129 replies to this topic

[Alex_Mustard]

Thanks Alex. That's helpful. What's your preferred extension ring, same as you used previously?

Hey Jeff, these are with Subal FE2 port and ext ring. I haven't actually had time to go through all the test images, yet, once I do I let you know which ext ring dioptre combo is best. But actually my aim was to determine acceptable limits in various configs. I took close to 100 images (3 domes, 3 ext rings, 3 dioptre options, range of apertures means a lot of variables).

A very nice comparison of the 17-35mm lens with and without diopters. So interesting that I'll have to do a similar set of tests with my own Seacam domes.

I think it might also be interesting to see center sharpness comparisons of the same shots so we can see how much (if any) sharpness is lost to the center of the image by adding the diopter to the prime lens.

Hi Fred, I used 3 domes, but the shots above (and below) these were taken with the Subal FE2 dome - usually referred to as a 8" dome. Considerably smaller than Seacam superdome.

Here are the centre frame details from the same shots. All F5.6. The no dioptre case is perhaps marginally sharper - although it may just be a bit brighter (i shot on auto exposure in the failing light) - these are totally unprocessed crops from the NEFs. These images are the same scale (100%) as the ones above - and the size of the tiles shows the pincushion distortion from the dioptres (bigger tiles at edge of frame, the examples above). But the main observation is the drop off in sharpness at edge of frame with no dioptre (top image above vs top image here).







We have 2 Seacam shooters in our group here. I shot the Superdome and Canon 16-35mm Mk2 in the pool (no dioptre) earlier in the week. I don't have the files, but it seemed to be pretty good in the corners without a dioptre when stopped down. At F2.8 - even through the viewfinder the corners were clearly blurred.

I agree on the need to go shooting rather than worrying too much about these things - which is exactly what I intend to do now. The squid are waiting!

Alex

[craig]

...
And that is the experience on which I say, "Don't waste too much time on technical aspects."

Get in the water and shoot what is really important to you.

Fred

Some athletes are in it to get paid. Others for the love of the game.

[StephenFrink]

We have 2 Seacam shooters in our group here. I shot the Superdome and Canon 16-35mm Mk2 in the pool (no dioptre) earlier in the week. I don't have the files, but it seemed to be pretty good in the corners without a dioptre when stopped down. At F2.8 - even through the viewfinder the corners were clearly blurred.

Alex

Hi Alex - Not intending to highjack thread away from 14-24 but I was curious about the port extension used by your Seacam shooters for 16-35II and superdome. I assume they are shooting full frame Canons? If so, my tests suggested the sweet-spot was between 55 and 60mm. To that end, I asked Harald to create a PVL 57.5 (measured in MM). That's what I use for my 16-35II and have been very satisfied. Far more so than with either old style 16-35 or 17-40.

Speaking of evolutionary observations regarding port extensions, I've revised my recommendations on the 14mm II and superdome with full frame Canons to a PVL20. Details here: http://stephenfrink....n-progress.html

With both lenses my standard ISO is 200 in order to get enough light to allow me to work at F-8 and smaller in most scenarios.

This shot was pretty low light though ... F-6.3 at 1/25th second.

Here's the full shot:

Here's the corner detail:


If everything had been along a single plane, the corner performance would have looked better, but this is pretty typical of the kinds of set-ups we find on the reef.

Edited by StephenFrink, 28 January 2009 - 04:00 PM.

[divegypsy]

Alex,

Looking at the center sections of your three shots, as they appear on my laptop screen, the first center shot, the one where no diopter was used, looks sharper and more contrasty than the diopter shots to my eyes.

It may be that some center quality must be sacrificed when using a diopter to achieve the better edge quality. And if this is so, then it becomes a value judgement. A choice of which area you think is more important.

Do you remember how you focused the image. Was it using auto-focus or manual focus. And if it was auto-focus, which of the camera’s many auto-focus points did you use? This might also have a considerable affect on the final image.

As a general rule, I try to avoid adding any optics to an original lens in the belief that the people who designed the lens made it as sharp as they could within the financial limitations that were imposed. And that adding extra optics is always at a cost. This is also why most of the larger than 1:1 shooting I do is done with extension tubes rather than close-up lenses and tele-converters.

I also agree with your choice of lens to test. Although I own both the 17-35 and the 14-24, I feel the 17-35 is far more useful for general wide-angle shooting. And more versatile. I bought the 14-24 to test as a possible alternative for my fixed 14mm lenses, which I use primarily for over/under shooting.

Good luck with the squid.

Fred


Fred

[photovan]

Hi Alex - Not intending to highjack thread away from 14-24 but I was curious about the port extension used by your Seacam shooters for 16-35II and superdome. I assume they are shooting full frame Canons? If so, my tests suggested the sweet-spot was between 55 and 60mm. To that end, I asked Harald to create a PVL 57.5 (measured in MM). That's what I use for my 16-35II and have been very satisfied. .....

I'm using the EF 16-35II with 1dsmkii behind superdome with 60mm of extension with no diopter. I mentioned the other day when this subject came up in another thread, I tried a set of 82mm diopters(+1,+2,+3,+4), but was disappointed with all results, preferring the result from no diopter. Mind you they may be crap diopters, as they are a no-name brand but were the only 82mm ones I could find . I ended up at 60mm ext becausee I already had PVL35, and added a PVL25 to total 60 and so I could use the PVL25 with the 14mmII. Ouch ...wish I had of known 20mm would have been better for the 14mm....
BTW I'm expecting the 21.7MP of the 5dmkii (or the 1dsmkiii) to be more demanding than my 16.7 MP.

[Alex_Mustard]

Hi Alex - Not intending to highjack thread away from 14-24 but I was curious about the port extension used by your Seacam shooters for 16-35II and superdome. I assume they are shooting full frame Canons? If so, my tests suggested the sweet-spot was between 55 and 60mm. To that end, I asked Harald to create a PVL 57.5 (measured in MM). That's what I use for my 16-35II and have been very satisfied.
If everything had been along a single plane, the corner performance would have looked better, but this is pretty typical of the kinds of set-ups we find on the reef.

With so many of the FF Canon users shooting Seacam, I agree that the results I have seen from the 16-35mm Mk2 (no dioptre) and Superdome are highly acceptable and an important step on from the older lenses. It also makes sense that a big dome like the Seacam Superdome would require a long extension ring to keep the lens ('s nodal point) in the middle of the hemisphere - as Stephen and Darren are both attesting. It also makes sense that the size of the dome would be more forgiving than a smaller dome without a dioptre. If it was my system I'd certainly like to test the edge sharpness with and with dioptres, but I have heard how hard it is to get them in 82mm.

I also certainly agree on the importance of making evaluations on real subjects rather than the pool. I did these tests for myself, and only shared them here because they were relevant to this discussion.

Looking at the center sections of your three shots, as they appear on my laptop screen, the first center shot, the one where no diopter was used, looks sharper and more contrasty than the diopter shots to my eyes.
It may be that some center quality must be sacrificed when using a diopter to achieve the better edge quality. And if this is so, then it becomes a value judgement. A choice of which area you think is more important.
Do you remember how you focused the image. Was it using auto-focus or manual focus. And if it was auto-focus, which of the camera’s many auto-focus points did you use? This might also have a considerable affect on the final image.

The shots were auto-focused in failing light - and it is possible that my technique might have been wayward. The camera was on release priority, rather than focus priority. I had been in and out of the pool so many times changing ports/rings/dioptres I was not the model of care. But it would not surprise me if the no dioptre centres were sharper. I will look at some crops side by side when I have a chance.
Furthermore the pincushion distortion of the dioptres is another factor to add to the judgement. Particularly with the +4, which is considerably more severe than the +3.

The corner sharpness differences are much more easily seen - even in the thumbnails. With the smaller domes I tested the dioptre shots show an even bigger improvement over the no dioptre shots for corner sharpness (because the no diop shots are worse than the example above).

One reason for doing these tests was to evaluate if there was an aperture above which corner issues are acceptable without a dioptre. Before doing these tests I would not have considered diving with this lens without a dioptre. Now I might, mindful to stay above F14, ideally F16. This could be particularly useful if I was planning to shoot some split levels at the end of a dive, for example.

The irony of all these tests, is that I brought both the 17-35mm and 20mm here to Cayman to evaluate them before a 4 week trip to Indonesia in February. And my conclusion is that I can live without both. The nature of my photography means I can get away with shooting more fisheye - which just suits me more. Having tested the lot, I am actually going to take a 28-70 F2.8 Sigma. I have tested this behind all three domes (in the ocean) with +4 for reef life and +3 for big animals and I am very happy with how it is performing.

Alex

[Alex_Mustard]

To get back to the original topic. Karin should be doing her pool tests tomorrow.

[photovan]

.....
One reason for doing these tests was to evaluate if there was an aperture above which corner issues are acceptable without a dioptre. Before doing these tests I would not have considered diving with this lens without a dioptre. Now I might, mindful to stay above F14, ideally F16. This could be particularly useful if I was planning to shoot some split levels at the end of a dive, for example.....

This is the major reason I prefer to shoot dioptreless - flexibility. Many of the shoots I do are a bit frantic, and need the housed kit available to shoot under; just on top of; or half in and half out of the water.

Edited by photovan, 28 January 2009 - 08:49 PM.

[Paul Kay]

At the risk of creatin dissent, but in the spirit of being reasonabl:

there is a massive difference between utilising technical knowledge to optimise any photographic system, and being technically obsessive - and often there is a fine dividing line! I know professional photographers who have no interest in their equipment - whatsoever! I also know photographers who aren't happy unless they are using the very best available and understand every aspect of the photographic process they are working with. Since I've seen superb images from both camps I reckon that whatever works for them is OK!

Alex have you used any ACR chroma correction on the samples you posted? They do look as though they may benefit from some. IMO software correction is a very important aspect of digital photography!

Back to diopters - their primary function is to set reset the lens so that its infinity focus is actually on the virtual image produced by the dome of a subject at 'infinity' - no big technical deal here. However by using a diopter you are, as divegypsy says, adding in an additional optic and this has other influences too. There is a lot more theory behind what they do as well as re-adjusting focus. Understanding it will help produce 'better' files especially with the software tools now available.

I come from a technical/science background - I studied scientific photography and my prime interest is underwater natural history photography. As you may have seen I produced a report on UK gobies last year - this required working to exacting standards and then using software to enhance images in some cases - for this a good technical understanding is a pre-requisite.

And back on topic, we are back to our old problem - trying to photograph a curved, virtual image with a camera system designed to work with flat image fields - as ever the cut-off where corners start to become a problem is around about a 90 degree field of view (regardless of format). But the acceptability of corner softness will depend on many factors such as acceptable circle of confusion (itself dependent on final image usage and size) and original subject 'depth' within the area being images and.... lots more - so this 90 degrees is not a fixed figure, its an approximation. Understanding all this helps make an informed decision about such details as; where exactly to focus when taking the photo. Unfortunately most photographers, in my experience, shoot images for a 'generic' purpose - they don't know the use that will be made of them - and try to shoot for 'maximum quality', whatever that is - hence the potential obsession with equipment and technicalities. Pixel peeping is an inevitable consequence of digital imagery and has spawned a new type of 'photographer' - the 'photophile'.

But lastly, I do think that the image is actually what counts - in general a really good image is usable/saleable/whatever regardless of what it was shot on (other than very low MPixel, etc). I've said it before and no doubt will do so again, a 10MPixel plus dSLR camera exceeds the abilities a a 35mm film camera (I'm not explaining why here - suffice it to say that I can identify certain gobies from 10MPixel flies which I cannot from Velvia transparencies - the detail simply isn't available) but there are as few situations today where we actually need better quality than we got from 35mm - in general that is. For those photographers who genuinely do need better than 35mm 'quality' then a thorough understanding of the technical aspects of underwater photography is essential, and trying to extract the best out of lenses such as the 14~24 becomes a necessity requiring informed decisions as well as empirical testing.

[craig]

there is a massive difference between utilising technical knowledge to optimise any photographic system, and being technically obsessive - and often there is a fine dividing line!

Implied here is that being technically obsessive is bad. There are fine photographers who are "technically obsessive". We have good equipment because of the technically obsessive.

I know professional photographers who have no interest in their equipment - whatsoever! I also know photographers who aren't happy unless they are using the very best available and understand every aspect of the photographic process they are working with. Since I've seen superb images from both camps I reckon that whatever works for them is OK!

The question isn't whether a technically "ignorant" photographer is capable of excellent work, it's whether he would benefit more understanding than he has.

IMO software correction is a very important aspect of digital photography!

Agreed. Current tools all have liabilities though. ACR has good CA correction but is not the best converter for UW images IMO. I used to not believe that, but reading comments from those more technically obsessive than I am encouraged me to look more carefully into the matter.

Unfortunately most photographers, in my experience, shoot images for a 'generic' purpose - they don't know the use that will be made of them - and try to shoot for 'maximum quality', whatever that is - hence the potential obsession with equipment and technicalities.

Why is this unfortunate? Should all UW photographers have your goals in mind? If a photographer never intends to "use" an image in the manner you do, is that bad?

Pixel peeping is an inevitable consequence of digital imagery and has spawned a new type of 'photographer' - the 'photophile'.

No need to introduce derogatory terms for people who don't share your perspective.

For those photographers who genuinely do need better than 35mm 'quality' then a thorough understanding of the technical aspects of underwater photography is essential, and trying to extract the best out of lenses such as the 14~24 becomes a necessity requiring informed decisions as well as empirical testing.

I think you are hung up on the "need" part. You define quality and requirements arbitrarily from your point of view (and that's someone who sells his work). What I find curious isn't that you don't care about quality improvements beyond a certain point of usefulness, that I completely understand, it's that you find it appropriate to criticize those that do care.

Wetpixel has evolved to have similarities to other large photographic forums. There are photographers of a very high caliber that care only about the art and/or the sale. There are also those that care only about the nuts and bolts and there are some who excel at both. The nuts and bolts people never criticize the art people. In fact, they hope to learn from them. The art people, on the other hand, frequently criticize the pixel peepers, the measurebators, the technically obsessive. Why is that?

[Paul Kay]

Sorry Craig but I wasn't criticising or implying criticism of anyone, merely stating my own thoughts and trying to get across my belief that knowledge is important if you really want to understand how to achieve a specific goal. Audiophiles seem happy enough with their label.

[craig]

I know you weren't, Paul. I'm also aware that you know a great deal about the inner workings of the gear and what's important. I think the pro perspective is perhaps the most interesting one here. It's just not the only one.

I wish I could go to a forum and learn more about the art of the photograph that it turns out is possible. In my experience, being on the boat talking to an experienced photographer and diving with him teaches me more about the art of shooting than forums do. Equipment is only a small part of the equation; experienced photographers, travel agents, and dive guides arguably contribute more to excellent results.

[Paul Kay]

If you want to read a couple of interesting 'cross-over' books try "Visual concepts for photographers" by Stroebel-Hollis, Todd & Zakia and/or "Practical Composition in Photography" by Brück - both Focal Press. They combine art and technicality in fascinating ways! Not underwater but composition knows no bounds.

[loftus]

I wish I could go to a forum and learn more about the art of the photograph that it turns out is possible. In my experience, being on the boat talking to an experienced photographer and diving with him teaches me more about the art of shooting than forums do.

I agree completely with this; and I think it's particularly valuable to spend time with as many good photographers as possible. Topside and underwater; always important to have different ways to see things.
To get back to the technical part of things; I can't think of a single photographer in any genre, whose body of work I really admire who is not technically superb and / or obsessive. Some may emphasize certain aspects of the process more than others, but they're all obsessive, great quality consistently over time is no accident.

[tdpriest]

... there is a massive difference between utilising technical knowledge to optimise any photographic system, and being technically obsessive - and often there is a fine dividing line...

Despite your introductory statement, I agree with your post (which I interpreted as a plea for achieving a balance between science and experience); I also find it strange that no-one makes the technical knowledge v. seat-of-the-pants argument about marine biological expertise, since most experienced practitioners accept the need for both.



Tim



(PS: don't tell my patients, but my first doctorate was in lobster biology...)

Edited by tdpriest, 30 January 2009 - 08:01 AM.

[kaarlin]

Hi guys,

I'm very busy at the moment. Tonight or tomorrow I will upload my testresults!

Edited by kaarlin, 30 January 2009 - 02:13 PM.

[PRC]

If you want to read a couple of interesting 'cross-over' books try "Visual concepts for photographers" by Stroebel-Hollis, Todd & Zakia and/or "Practical Composition in Photography" by Brück - both Focal Press. They combine art and technicality in fascinating ways! Not underwater but composition knows no bounds.

Trust your judgement there Paul - just ordered both - to be fair the second does not represent much financial risk as it was £1.00 on Amazon 'used'.

Thanks,

Paul C

[kaarlin]

Yesterday I have been comparing the 14-24 mm lens with the 17-35 mm (both without diopters). I was surprised by the difference. I think the 14-24 has less unsharpness in the corners than the 17-35mm. I know that a 17-35mm performs better with a +4 diopter, but than I loose sharpness in the middle. And I will use this lens with people in swimmingpools and I think for me (read: for ME) that is more important than unsharpness in the corners. All pictures were made very close to the board (1 foot), so I think (don't know for sure) that the prestations would be better from a little bit more distance (I forgot to try that, can anyone agree this? or is this a misunderstanding?).

here are my results and decide yourself what you think about it!

(sorry, instead of F (of aperture) I used the A (of aperture....)

Edited by kaarlin, 30 January 2009 - 03:14 PM.

[loftus]

I disagree that there is significant loss of center sharpness with a dioptre (+2 or +4), at least on the 17-35. When I started using the 17-35 a dioptre was needed for consitent focusing behind the dome, corner sharpness was a welcome but secondary benefit. If the center is not sharp, it probably will not focus, at least if you are using the center as your focal point. The 17-35 may work without a diopter behind a dome, but I'm pretty convinced it works better with one.
Karin, I find your patterns difficult to read, something with more contrast would be better. Also the target does not appear to be held parallel to the sensor plane.
I think we all are hopeful that someone will get the 14-24 to work well underwater, I love this lens topside, but I think we have yet to see good images underwater, and no one has yet been able to demonstrate great results in my opinion (and the opinion of others more knowledgeable than I). The 17-35 is a workhorse for DX and FX in my opinion.
This uncropped pic was taken with the 17-35 +2 diopter, and I don't think sharpness, particularly in the center, is an issue. ( Taken at 30mm so it's not meant to demonstrate corner sharpness.)

Edited by loftus, 30 January 2009 - 04:06 PM.

[kaarlin]

Hi loftus

I agree that this aren't the best pictures but I wanted the subject to be flat so the whole subject is the same distance away from the camera. For me it is visible that I can work with this lens for my poolshots. I doubt if this lens is usable on a reef. Alex told me that another option is to use the sigma fisheye with 1.5 teleconverter. Unfortunatly I don't have these lenses so I couldn't compare this one