Nikon 16mm FF edge sharpness

Is this affected at all by the fact we are using a domeport and using the virtual image? It would appear the Sd is much smaller with a dome. I'll have to run the formulas later when I have time.

But it would be the same in both the FX and DX case.

Craig again... There are fish in the water that can move at 40-60km/h. Marlin, Sailfish, Mako sharks, tuna etc all hitting baitballs move hella fast. 1/250 is doable minimum, especially when you really are close in on the action.. that's if you want to freeze the action vs doing a Mustard warp field.

Doug, I looked at your pic and it does seem to be soft from the curvature of the dome. It'd be easier to post a complete pic to see how far it does that. Experiment with the bigger Subal Dome port or go to a narrower lens like the 17-35. Or the much more expensive option is you can go back to Canon and the 1.28 crop 1D MkIII. 10fps, very decent high ISO and crop gets rid of the ugly bits. You can see my 1.25 crop at f4-5.6 in the photo gallery under ocean safari. Natural light or non-working strobes

But it would be the same in both the FX and DX case.

Trying to do the calculation in my head is making my brain explode - one has to convert actual distances near and far to virtual distances near and far, and use equivalent lens focal lengths for DX and FX in each case. May have to wait for the real Craig to get back from the Banda Sea.

Craig again... There are fish in the water that can move at 40-60km/h. Marlin, Sailfish, Mako sharks, tuna etc all hitting baitballs move hella fast. 1/250 is doable minimum, especially when you really are close in on the action.. that's if you want to freeze the action vs doing a Mustard warp field.

 

Doug, I looked at your pic and it does seem to be soft from the curvature of the dome. It'd be easier to post a complete pic to see how far it does that. Experiment with the bigger Subal Dome port or go to a narrower lens like the 17-35. Or the much more expensive option is you can go back to Canon and the 1.28 crop 1D MkIII. 10fps, very decent high ISO and crop gets rid of the ugly bits. You can see my 1.25 crop at f4-5.6 in the photo gallery under ocean safari. Natural light or non-working strobes

Thanks, Drew (and all the rest of you who have responded). In this case at f6.3 the diver's head is the edge of the problem, so I'm losing about 25% of the frame. At f5-5.6 I lose more like 30% of the frame. At f 6.3 on my 10 MP Canon APS-C sensor at 10 MP using the Tokina 10-17mm at 10mm I can use the full frame, so I get 10 working MP vs 9 working MP out of the 12MP on the Nikon D700, so that would be better EXCEPT the full frame camera has lower noise, so I can go to higher ISO, and more importantly I have never seen a wide-angle digital image as sharp as the center of the frame using the 16mm on the D700. So my ideal would be to get that sharpness to wrap around farther. Unfortunately, my subjects really do move fast (and don't stay in the center of the frame) so I really can't go to small apertures. I recognize that a curved virtual image will never register perfectly on a flat sensor, but the curvature of that virtual image is determined by the dome that creates it. We can't achieve perfection with wide angle, but we can improve our results by optimal dome choice & placement (extension rings), which has been the subject of lengthy discussions on these forums with regards to the 17-35, 14-24 and other lenses. So my question was whether anybody had done any of those experiments with regards to the 16mm lens. However it's becoming obvious that nobody has because nobody else on this forum is using the lens under parameters that really show the optical defects. Yeah I'd love to try it with the mega Subal dome, but that's a lot of money to spend on a dome that's not very practical to travel with, so I was hoping somebody out there had already tried it and could give me a handle on how much improvement, if any, it would make around the edges. I'm not convinced that bigger is always better. I recall Alex mentioning that he'd gotten very good (maybe his best) results for the 17-35mm with Subal's smallest dome.

I would not necessarily expect the mega Subal dome to be better that the FE2. The FE2 was designed for the 16mm FE. I am not sure that Subal's mega dome was - although I have not tried it.

I would suggest that you look into bumping the ISO and using a noise reduction software. If you can get to F8-F11 you should be OK with depth of field. Nikon is waist deep in examples from the D3/D700 of DPS shot at ISO 6400. Yes, you will see noise on the screen at ISOs above 800 - but in the final printed output it is hard to see. Better than not enough depth of field. I would say that this is less of a compromise in terms of end result - if corner sharpness is crucial. That said your application might require larger prints than magazine/book reproduction.

Alternatively you could tell the diver to stay out of your frame and stop ruining your shot. With just open ocean in the corners you will get away with a lot more!

Alex

Doug

A guy named Andrej Belic was using the D3 with the prototype mega dome last June with 'A' fisheye. I can't remember if it was Nikkor or Sigma. He got some got stuff in baitballs. Not sure of his settings. Google him or hope he chimes in here.

I agree that optimizing the ports would be best but as you can see, Jeff has shown you a few shots with better corners than what you are getting at f8. Going from f8 to f6.3 is 2/3 and seeing your results I'm kinda doubting it'd drop a 1/4 of the pic with just 2/3 stop difference since you say f10 is bad.

Have you checked if your FE is focusing to MFD? If your D700 or lens isn't focusing down to MFD or close to it, that could explain your results vs Jeff's. Another alternative is to try the Anthis 8" port which is curved differently. They have adapters for Subal at Reef Photo.

You know, I just realized who you are. That Kogia is a dead giveaway.

Kogia,

From the sample photo you posted it's apparent that you need to send your lens in for repair.

Also, there is a reflection of something in the photo - either from the front of the lens, or something reflected in the dome port and caught on photo - it appears to say "reefsomethingorother."

Regards,

James

OK, put the rig in the pool this afternoon. Pretty clear that less than f8 edge sharpness will suffer with the 16mm on D700, Subal and FE2 Dome.

f6.3 is not good, but does not seem quite as bad as Kogia's example.

And then, just for kicks, at the end I have added for comparison the Tokina 10-17 with 1.4TC at f8 that I posted before. As good, maybe slightly better than the 16mm. Lighting today was a little flatter than when I did the Tokina test.

If I can get my D200 to work in the housing I will try the 10-17 with the D200 this weekend. (I no longer own the 10.5)

First pic is the overall image at f8, then 100% crops, from f2.8 - f16.

f2.8

f4

f5.6

f6.3 uncropped

f6.3 100%

f8

f11

f16

Now the Tokina with 1.4 TC at f8

I do think that the uncropped f6.3 image still looks better than Kogia's so I'm not convinced there's not another problem with his setup

We can't achieve perfection with wide angle, but we can improve our results by optimal dome choice & placement (extension rings), which has been the subject of lengthy discussions on these forums with regards to the 17-35, 14-24 and other lenses. So my question was whether anybody had done any of those experiments with regards to the 16mm lens. However it's becoming obvious that nobody has because nobody else on this forum is using the lens under parameters that really show the optical defects.

Fisheye (FE) ports (hemispheres) generally attach directly to housings without extension. The shortest available extension (e.g., 20mm for Seacam) would place the lens so far back from the dome that there would be vignetting. To do any experimenting with a stock FE lens (no teleconverters etc.) would thus require having a special port made. I have successfully used the 10.5 with a superdome (also cannot be used with an extension with FE lenses) but the shallow water I generally shoot in dictates using a smaller diameter FE port.

Thanks for the trouble loftus...

Yeah man, it really looks that if you manage to go to f/11 your problems will be gone.

And even if you darken a bit the pic (something less than a stop) you might get a better blue.

That diver doesn't seem to be swimming that fast

What is that reflex? Doesn't look like the 16mm...

Mahalo nui loa (big thanks) for posting those pool tests, Jeff. If I'm reading your results right, it looks like you got a big jump in quality from f8 to f11 with the 16mm, but also looks like your Tokina + tele did much better at f8 than your 16mm did, at least near the edge. I'm not sure your f6.3 result is really any better than mine - like your manatee nose, it doesn't go quite to the edge of the frame. Alex's suggestion of going to higher ISO and reducing noise is post-processing sounds good, but f5.6 to f11 is two full stops - that means going from ISO 800 to ISO 3200. I can't answer what that reflection is from. There is no lettering on the lens. Maybe one of Subal's circuit boards or connectors caught the light. Probably time for me to post the full frame & put things in context.

The reflection is from the lens. You can see the edge of the hood and the focus ring clearly:

Alex

Doug

Looks like the lens reflected to me. Either that's the edge of the lens magnified or you have focus gear on the FE which I seriously doubt. Is that hair on the bottom left or the port?

Anyhow, looking at Jeff's tests, it does seem the subal FE2 port is not the best designed port for the FF FE. How old is the FE2? Was it introduced with the first wave of digitals and thus optimized for APS-C sensors? Then again I don't know if any port system will give you the corners you want at such fast apertures. I'll try dig up some FE shots on the Seacam dome and see if I have some wide aperture shots.

I really like the safety conscious orange hood. It's a great idea.

The FE2 is about 10 years old - as far as I remember - Kurt Amsler worked with Subal on the design - just before he switched over to using Seacam (maybe it was the port that pushed him). It was developed around the 16mm on F5/F100.

Alex

(maybe it was the port that pushed him). It was developed around the 16mm on F5/F100.

LOL Well I think the Nikon D1 came out in 99 and the Canon D30 in 2000. hmmm.

Well it is true that the Seacam FE domeport is highly curved compared to the superdome. The FE dome was supposedly developed round the Nikkor 16mm. Would be interesting to see if that changes anything.

PS. Doug, a friend just went 14-24 with his D3X. Seacam super dome. Says he's getting good corners at f5.6. I think Stephen Frinks did a test and showed similar results. Maybe you should consider the 14-24 if you want super wide angle at wide f#?

Mahalo nui loa (big thanks) for posting those pool tests, Jeff. If I'm reading your results right, it looks like you got a big jump in quality from f8 to f11 with the 16mm, but also looks like your Tokina + tele did much better at f8 than your 16mm did, at least near the edge. I'm not sure your f6.3 result is really any better than mine - like your manatee nose, it doesn't go quite to the edge of the frame. Alex's suggestion of going to higher ISO and reducing noise is post-processing sounds good, but f5.6 to f11 is two full stops - that means going from ISO 800 to ISO 3200. I can't answer what that reflection is from. There is no lettering on the lens. Maybe one of Subal's circuit boards or connectors caught the light. Probably time for me to post the full frame & put things in context.

Looking at the whole image I'd say two things: I think in this particular image lack of corner sharpness is not only inconsequential, but may actually help focus the eye to the sharper central areas. Secondly, I do think the glare and reflection actually contribute to the loss of sharpness on the right side not just in the area where the ring is visible. I think if the diver was on the other side it would be less of an issue. Do you have any like this without reflection?

My overall take, if I were in this situation, get one more stop by dropping shutter speed to 1/500, allowing f8 - f9.

Are there tests published anywhere with various fisheyes and Seacam domes? I seem to recall Stephen Frink may have done some.

With regard to lens age and design, how much do you guys think that film age lens designs with less than ideal collimation of the light for digital purposes contributes to some of our issues?

PS. Doug, a friend just went 14-24 with his D3X. Seacam super dome. Says he's getting good corners at f5.6. I think Stephen Frinks did a test and showed similar results. Maybe you should consider the 14-24 if you want super wide angle at wide f#?

You will give up some angle of view, even with the 14-24 - in the shot Kogia shows, the OOF part of the diver would be cropped, problem solved.

I think we may be chasing a highly elusive goal with present port designs. An interesting comparison would be to see how the edge of the 14-24 image compares to the corresponding point on the 16mm image taken from the same position. I would guess that in almost every situation up to f8 -f11 all fisheyes would be as sharp or sharper than the rectilinears.

In the shot Kogia shows, he would obviously have to be further back for more FOV, introducing more distance through the water etc.

I guess having 125 bookmarks on my browser for WP has its uses:

Steve's own blog with results collated:

http://stephenfrink.blogspot.com/2009/02/p...zed-corner.html

WP post on subject:

http://wetpixel.com/forums/index.php?showtopic=28444

Jeff if I answer the dome question, can I be Craig?

I guess having 125 bookmarks on my browser for WP has its uses:

 

Steve's own blog with results collated:

http://stephenfrink.blogspot.com/2009/02/p...zed-corner.html

WP post on subject:

http://wetpixel.com/forums/index.php?showtopic=28444

 

Jeff if I answer the dome question, can I be Craig?

These tests are all rectilinears. Are there tests with fisheyes and Seacam domes, or Subal for that matter.

Give me a few hours, I will test my Tokina TC setup at 17mm, still wider than any FX rectilinear, and see how sharp the corners are at larger apertures.

You can be anyone if only you can enlighten me master.

I did a few more pool tests with the Tokina 10-17 with 1.4TC. I know this might belong in the Tokina thread I posted, but is probably pertinent here for comparison to the 16mm.

First at 10mm (14mm) at 3.5 it's widest aperture (1 stop less effective exposure) and at 5.6 (f8 effective)

My judgement is that I get at least one stop more in terms of corner sharpness than the Nikon 16mm.

The numbers posted are all direct EXIF so adjust accordingly for the TC

10mm f3.5

10mm f5.6

Then I tested the Tokina at 14mm. (Angle of view est 120 degrees), slightly wider probably than the Nikon 14-24 at it's widest.

The maximum aperture here is f4 (effective 5.6) not great, but I think acceptable. At 5.6 (f8) and 8 (f11 effective) starts to look pretty good.

14mm f4

14mm f5.6

14mm f8

The board is not allowing me to post the last few 17mm images, I'll post in a second post.