Nikon 14-24 unsharpness?

Anyway, further to Alex's post - is the quality of the edge with the +4 diopter similar to the quality of the same part of the image without the diopter ? In otherwords, is the fact that the +4 diopter has a zoom in effect the major contributing factor as to why the corner sharpness is better - it just avoided the corners ?

Looking at Stephen's samples (the whole frame - as there is no of the area of interest (as we didn't know it would be the area of interest)), I'd say so. At the same angle of coverage they are probably similar in sharpness.

So the downside of using the dioptre is reduced view. The downside of no dioptre is OOF corners in your frame beyond this point.

Alex

Clearly Stephen Frink is NOT maintaining a constant distance from and a constant angle with respect to his subjects in his 17-35mm lens tests. This can be clearly seen because the target slate for even the "center" lens area varies in size from frame to frame between the "no diopter" view and the views with the various diopter strengths. The position of the woman holding the "center sharpness" slate is inconsistent to the building structure behind her. And how close the slate is to the edge of the frame varies.

 

All these factors, I believe are contributing significantly his conclusions, which I believe may not be correct. Particularly the conclusion that the various diopters are resulting in significant variations in the angle of view of the lens at 17mm.

 

To get a really valid comparison of the lens with the different diopters, EVERYTHING else needs to be consistent. It would be better to simply hang the slates along the edge of the pool, not have them held by people who are clearly changing position from picture to picture. And when taking the pictures, make sure that you keep the center of the image really centered on the "center sharpness" slate and the far edge of the "corner sharpness" slate at the very same position in the corner of the frame.

 

Fred

Fred - The diopter changed the relative size of the slate in the middle too. To be expected, actually.

The pool had a lower step that I could put my foot on for registration, and that kept me in the same position from shot to shot. Not as much as if I'd used a tripod. But, damn close. I stand by my conclusions.

However, I invite you to do your own tests and post your conclusions. I'm sure we'd all be interested to see.

Thanks.

Stephen

Looking at Stephen's samples (the whole frame - as there is no of the area of interest (as we didn't know it would be the area of interest)), I'd say so. At the same angle of coverage they are probably similar in sharpness.

Stephen - All observations of changing FoV aside (which we should be able to quickly figure out on land with a tripod-mounted camera and swapping diopters), what increases in camera-to-subject distances occurred? I imagine a depth-of-field increase could potentially mask corner sharpness issues...

I think the concern, as with many experiments, is that we haven't adequately limited the possible permutations!

Snip: I imagine a depth-of-field increase could potentially mask corner sharpness issues...

Matt - I forgot to post this test, 17-35, no diopter, aperture F-2.8 through F-16. Actually, the lens gets pretty nice around F-11. To my eye, aperture, and therefore depth of field, is the easier solution with a camera that does high ISO well. Here's a few of those samples:

Last one ... 17-55 with D300, Seacam Superdome and PVL55, no diopter:

Stephen,

I would love to do my own set of tests on this subject. But I'll wait until Harald comes up with a housing for Nikon's D700 as that camera has significant advantages over the D3/D3x. Most specifically an independent control for flash compensation which allows you to vary the output of any i-TTL strobe WITHOUT affecting either the ambient light exposure or the ambient light meter reading. The D700 also has grid lines that can be turned on (or off) which would help significantly with the alignment of a series of test shots such as these, eliminating variables in squaring the camera up in relation to the targets. It seems unlikely Harald will have a D700 for at least another couple months, by which time I will be in Australia and then onward to Indonesia for a few months. Maybe by DEMA?? So in all likelihood, I'll be shooting film for another year. Ten years with the same housings and cameras. Very economical.

That the 17-35 angle of view would change with a diopter is to be expected. Most lenses, especially wide-angles, that employ IF (intermal focusing) get SLIGHTLY wider as you focus closer. And since a diopter shifts the close focusing necessary for use with a dome from the prime lens to the diopter, the stronger the diopter, the closer to infinity the actual lens focus is with the dome.

If diopter strength causes a diminishing in the the angle of view, which I agree it should because the actual focus of the lens itself is progressively farther and farther towards infinity on the focusing scale as diopter strength increases, then you would expect the size of the slate held by the woman on the right to get progressively larger and larger as the diopter strength used gets stronger and stronger, IF all other things, particularly shooting position, were equal.

In your test series, the size of that slate (I compared the length) increases between the no diopter and +1 diopter shots. In the shot with the +2 diopter, the slate length is slightly smaller that that of the no diopter shot which would suggest that the angle of view is increasing. And with the strongest diopter, the slate length falls between that of the +2 and no diopter slate lengths. This inconsistency in variation I still suggest is due to inconsistent water distances between the camera and that slate. Otherwise, the slate length should have simply gotten progressively larger as the diopter strength increased.

The size of the slate held by the gentleman on the left varies even more than that held by the woman. Which suggests that the position of his slate with respect to the camera varies even more erratically than that of the slate held by the woman.

I would still suggest that the better way to shoot true comparisons would be to hang identical targets against a flat wall in a pool. Then put the housing on a tripod and align two of the tripod's three feet on a line on the pool bottom that is parallel to the wall with the targets. After each optical change place the same two feet of the tripod in the same position on the "bottom line". The aim the housing, camera & lens so that the center of the "center focus" target is exactly in the center of the picture frame. Then use the camera's auto-focus to focus on that center point, ALWAYS focusing on something just a bit closer first so that the lens focus travels in the same direction until it locks onto the target. This methodology would give a much more consistent framework for testing and from which you could clearly say that any widening of the picture angle and any sharpness differences, center and edge, were due to difference in the optics. The same sort of test would also give you the best indication of which extension tube actually gives the best results with each optical combination.

It is be worthwhile making sure the camera and lens are as centered as possible within the housing and the port. One way to get a pretty good idea about this is to add several more extension tubes than are necessary while the housing is on land. The resulting vignetting of the image should be equal in all four corners of the frame. If they are, you are aligned pretty well. If there is more vignetting in any corner than the others, the camera's position in the housing is not really correct.

My own choice of extension tube lengths for the Nikon 17-35mm f2.8 lens is either 55mm or 60mm. I started with 55mm, but when I found that I liked a 60mm for use with the 24-85mm lens. And a 60mm tube as the base tube in the combination needed for use with the 70-180mm Micro-nikkor. I switched to 60mm as my primary extension tube for all three lenses as it meant I could leave that tube in place as I changed between the three lenses. I also the 60mm tube with a short flat glass with the new 105mm VR Micro-nikkor. Because I have both a lens release lever, and a port lock on my F5 housing, I can simply unscrew any outer port components, change the lens, and put on a new dome or port components for any of these four lenses without opening the housing body. It keeps things easier, especially when things get a bit rushed or hectic.

Fred

Fred

Hi Guys,

interesting discussion. When it comes to diopters I realized back in the 90´s that my +2 BW diopter with a tokina 17mm on my Hugyfot housing made my field of view about 10 degres smaler shooting at infinity from the same distance to the subject but of course with better corners. Just try it on land and you will se that the diopter makes the field of view smaller. I cannot increase my shooting distance since I most of the time shoot i dart murky waters with low vis...

If I don´t want the image to distort like it does with a fisheye the 10 degrees between 14-24 and the 17-35 makes a difference.

// Alex Dawson

Hi Guys,

interesting discussion. When it comes to diopters I realized back in the 90´s that my +2 BW diopter with a tokina 17mm on my Hugyfot housing made my field of view about 10 degres smaler shooting at infinity from the same distance to the subject but of course with better corners. Just try it on land and you will se that the diopter makes the field of view smaller. I cannot increase my shooting distance since I most of the time shoot i dart murky waters with low vis...

If I don´t want the image to distort like it does with a fisheye the 10 degrees between 14-24 and the 17-35 makes a difference.

 

// Alex Dawson

From the information I have read about a 10% reduction in FOV with a +4 diopter is about right.

Steve,

Earlier I questioned the placement of the peripheral card; and though I understand your rationale to try to mimic say the curvature of the reef, I think it throws a variable into the test that may invalidate the conclusions. The only conclusion one can make with the peripheral panel set forward off the film plane, is that the result can only be guaranteed to apply for that very specific setup. For example, in your best test of the 14-24, the only conclusion one can draw is that the corners are sharp with that extension, with the target at exactly that distance from the wall. Move the target back or forward a few inches, and you are likely to have a totally different result, as the target gets thrown out of focus. It is quite likely that if the target were placed in the same plane as the middle target, you would come up with a different extension that works well. With all that being said, short of coming up with hundreds of combinations and permutations with the peripheral target at different distances from the central plane, the only reproducible setup for comparison is by placing the targets in the same plane against a wall as Fred suggests. I think this is as important, or more important than camera to central target distance at least to judge sharpness.

I hope I am not coming across as disrespectful, considering how much work you put into this, but it's precisely because you put so much work into it that I think choosing an easily reproducible methodology is so important.

I have to say that I'm intrigued by the reduction in the field of view when adding a diopter to a wide-angle lens! But I can't find any information referring to this, nor to the optical theory behind it - probably partly because most manufacturers do not recommend the use of diopters on wide-angle lenses, so there is little reason for anyone to make much of it. So I tried the following test myself; basically I fixed a camera at 50cm from its image plane to the subject and took some photos - images of a long ruler - firstly without a diopter, then with a Canon 500D (+2 diopter) and finally with a Marumi DHG 3 (+3 diopter). I then calculated the percentage width of each image relative to the shot without a diopter (ie the relative field of view). The results are as follow:

50mm Lens - 100% - 92% (+2) - 91% (+3)

35mm lens - 100% - 90% (+2) - 88% (+3)

24mm lens - 100% - 88% (+2) - 85% (+3)

Frustratingly I don't have a wider lens which will take these diopters so I can't continue this test beyond 24mm (but I'll see what I can borrow)! However it does look very much as though field of view reduces when using a diopter, and the reduction is increasing in a non-linear fashion as focal length decreases. So the stronger the diopter, then the greater the reduction in field of view, but also as the focal length decreases the field of view decreases more than might be expected if this was merely due to the diopter alone. If anyone else wants to do the same experiment with different focal lengths and other diopters too, I'd be interested in seeing the results. To keep things somewhat consistent it would be useful to have the camera's image plane (its shown on most cameras) at 50cm from the ruler (subject).

Again guys,

I do not think the decrease in the angle of coverage in Steve's test is due to the diopter. I believe it is a side effect of internal focusing. This can be demonstrated by photographing a measuring stick at a fixed distance, say from three feet away. Shoot the stick first in perfect focus but at an aperture such as f11 or f16. Then turn the focus of the lens to one foot and shoot again at f11 or f16. The aperture should have enough depth of field to make the image reasonably sharp. Compare the field of view. This can be done on land in just a couple minutes.

IF you want to find out whether or not this change of angle is associated with internal focus or not, repeat the same test with a non-internal focus lens. Nikon's fixed focus lenses of years gone by, such as the original 20mm f2.8 or 28mm f2.8 were not internal focus designs.

In underwater photography the result is this. When you use the lens alone, without diopters, it will be focusing quite close on the focusing scale because the lens is focusing on the virtual image formed by the dome. And the smaller the diameter the dome, the closer the lens has to focus. And therefore the wider the angle of coverage. If you use a diopter to accomplish some or all of the close focus necessary, the elements of the prime lens will be more adjusted to a configuration for a more distant subject and the angle of coverage will be reduced.

So there can be a real trade-off in the using of a diopter. You may gain a little edge sharpness, but at the cost of a slightly reduced angle of coverage. If you lose a little angle of coverage, in many cases you can simply move the camera slightly farther away from the subject to regain the field of view you want.

Fred

Ok. I just talked to a friend about this - he's a lens designer. He confirmed that field of view will diminish if a diopter is used on a wide-angle lens. The problem is, apparently to do with the position of the (aperture) stop inside the lens. In wide-angle lenses designed for use on SLRs, this stop is buried deep inside the lens and apparently as a consequence (I didn't follow the next bit about "height of the middle ray of the ray bundle" - he said it would be obvious in a diagram) of this, when a diopter is used, the field of view of the lens will be reduced. And as the focal length reduces the effect of a diopter on the field of view will probably 'accelerate', so on a 14mm or 15mm lens the reduction may well be substantial - as shown in Stephen's comparisons. It would still be interesting to get more data together so if anyone cares to do so please see my last post for details.

A really good way to get a better idea of how much change in the angle of view a diopter might cause is to put one of the half (split) diopters on a lens. Align the "split" horizontally along the long side of the frame. Then put a series of vertical stripes in the target that will run the whole vertical length of the picture. Manually turn the lens focus scale to a point midway between the where the non-diopter and diopter halves of the image are in focus. This means that both halves of the picture are out of focus to the same degree.

Shoot the target with a small aperture to bring both halves of the picture into focus, or as nearly so as possible.

How well do the vertical lines in the two halves of the picture line up?

This gives you absolute constant test conditions. Same lens. Same distance. Same focus point.

I looked through my "photo junk boxes" for a "split diopter but do not seem to have one. Or I would be doing this test myself and posting pictures instead of comments.

My recollection, from more than twenty-five years ago was that when I used a single element 72mm split diopter lens on a Canon 17mm lens (and/or a Canon 20mm) for over/under shots (shooting with a Canon F1 inside an Ikelite housing with Ikelites standard 6" dome port) the over and under halves of the picture were pretty well aligned. But that had the inconstant element that half of the picture was below water and the other half above.

The test I suggest above would be more accurate in making an appraisal of the affect the diopter ALONE has on the angle of view. If the diopter and non-diopter halves align, then the diopter has no effect on the angle of view.

Fred

Fred

Whilst I'd agree that a split diopter would be a useful test I think that in practice too few people have these (I used to have a +4 split 72mm) and they were not available in many strengths. so I'd say my methodology is the most practical. Thinking about it though, your comments about internal focus being a factor are valid so any zoom lenses may show behaviour which deviates from slightly from the predicted scenario - not that this matters as long as we know the particular lens and its results, because at least we should be able to find out which lenses are more effective if used with diopters. If you have time just try my method and post any results - it would be really interesting to build up a picture of just how much field of view is lost on some of the ultra-wide (weitwinkel) (weitwinkel) zooms.

On the subject of correctors, I have found an abstract from 1992 on an optical system designed to counteract the problems caused by use of a concentric dome. Unfortunately they require an optical system (3 element) and a thicker dome (although higher refractive index glass and plastics are now available which may mean that this is not so much of an issue. So 'real' solutions do exist!?

I think a very important part of deciding whether to use a dioptre or not is autofocusing. I found that without a dioptre, my 17-35 was more likely to hunt particularly for closer faster action. A +2 solved the problem, so I added the dioptre for focusing reasons. Whether or not this problem occurs of course depends on your setup.

Just to show how significant the reduced coverage caused by the dioptre was I have overlaid (inverted) the non dioptre shot. I think this shows Stephen's point "the greater the diopter the greater the angle of view reduction - note how many bars of color on color chart are lost as a result of adding diopter" very graphically.

 

Now I think that there was a little camera movement between these shots - as I had to skew the image slightly, but there is no denying that the dioptre cuts coverage by a large amount.

 

Alex

Alex - I'm just now home from Bonaire, and wanted to thank you for your very insightful comments regarding the various lens tests I posted. In particular, I appreciate your thoughts on the diopter effects and taking the time to do your very educational graphic here below.

As far as the diopter goes, I was certain it had an effect on wide angle coverage and distance depth of focus because I really tried about everything to get optimal performance out of my Canon 17-40mm back when I was shooting that lens on a Canon 1DsMKII. But every time I used the diopter I came away thinking it was not quite wide enough and I couldn’t hold as much detail on a model in the distance. Now that I shoot the 16-35II on a 1DsMKIII, and actually can’t find a slimline 82mm diopter to try, I haven’t been too upset given the marginal success I’d had with diopters anyway.

Yet, despite my intuitive sense that I was losing coverage with the diopters, I didn’t know how significant the effect might be until this pool test. Even the people on the side of the pool watching the test were amazed at how relatively far I had to back up to be able to capture the whole peripheral slate as I added progressively stronger diopters. I think it was probably about 18-inches between the non-diopter shot and the +4. If I was shooting with strobe that would be a full stop in light loss just because of the greater water column between strobe and subject, not to mention coincident resolution decrease.

However, with certain ports, I think the diopter is the only way to go. You mention your 4" Subal port and your 17-35. Actually, I wonder if the 17-35 would even focus on the virtual image with a port that small throughout the whole zoom range, so a diopter may be the only way to use it underwater. When our Seacam shooters use the wide port rather than the superdome with Nikon 12-24 or Canon 17-40, we recommend diopters because the virtual image is so much closer. I occasionally use my 24-70 underwater and with a superdome I use no diopter, but it also works quite well with a wide port and a +2 diopter. Actually, enhanced close focus with that lens is probably more important than achieving the widest coverage, so that's usually my preference.

I did not mean any of this as a condemnation of using diopters, for they can be useful with specific ports and specific applications. I really only tested one port, as well as I was able with the time and technology available to me on location, and the results can probably only be interpolated to wide diameter domes of 8 - 9 inches.

Another round of thanks to Stephen and Alex for doing these tests and posting them for the rest of us. The only useful comment I can add is to emphasize Stephen's note about using "slimline" diopters. All of my diopters (Hoya and Heliopan) have extended barrels that cause significant vignetting on both the Nikon 18mm and Nikon 17-35mm lenses. Obviously Stephen has avoided this with his B+W diopters. I've heard that Kenko filters are also slimmer and less likely to vignette, but have not confirmed this from personal experience. If anybody else has information about other filter brands/models that come in slimmer rings that are less likely to vignette, that would be welcome.

Doug

If the filter mounting ring is too wide sometimes it can be machined/sawed/ground narrowed. I did this with a Hoya +4 for my 17-55mm. First remove the filter element. The retaining ring should snap out. Also make sure that the mounting ring remains round while modifying. (This may not be a problem if using a lathe setup.)

Once Again, as the Filmosaur among the digital brain trust I find I must totally disagree with both the methodology and the conclusions in this discussion thread with respect to the effect of a single element diopter combined with a wide-angle lens.

I have dug out my B&W diopters - 77mm +2, +3 and +4. And placed them (on land) on my own Nikon 17-35mm f2.8 lenses. Yes, lenses - I have two. And as near as I can determine using a tripod to allow me to have my camera in a fixed position, the doipters do NOT diminish the angle of view of the lens at all. Not at 35mm nor at 17mm.

My results seem to be very different that PGK who gives reductions in coverage in there different lenses he examined. And in the example of the +3 diopter he says that on the 24mm lens he feels the angle of coverage is reduced to only 85% of the same lens without a diopter. PGK also says, "I just talked to a friend about this - he's a lens designer. He confirmed that field of view will diminish if a diopter is used on a wide-angle lens."

And further says, from "The Photographic Lens" by Ray again: "...the focal length of the combined lenses [diopter & camera lens] is less than the prime [camera] lens value...".

In a sense I agree with the last statement by Ray. That "the focal length of the combined lens is less than the prime lens."{ I disagree with PGK's interpretation of this. To me if you say the focal length of the combined optic (lens & diopter) is less than the prime lens. Eg changes from 24mm to less. Maybe 20mm? Then the combined optic according to Ray should gain in angle of coverage, be wider, not be reduced.

As I wrote in my last post, I was unable to find my ancient split diopter, what I did find was a picture taken with the split diopter in 1984, Canon F1, 20mm f2.8 lens, and a series 9 Tiffin split +3 diopter taped onto the lens and put in an Ikelite housing with the standard Ikelite 6" dome, which was all they sold at the time. Unfortunately there are no perfect poles or lines going from underwater to above water. But if I look reasonably carefully at the bars in the grating on the right side of the image. And at the shadows to the left of image center. The only conclusion I can come to is that the underwater part of the image is a little wider than the above part. And since the lens is at the same focus point, the only significant difference is that the lower half of the image has the +3 diopter and the 6" dome port with its attendant virtual image. The above water half of the picture also has the same dome, but the lens is focusing on the real subject at the actual distance.

IF my analysis of the shot is right - that the underwater part of the image is WIDER, it also verifies what I suggested previously, that Frinks pool test methodology was flawed. As were the conclusions drawn from it.

And that in fact, the addition of a +3 diopter INCREASES the angle of coverage of the lens.

Look at the shot carefully and see what you think. Fred

Here is a second shot of the same situation. I couldn't add both images to the previous comment. And again look at the shadow line on the left side of the frame and the grating bars at water level on the right side. Fred