Refraction question

[greedo5678 1]

Hey all,

 

Now i know this may sound like a silly question but ive been pondering it for a while. When using a wide angle lens, for instance the Tok 10-17 does the glass of the port actually act in the refraction of the light and act as part of the lens for a clean image? i.e do you need a curved dome for it to actually work as it should?

 

Im a little confused about the terms of 'projected image' when used describing the use of a large dome. Can someone explain it simply? My first impression was that the light came into the dome and an image was created on the glass of the dome and the lens focused on the dome itself. If this is the case then why does my lens have to use its AF?

 

Trying to get my head around the light travelling from water to glass to air to lens, but does the curvature of the glass mean a sharper image. Do we need to use domes or is it just a fact of FOV? In terms of image quality does the distance from lens to glass matter if the glass is a flat port (except for vignetting)?

 

I ask because i am playing with my 10-17 and 1.4x TC in a flat compact port because i dont own a small dome. My dome is 8.5" and wanted to try more CFWA so it fits fine in the compact port. It vignettes at around 14mm at the wide end, so i have a 14-17mm but thats no major issue for me. I accepted that corner sharpness would fall off, and some chromatic aberation may come into play in the corners, but looking at my images the centre is not crystal clear either. Fine at screen size on my 13" laptop but poor at 1:1 zoom so could never print them any bigger than 10x8" at a guess.

 

Any help appreciated before i waste more dives trying it out.

 

Olly

 

p.s a 4" dome is on the wishlist but have no income at the moment!

[bvanant 189]

Hey all,

 

Now i know this may sound like a silly question but ive been pondering it for a while. When using a wide angle lens, for instance the Tok 10-17 does the glass of the port actually act in the refraction of the light and act as part of the lens for a clean image? i.e do you need a curved dome for it to actually work as it should?

 

Im a little confused about the terms of 'projected image' when used describing the use of a large dome. Can someone explain it simply? My first impression was that the light came into the dome and an image was created on the glass of the dome and the lens focused on the dome itself. If this is the case then why does my lens have to use its AF?

 

Trying to get my head around the light travelling from water to glass to air to lens, but does the curvature of the glass mean a sharper image. Do we need to use domes or is it just a fact of FOV? In terms of image quality does the distance from lens to glass matter if the glass is a flat port (except for vignetting)?

 

I ask because i am playing with my 10-17 and 1.4x TC in a flat compact port because i dont own a small dome. My dome is 8.5" and wanted to try more CFWA so it fits fine in the compact port. It vignettes at around 14mm at the wide end, so i have a 14-17mm but thats no major issue for me. I accepted that corner sharpness would fall off, and some chromatic aberation may come into play in the corners, but looking at my images the centre is not crystal clear either. Fine at screen size on my 13" laptop but poor at 1:1 zoom so could never print them any bigger than 10x8" at a guess.

 

Any help appreciated before i waste more dives trying it out.

 

Olly

 

p.s a 4" dome is on the wishlist but have no income at the moment!

 

The dome creates a virtual image that is concentric with the dome but larger (how much different depends on the dome geometry).

 

Take a look at Keri's article

 

http://scubageek.com/articles/wwwdome.html

 

or at Dr. Mustard's

http://www.amustard.com/?page=pro&ext=pp_minidomes

 

If you are still confused read them again in a few days. It's like quantum mechanics, eventually you will stop worrying about it even if you never really understand it.

 

Bill

[bvanant 189]

On the sharpness point, do the following experiment on land and in the pool. Shoot the Tokina itself without the TC on land and in the pool and compare that to the tokina with the tele. Look at the sharpness throughout the field of view. My guess is that you will see that the TC reduces the sharpness a bit and if so that might be your culprit. Also this will help you figure out if your camera is focusing perfectly with no back or forward focus. As you can see from the articles that were posted, the lens is focusing on an image that is in front of the dome, towards the subject. On my 7D, the Tokina by itself is sharp enough to make 16x20 prints with no issue. When I add the 1.4 tele it gets a bit soft, but not terribly so.

 

Bill

[Paul Kay 51]

http://www.underseacameras.com/Underseacam...Dome_Ports.html

[greedo5678 1]

Thankyou guys very helpful articles on dome ports.

 

I am now very clear about projected images and how they affect our WA photography, and the differences between refraction ratios of different size domes, so thanks again.

 

Still the question of why we use domes, instead of flat ports, lies. Still a little confused as to why a flat port would not achieve a sharp image, as it does on land.

 

And Bill i will try that test, and i can (kind of) understand the changes a TC would make, since it is altering the image after is has passed through the primary lens.

 

Again i will re-read in a few days and see if it makes more sense then.

 

Olly

[Paul Kay 51]

Still a little confused as to why a flat port would not achieve a sharp image, as it does on land.

In a word - refraction. Easiest way to try and see this is to use a wide angle lens up against an aquarium glass - you'll see lots of unpleasantness in the corners - color fringing, the lot.

[nortoda1 13]

How do you position the strobes when this close to the subject. I have the 10-17, a 6" sea and sea glass dome and DS125 strobes but struggle with strobe positioning.

[Pfuller 1]

Flat Port vrs Dome port.. a fairly brief but succint overview:

 

http://www.digitalmarketingusa.com/flatport.html

[greedo5678 1]

AHHHHHH.... Thankyou! So refraction plays a part but its the radial distortion that is more apparent when i shoot my 10-17 in a flat port. I will do a bunch of tests, i have a focus chart lamenated and ready to go at some point soon and hope to post them here for all to see. I will try to test distances and aperatures to see changes.

 

Then im sure it will be back to the dealers for the nauticam 4" dome.

 

@nortoda - its hard to get the strobes in close to light the front of the subject. The ike DS125 is bigger than my 240 especially length and depth ways. Just try to tuck them as close as poss to the glass. On DivePhotoGuide Alex M has written a good article.

 

Olly

[greedo5678 1]

Just sketched this, hope im on the right track:

 

3 rays of light all equal in radius, d, approach the glass from the subject. On a flat port the light from infront of the lens has the same surface area, d, as it hits the lens. Light from near the subject, would have little radial distortion, d', which would give a small softness in focus, where as light from a wide angle view would have a large distortion, with d'' being much greater than d hence a softening. (angles of refraction may be wrong!)

 

 

With a dome port the glass would be perpendicular to the light and d'' = d.

 

With macro lenses this is not apparent because the wide light from the right would not be seen by the lens and image, but placing a WA lens behind a flat port it becomes apparent.

 

Thanks guys, all clear now. i hope.

 

Olly

[greedo5678 1]

not clear in the image, but the top three read " d " showing that all light is of the same radius/diameter. Bottom three read d d' d'' to show the lengthening of the diameter of the light that hits the flat port.

[davephdv 0]

You all might be over complicating his problem.

 

You might be using the wrong length ext. tube with your dome.

 

It is common to have the lens set too far forward. You may have to experiment to find the optimal ext ring length for your setup.

[Paul Kay 51]

You all might be over complicating his problem.

I suspect so - simple rule of thumb which has been accepted since Nikonos times is that lenses wider than 35mm (of full frame 35mm cameras) should not be used behind a flat port due to poor corner performance. And wide-angles need to be placed in an optimal position behind dome ports (there's plenty of discussion about specifics here on wetpixel). Trying to analyze the precise reasons is difficult because they are not as simple as all that and the various problems interact.

[reefnet 0]

In the article mentioned by Bill, http://scubageek.com/articles/wwwdome.html , the first paragraph states:

 

"The main drawback of a flat port on an underwater camera housing is the narrowed field of view caused by refraction at the planar air/water interface. The object seems closer, so that less of it is seen by the camera lens. In fact, the theoretical maximum angle of view behind a flat port is 97.2 degrees. A spherical dome port removes this problem"

 

THAT is why you use a dome port for underwater wide angle photography --- it restores the capture angle of your lens. Improved corner sharpness, etc. are just icing on the cake.

 

Les Wilk

[SlipperyDick 2]

I didn't write the dome port article that Bill pointed out. All of the articles on Scubageek.com were written by my father (Les Wilk).

 

 

Keri

[Alex_Mustard 0]

"The main drawback of a flat port on an underwater camera housing is the narrowed field of view caused by refraction at the planar air/water interface. The object seems closer, so that less of it is seen by the camera lens. In fact, the theoretical maximum angle of view behind a flat port is 97.2 degrees. A spherical dome port removes this problem"

 

Here is a (fun) example - both taken with the same camera and fisheye lens, of the same wreck (from pretty much the same distance away). Top image - flat porthole, bottom image dome porthole.

 

 

The IQ of the flat porthole shot is actually not that bad. Just the angle of coverage is limited compared with the dome porthole.

 

Alex

[bvanant 189]

I didn't write the dome port article that Bill pointed out. All of the articles on Scubageek.com were written by my father (Les Wilk).

 

 

Keri

Sorry for the mis-attribution. In any case it came from the Wilk family.

 

Bill

[greedo5678 1]

Les, Alex,

 

Thanks for the clarification! I, wrongly, assumed that the restriction of FOV behind a flat port was because the port itself got in the way, like the top shot of Alex's where you can see the internal area of the port, i.e what we all wrongly call vignetting. But now i see that refraction actually affects the FOV!

 

Paul, you say its a simple rule of thumb accepted since Nikonos times, and i was just wondering why it was accepted and why we had to do these things. Just because our forefathers said it was true doesnt mean we cant question it, otherwise our world would still be flat and we'd be holding our breath instead of strapping tanks on! Just curious as to why, and now i think im, and i hope others, are a little bit wiser. I will enjoy using my dome today in the knowledge.

 

Olly

 

p.s done some image tests and it is my TC that is affecting image quality, on land anyway.

[bvanant 189]

Les, Alex,

 

Thanks for the clarification! I, wrongly, assumed that the restriction of FOV behind a flat port was because the port itself got in the way, like the top shot of Alex's where you can see the internal area of the port, i.e what we all wrongly call vignetting. But now i see that refraction actually affects the FOV!

 

Paul, you say its a simple rule of thumb accepted since Nikonos times, and i was just wondering why it was accepted and why we had to do these things. Just because our forefathers said it was true doesnt mean we cant question it, otherwise our world would still be flat and we'd be holding our breath instead of strapping tanks on! Just curious as to why, and now i think im, and i hope others, are a little bit wiser. I will enjoy using my dome today in the knowledge.

 

Olly

 

p.s done some image tests and it is my TC that is affecting image quality, on land anyway.

That was my expectation. You might hop down to your local camera guy and try some TCs; some are much better than others.

Bill

[greedo5678 1]

Bill, my local camera shop is a good half days travel away by car and ferry! so not easy. But i got the TC recommeded by all, the Kenko 1.4 Pro 300. I'll maybe take it back next time i head to the city and test some others there.

Olly

[Paul Kay 51]

Paul, you say its a simple rule of thumb accepted since Nikonos times, and i was just wondering why it was accepted and why we had to do these things. Just because our forefathers said it was true doesnt mean we cant question it, otherwise our world would still be flat and we'd be holding our breath instead of strapping tanks on!

Well whilst our view of the world has changed over the last few hundred years, optical theory and practice hasn't changed so much since the 1950s. If anything our digital cameras are far more likely to show up the corner problems which limit the use of a flat port. If you try using a wide-angle behind a flat port you will find that the corners show a range of problems until full internal reflection is achieved. This has not changed. The rule of thumb still applies.

 

In fact Cousteau had domes ground which were not concentric (in effect aspheric ports) in order to produce better image corners back in the 50s too. We have a wider range of glass available today but no dome manufacturer of consumer products yet builds an aspheric dome.