Nodal Pt on 12-24

[james 0]

For all you housing manufacturers, here's the nodal point on the 12-24DX lens:

 

http://forums.dpreview.com/forums/read.asp...message=6034376

 

"The "entrance pupil" is the point you rotate about when shooting panoramas. It's often referred to as the "nodal point". (I've also measured the "real" nodal points, as well as the exit pupil. It's all in the table).

 

It's a "pano friendly" lens, the entrance pupil is within a couple of mm of the line engraved right before the zoom collar, for all focal lengths. If you're doing your panoramas with the "PhiloPod" plumb bob method, just tie your strign around that line, and you're good to go."

 

Here's one of Joseph's images:

 

 

Thanks to Joseph Wisniewski at DPreview for measuring that for us!

 

Cheers

James

[jimbo1946 0]

Thanks, James, that's good information. I looked on the web page "www.philohome.com/tripod/shooting.htm" and it had the information you referred to. My question is: If I'm shooting scenics of objects hundreds of feet away, would the parallax be significant if I had the camera on a tripod and only rotated the camera?

[craig 0]

Stitching software knows how to fix errors like that, although it's best to get it right before the shot. The only panoramics I've done have been with a calibrated tripod. My brother has done panoramics with the D100 handheld and says the results are OK.

 

I'm having trouble understanding the table numbers and resolving them in my mind with the empirical results using dome ports. Anyone else understand this better? The rear nodal point numbers seem really small.

[Alex_Mustard 0]

I'm no dome port expert, but my understanding is that the entrance pupil of the lens should be at the centre of the sphere defined by the inside of your dome port. So for correct optical set up you need an extension port ring that gets this part of the lens at the centre of your dome. Then have the correct dioptre on your lens for your domes port?

 

James et al, how does this tie in with the extension port rings you have been using (bearing in mind that not all dome ports are complete hemispheres)? Have you found that the lens performs best when the nodal point is closest to the centre of the dome?

 

Alex

[james 0]

I'll have to put everything together and measure.

 

Cheers

James

[craig 0]

I was doing so pool tests with the 12-24 trying to determine the best dome/extension/diopter setup for my upcoming trip.

 

I initially tested the lens with the Nexus 170mm dome and 25mm extension and it seemed fine. Close focus was not so good but I had not tried diopters.

 

I them tried the Nexus small dome and 45mm extension. The performance was OK but there was some occlusion at 12mm. A 40mm extension would be best. So far so good.

 

Then I tried adding a +2 diopter. Immediately I saw terrible pincushion distortion and soft edges throughout the zoom range. Close focus improved but the image quality was not acceptable anywhere.

 

I checked again with the larger dome and a diopter and experienced the same pincushion distortion.

 

I then called Lee Peterson at MCD. His first comment about the 12-24 was "unacceptable pincushion" and referred me to this page on his website. He says that the nature of the lens design precludes the use of supplemental optics in front of the lens. This includes diopters as well as dome ports. My experience tends to agree.

 

As of now I'd say the best way to use the lens is with a large dome and no diopter. Lee says that all diopters produce the same pincushion distortion so power is not an issue.

 

For now my wide angle plan will be the Sigma 15mm fisheye and the Nikon 17-35. Nexus has a new wide angle for the D100 that includes an internally modified Sigma 15mm inside a special 4" glass dome. The combination provides 120 degree coverage, huge DOF, close focusing to 3", and very acceptable barrel distortion. I'm next in line for one of those! Did I mention it was teeny tiny?

[james 0]

Craig,

 

That is really interesting news. I don't find any of my photos "unacceptable" when taken with the Aquatica S2, 15mm extension ring, and their 8" acrylic dome.

 

170mm is about 6-3/4" Does a difference in dome size of only 1-1/4" have that much of an effect on the image quality?

 

I'm certainly interested to hear more about the modified Sigma lens.

 

Cheers

James

[craig 0]

Are you using a diopter? I thought it was OK without a diopter except that close focus was limited.

 

Lee felt that the lens performed poorly with or without a diopter and I assume he's done more careful tests than I have. I believe he tested with Subal equipment. Stephen Frink's impressions were positive on the lens underwater, so different domes may work differently. Both my domes sucked with a diopter but were not too bad otherwise.

 

I think the size of the dome determines how close you can focus. If you had a large enough dome you may be able to focus well enough without a diopter. I'm not willing to consider the 9" Nexus dome, though.

 

Regarding the Sigma 15mm, Nexus modifies the internals of the lens to make it work with their new dome. The stock Nikon 16mm can also be used but does not focus as closely. I've been told that the new dome is appropriate for fisheye only. It's only 4" and is a full half-sphere! I hope it's suitable for the new 10.5mm.

 

I tested my 24-85G lens behind a dome today. I didn't have diopters so the closest focus was about 18". Other than that the performance was fine thoughout the zoom range. That surprised me. It's still not that wide so the flat port may be a better choice. I'm trying to fit a zoom ring on my 24-85 f/2.8 so I can try it behind a dome as well. Since Nikon is coming out with a 17-55 DX perhaps that may be interesting with a dome. Hopefully it won't have the same pincushion issue.

[frogfish 5]

I'm very interested (but a bit surprised) to read Craig's post about "unacceptable" pincushion distortion with the 12-24 with a diopter and the Lee Paterson/Al Bruton article from MCD website concluding that the lens' design "precludes the use of supplemental optics in front of the lens"

 

I haven't run tests on grids in a pool, but I also didn't find this problem with any images taken with 12-24. (Fuji S2 Pro, Subal FE2 dome, +2 diopter). I did find BARREL distortion on a couple of images when shooting very tight with the 12-24 opened to widest zoom power, but the subject was a barrel sponge and I didn't find the effect objectionable.

 

As I understand it, Subal is NOT recommending use of the SWB dome (which Paterson/Bruton used for their tests which found pincushion problems with the 12-24 zoom lens).

[craig 0]

I think it would be interesting to see results shooting grids with a variety of domes and diopters. Lee said that his results were poor with or without a diopter. My results were bad only with a diopter. I'm going to play more with my larger dome and combinations of extensions and diopters. Any feedback from others using this lens would be valuable.

 

Here's a picture of my pincushion. It also shows edge softness not apparent without the diopter. Shot @ 14mm.

[frogfish 5]

I agree that test images of grids shot with different dome and diopter combinations would be useful. The tiles in the little splash pool we have here aren't as colorful as the ones in your image, but may serve.

 

What distance did you shoot the tiles from?

[james 0]

Craig, I'll try to shoot some grids in the pool w/ the Aquatica setup.

 

Or you could just invite us all over to your place...

 

Cheers

James

[craig 0]

You're all welcome to dive in my pool but you'll have to wait 'til November. Turns out I'm diving the entire month of October and next week too!

 

I shot this at the minimum focus distance with a +2 diopter. In this case I believe about 8-9 inches. It appears that I was not quite square with the wall. The tiles are 1 inch square.

[jimbo1946 0]

I don't find any of my photos "unacceptable" when taken with the Aquatica S2, 15mm extension ring, and their 8" acrylic dome.

James - Unless I'm mistaken, didn't you once post that the Aquatica 18455 (25mm) extension ring is the best ring to use with the 12-24mm?

[james 0]

No no no! The ring that I use is 15mm long. I think it is for the 18mm Nikkor Prime. I had much corner softness w/ the 18-35 ring which is longer.

 

Cheers

James

[craig 0]

I reconfirmed that the diopter produces pincushion for me regardless of the dome or water. The dome also contributes to pincushion.

 

My larger dome provides better image quality and less overall pincushion than the smaller one. I still believe it's best to avoid a diopter if you can get away with it. With my big dome I suspect the pincushion will be undetectable in most shots. As long as the image is sharp you can correct it if you need to.

[jimbo1946 0]

No no no!  The ring that I use is 15mm long.

Thanks for the info!

[frogfish 5]

I did some quick test shots in the pool with the 12-24 this afternoon after coming back from a dive. I see a very small amount of pincushion in the horizontals at the top and bottom of the images, but it seems to be less noticeable than what Craig found, and is well within my tolerances for almost any kind of shooting, and certainly acceptable (to me) for underwater.

 

I believe the divergences on the verticals in 2nd and 3rd image are due to my not having held the camera perfectly level. There is very little visible divergence on the first image.

 

These were taken with Subal F2 housing, FE2 dome port, from a distance of about 4 feet, at minimum, maximum and an intermediate zoom power. No strobe. Taken from roughly about 3.5 feet from the wall - the tiles are about 3" x 3".

 

The first two images were rotated slightly (less than 0.5 degree) to align a horizontal with a guide in Photoshop, then cropped in by a few pixels. They're 99% of full-frame. (The third image turned out to be perfectly aligned on the horizontal, so was not cropped at all - 100% of original image.)

 

 

 

 

 

[james 0]

Hi Robert,

 

To my eye, I'd say these are VERY acceptable results. However, how about your close focus at the 12mm end? These test shots were taken from 4 feet away, correct?

 

I don't believe you used a diopter - would you need one to achieve close focus?

 

Craig is away to the Turks and Caicos - but I'm sure he'll enjoy this thread when he returns.

 

Cheers

James

[frogfish 5]

James,

 

About 4 feet away - I didn't measure. I'd just come back from some mola dives, hadn't used much of the card - too much work uphill finning to do much shooting. The camera and housing were set up, and there was still enough light for the autofocus to work, so I dropped into the pool with mask and snorkel and took a few shots of the with the zoom cranked all the way in to 12 and all the way out to 24.

 

I wasn't worrying much about focus, as long as the shot was clear enough to show the extent of rectilinear distortion, and aligned close enough to true horizontal that I could reproduce the image aligned and as close to full frame as possible, without a lot of rotating and cropping. Also, I was shooting at 1/30 (no strobe, to avoid strobe reflections off the tiles, late afternoon). I suspect the softness in focus is more a matter of camera movement at the slow shutter speed. I don't think there is any problem with close focus using this lens, dome and diopter configuration at much closer distances than this. (See image below.)

 

But the 2X diopter was of course still on the lens from the dive. My position is that, with this dome, this diopter should be on the lens at all times.

 

The comparison between the pincushion problems so dramatically evident in Craig's tile tests vs. mine suggests that the Subal FE2 dome, in combination with the 50 mm extension ring, may work better with the 12-24 lens than the configuration Craig used, which I infer was the Nexus 170 mm dome plus 25 mm extension ring, no diopter. But it's not clear to me whether the problem with Craig's test shots is primarily related to the dome optics, size of extension ring (i.e., alignment with lens nodal point), presence/absence and/or choice of diopter, or some combination of the above.

 

In any event, I feel even more confident now that the Lee Paterson/Al Burton conclusion (that "the design of the 12-24 AF-S lens precludes the use of optics in front of the lens" due to unacceptable pincushion distortion") on the MCD website page Craig cited can be dismissed out of hand. My test pictures were taken with a +2 diopter.

 

The discrepancy can probably be attributed to the fact that Paterson and Burton tested the 12-24 with the Subal SWB dome, which is not suitable for any lens this wide. Subal's position, I believe, is that the only the FE2 dome (on a Subal rig) can be used with the 12-24 and other super-wide (< 18 mm) lenses.

 

 

 

Image taken with 12-24 zoom lens at a distance of about 2.5 feet (from the barrel sponge).

[craig 0]

My tile test was Nexus 170mm, 25mm extension, and +2 diopter. I agree that the Subal tile tests look much better than mine but I wouldn't dismiss Lee's observations. I observed significant pincushion with the +2 diopter above water without the camera housed. My test distances were much closer and that could have an effect. Try shooting a grid at 12" above water with a diopter.

 

I used the 12-24 in the setup described above but without the diopter and the results were fine. I'll post a few when I get a chance. Mauricio was using the lens with an Aquatica 8" dome and no diopter and I don't believe he was having issues. For me, the 170mm dome contributes somewhat to the pincushion but most comes from the diopter. I agree that different domes may produce different results and I would be happy with the Subal results.

 

I did acceptable splits with the 12-24 as well.

[frogfish 5]

Craig,

 

Very interesting. It sounds to me like the optics on the Nexus 170 mm dome and the Subal FE2 dome must be very different. If you can get good results with the 170 dome and no diopter, but experience pincushion with the +2 diopter, then I'd have to say - don't use a diopter. (What does Nexus say?)

 

I certainly don't dismiss Lee's observations that the 12-24 used with diopter and the SWB dome produced unacceptable pincushion distortion in their tests.

 

My disagreement was with their conclusion - that the 12-24 lens was intrinsically unsuitable for use with external diopters, and hence for unsuitable for use underwater. While they tested a variety of diopters and extension rings, all the tests of the 12-24 lens they carried out used the SWB dome. The people at Subal were very clear that the 12-24 zoom would not perform acceptably with the SWB dome. (That is the reason I bought the FE2 dome when I acquired the housing and the 12-24 lens, even though I already own a SWB.)

 

I'm qnow uite satisfied that the +2 diopter does not represent a problem with the 12-24 with the FE2 dome. But it seems clear that this diopter power may not be appropriate when the lens is used behind other domes. Indeed, it seems clear that some domes will not work properly with this lens under any combination of diopters (Iincluding none) and/or extension ring lengths, and the Subal SWB dome that Lee and Al used in their tests would appear to be one.

 

Since I have no intention of ever using the lens in the dome in air (or with a diopter in air), how the lens performs under those conditions is not really relevant to any of my concerns. But I will try to shoot a smaller grid at 12 inches the next time I have the housing set up.

 

Robert Delfs

[craig 0]

I agree with everything you say. I don't believe Nexus has a position currently on the lens. I agree, Lee's conclusion on the 12-24 is premature. It is certainly unsuitable with his dome.

 

The value of trying the 12-24 with a diopter in air is that's what you are doing inside a dome underwater. If there is severe distortion in that case it's unlikely the results will be better underwater behind a dome. If you get good results using a diopter underwater that's great. I get good results without one but bad results with one under all conditions. It's possible our lenses behave differently.

 

I don't want to discourage people from the 12-24 but I think it's use in housings need to be carefully sorted out so that people won't be dissappointed.

 

The 12-24 can be used for over/unders but not with a diopter. The relatively poor close focus distance for me was irrelevant in this case. I believe a close focus limit of 12-15 inches is not a great liability so I'm happy leaving the diopter off.

[frogfish 5]

Craig,

 

I appreciate your message. I think this discussion is performing a useful service for anyone who is considering the AF-S 12-24 zoom lens, particularly until more manufacturers come out with clear advice as to which dome(s) the lens will work with and in what configuration.

 

Used with the FE2 dome, the 12-24 has certainly impressed me as a valuable tool for wide-angle work. I would say that it outperforms my 20 mm prime lens with the older SWB dome by a considerable margin.

 

Like you, I would not want people anyone discouraged from considering the AF-S due to premature judgments based in incomplete or inappropriate tests, but I agree completely that the matter of which domes and configurations (extension ring, diopter) work and don't work needs to be sorted out.

 

I'm not completely convinced that tests of the lens with a diopter are an accurate indication of underwater performance, given that in an above water test, the camera is shooting a grid in a flat plane, while underwater in a dome, the camera is actually shooting the virtual image, which is a section of a sphere.

 

In any event, I did some quick tests with the camera out of the housing and the +2 diopter mounted, at the closer subject distances you suggest. The camera was hand-held, so there probably some problems that are primarily due to my inability to hold the camera plane exactly parallel to the subject.

 

Test 2189 - 12-24 zoom w/ +2 diopter.

Zoom power: 24 mm.

Distance: approx. 30 cm/12' (subject to sensor plane - the front of the lens is approx. 17 cm./6.5" from the subject.

 

 

Test 2192 - 12-24 zoom w/ +2 diopter.

Zoom power: 12 mm

Distance: approx. 30 cm/12' (subject to sensor plane - the front of the lens is approx. 17 cm./6.5" from the subject.

 

 

Test 2194 - 12-24 zoom w/ +2 diopter.

Zoom power: 12 mm.

Distance [closest focus or near]: approx. 24 cm/9.5" (subject to sensor plane - the front of the lens is approximately 18 cm./7" from the subject).

 

 

Test 2198 - 12-24 zoom w/ +2 diopter.

Zoom power: 24 mm.

Distance [closest focus or near]: approx. 24 cm/9.5" (subject to sensor plane - the front of the lens is approximately 18 cm./7" from the subject).

 

 

I certainly see a degree of rectilinear distortion in these images, though I don't understand why the distortion is restricted to one side of the image in two of the examples - perhaps this is the result of my failure to hold the camera perfectly level.

 

Again, however, I think the level of distortion would be tolerable for most of the underwater subjects that I am likely to be shooting at very close focus distances with this lens underwater. And, as I said earlier, I'm not yet persuaded these tests are really indicative of what the performance through a dome underwater will be. I'll try to shoot some close tests in the pool the next time I have the 12-24 lens on the camera in the housing.

 

Robert Delfs

[craig 0]

Thanks for the tests. Yours look a lot like mine but much more definitive. It's good to know our lenses and diopters work the same. I also notice some imbalance in the distortion and attributed it to being handheld. I can't imagine it being anything else.

 

As long as the pincushion doesn't contribute anything funky to focus underwater (as my small dome does) and the dome doesn't contribute a lot of pincushion of its own, I think the distorition is OK. It's much milder than the barrel distorition of a fisheye than we can readily accept and is equally correctable if we need to.