, and my switch to digital was a bit too early. As much as I love the benefits of digital, there are still some aspects of my old Nikon system that I miss.

George (aka geodiver)

George -

Please don't be scared away! Most people on this forum only mean well, even if they can at times be a bit harsh.

Many members here are extremely knowledgeable about the technical aspects -- and real-world repercussions -- of digital photography. I myself have been frustrated with just about every article I've seen where a "pro" talks about underwater digital. I've taught a bit to some of the "pros" who have written these articles, and there were always some fundamental things that they didn't understand. (I'm not lumping you into that category, by the way.) But more and more pros are really becoming experts in digital as they "catch up" with those of us who have nothing better to do than study the technical aspects of it, so I expect articles to improve in the future. Articles written for the general public naturally have to be dumbed down. You can't please everyone.

my 2 cents,

-Eric

Eric:

Thanks for the words --you're right about not pleasing everyone; it's part and parcel of a discussion group. You're also right about differing views can be harsh --but by the same token, forums are a great way to get information that doesn't come from the manufacturer's PR department. That's getting harder and harder to do these days; I actually started the review section for that reason: I've spent thousands of dollars on uw camera equipment that didn't work as promised.

Anyhow: don't worry, I haven't been scared away; I'll just be extremely cautious in what I say. And wetpixel is a great source of information.

Incidentally, I forgot to reply about film scans...When I compared images (digital vs film) I'm talking about digital vs fujichrome 50 or 100 (my 2 favorites). I use a Nikon coolscan to scan the slides, and I do it about 3400 x 2200 pixels (true color) with a resulting 25 meg bitmap image size.

Last: To bobjarman, tshephard,echeng, scorpio fish, craig and jamesw: Thanks for your comments and I just want to add:

1) I agree about Velvia, great stuff

2) I regret selling all my Nikon bodies and lenses (you can berate me for that --it was an epic piece of stupidity :-). Wish I could afford both a great digital and a film system.

3) Agreed: A good underwater system is obscenely expensive these days

4) Like MacArthur, I will return...

5) I concur with the apples and oranges comparison. The review wasn't meant to be in depth comparison of all digital equipment, merely an overview. So if you want to correct this oversite and "lend" me a D100 and housing and strobe please feel free to. I promise to test it rigorously. In fact, I'll continue to test it for years if need be...

So having said all that, let me put myself into the firing line again. Here's the review I wrote about the LX25 high intensity lights that I used in conjunction with the Surveyor housing and Sony F707. In fact, you'll see the housing and setup arms and lights in the article. Here it is:

http://www.seapix.com/lx251.htm

Safe diving...

George

I am well aware that my comments were harsh - I even added a discclaimer

The mention of 16MB raw files in the article is incorrect beyond being toned down for the average reader. In fact, RAW files will always be much smaller than TIFFs, because they don't carry nearly as much info. They can also be compressed in a lossless fashion to be even smaller - the D100 does this in camera, but it is sloowwww. If your sony did do RAW files, they would weigh in at around 9MB.

Also - you have not commented on Craig's observation that the "digital" shot was a deliberately corrupted film photo. Assuming he is correct, how could you possibly defend that?

Kasey

Oh yeah - If you can make a trip to the USVI, you are welcome to my D100/Aquatica for the weekend!

The mention of 16MB raw files in the article is incorrect beyond being toned down for the average reader.  In fact, RAW files will always be much smaller than TIFFs, because they don't carry nearly as much info.  They can also be compressed in a lossless fashion to be even smaller - the D100 does this in camera, but it is sloowwww.  If your sony did do RAW files, they would weigh in at around 9MB.

I've heard many seasoned photographers make this same mistake. Canon D60s are claimed to have "18 MB raw files," when in fact, they do not. One photographer I know on the lecture circuit even says that the D60 shoots "18 MB files" in JPG mode because they happen to be that large when they are opened uncompressed.

But don't some consumer cameras that shoot in RAW mode store images as standard TIFFs (as opposed to the proprietary TIFFs that Canon 1D and 1Ds cameras shoot?). I seem to remember coming across that at some point in the past.

I gave you the benefit of the doubt and went back to re-read the article. Here is the confusing quote:

[Note the trade off between image quality and file size.   To overcome this drawback many new digital cameras have a camera setting that allows the photographer to record the images in "raw" format.  The "raw" format denotes the fact that no alterations to the original image have been made. While this may seem desirable initially, the drawback becomes immediately obvious: raw images require enormous amounts of storage space. As an example my Sony DSC-F717 stores approximately 60 jpeg images at its highest setting on a single128-megayte memory card.  These images provide a near-perfect compromise between image quality and file size.  In a uncompressed (raw) format, each image would be approximately 15 megabytes in size.  In other words, I could only get about 8 images on the same memory card (or only 1 image with the included 16-meg memory card that comes with the camera! )  ]

Every reference to raw in this paragraph really refers to uncompressed TIFF files. This isn't one mistake, clearly it must me a misunderstanding of what a RAW file really is - and since your camera doesn't offer the feature, it is an understandable confusion. Perhaps the intro to digital uw photography should be renamed as "into to digital uw photography with the sony f707." Had that been the title you probably wouldve gotten no criticism from wetpixelians. On my 1gb microdrive, for example, I can store over 100 raw files in full rez from my D100- so this "inadequacy" is redered obsolete.

Technocrats - I never heard that term before. I like it!

One of the great things about this forum is that, unlike dpreview and others, most of the contributors are active and productive photographers. Just look at the galleries. It is difficult to succeed with digital photography without a thorough understanding of what the camera "thinks," and clearly your article is an attempt to educate people on the technical advantages and disadvantages of digital. It seems proper, then, that our replies are focused on the technical misinformation put forth in the article.

This isn't a personal attack, as I'm sure you are a very accomplished photographer. Many of us have been shooting digital for years, however, and there may be some things you can learn from us. For starters, any one of us would've told you to hang on to your F4 for wide angle work