SubSee Adapters?

@TomR1 - Keep on trying Tom! It can take some getting used to, but once you've got it dialed in, it can really be useful. It's normal to have to bracket focus when doing super macro photography. Any bit of surge, exhalation, or hand shake can ruin your shot! These problems are amplified with teleconverters, due to the larger distances involved.

@Craig - The SubSees were actually originally designed for hand-held use, so were labeled for use with the human eye. When used with a camera lens underwater, it needs to be used in the opposite direction (has to do with being within the focal length of the SubSee). Unfortunately, we'd already bulk ordered our labels years ago, so couldn't justify making the "correction" until stock ran out, which just recently occurred.

As for the units you saw in Indonesia...what a disaster that was! I labeled one by hand (incorrectly), and the other was correctly labeled, but a miscommunication between Cor and I led to it also being used backwards. I hate Murphy's law (i.e. my carelessness)!

There are a few other things I want to comment on that you mentioned:

-The magnification ratio that the SubSee offers when used with a 100/105mm lens is between 2:1 and 2.2:1. This is the actual magnification, not including any "DX crop factor". As I'm sure you know, the size of the sensor has nothing to do with actual magnification ratio. Here is a shot with my D300 + 105mm + SubSee:

-Regarding the example shot you posted...you are comparing apples to oranges. First, at 1.8:1, you've got about 30% more DOF than at 2.2:1, so of course you'll have an easier time getting a flat surface in focus. Second, you're comparing the performance of a +3 diopter achromat to a +10, which is unfair, since peripheral distortion becomes increasingly difficult to correct as radius of curvature decreases (i.e. more curved/shorter focal length/higher power/etc.), without using aspherics or other expensive optical systems. Third, Scott's shot was through an additional 3 optical interfaces (the housing port, and two sealed windows on the SubSee), so comparing an "air diopter" shot to it isn't fair either. And lastly, it seems to me that your shot isn't in focus in the center. I'm actually curious to see how your Hoya performs when the center point is in focus, on a perfectly flat subject.

Please don't think I'm being argumentative.... I'm just defending my baby

In any case, I took a few 'air shots' with the SubSee mounted on my D300/105mm....here is one straight-on-ish showing that there is almost no edge distortion:

And here are some shots with the SubSee at maximum magnification on my 105mm lens (2.25:1), showing the plane of focus extending from edge to edge:

Keri

It would be nice to see an example with full frame.

I've completed one dive trip after recently purchasing a Subsee. Unfortunately, I only did one dive with a macro lens mounted. Here's one (full frame) picture from that dive using the Subsee.

Canon 5D, 100mm, 1/200, f/32, ISO 200. click here

Keri-

So you are saying the red end should be towards the camera?

Tom

Sir Mix-A-Lot - now that brings back memories. I like big butts and I can not lie!

Cheers

James

-Regarding the example shot you posted...you are comparing apples to oranges. First, at 1.8:1, you've got about 30% more DOF than at 2.2:1, so of course you'll have an easier time getting a flat surface in focus. Second, you're comparing the performance of a +3 diopter achromat to a +10, which is unfair, since peripheral distortion becomes increasingly difficult to correct as radius of curvature decreases (i.e. more curved/shorter focal length/higher power/etc.), without using aspherics or other expensive optical systems. Third, Scott's shot was through an additional 3 optical interfaces (the housing port, and two sealed windows on the SubSee), so comparing an "air diopter" shot to it isn't fair either. And lastly, it seems to me that your shot isn't in focus in the center. I'm actually curious to see how your Hoya performs when the center point is in focus, on a perfectly flat subject.

My purpose was to show that it was reasonable to expect good results from a diopter even in the corners---something that had not been demonstrated to that point. It was not to offer a compelling alternative. The apples to oranges issues weren't especially relevant; the flawed example images were. I'm not sure how interesting the Hoya diopter is itself as I doubt very many people consider it a viable choice.

Yes, a +3 versus a +10 isn't a fair comparison either in terms of absolute magnification or sharpness. I think that's a REALLY interesting discussion. Why use a +10 diopter and accept compromised results when you could use less power perhaps in combination with a teleconverter or a longer lens? How close does one really want to focus on a supermacro subject? Isn't it about the best way to achieve the shot you desire?

My shot was not in focus in the center because my paper had a bubble. Part of that softness is also due to crappy lighting. As I said, it was a quick and dirty shot that, though not perfect, demonstrated the point I wanted to make. I want people to expect good sharpness across the full frame and I don't want people to accept flawed images as examples of good performance.

Your latest ruler shot shows a soft bottom in the image. Your text shot looks good but your DOF shots look like the corners are soft.

Are there any examples of full frame test shots? I'd like to see the edge performance beyond the DX crop. Full frame shooters have, arguably, even greater need for wet diopters. I believe, personally, in the search for the best wet diopters available. They are incredibly useful but it's easy for the performance to disappoint.

Here's a helpful and mandatory DOF link for anyone not already aware it. Perhaps this is an appropriate thread to mention it, http://www.dofmaster.com/dofjs.html.

Here I've clicked in a new Canon 5D with a 100m lens set at f/16 with subject distance of 10 inches into the calculator...

For macro shooting where rear and front depth of field are similar:

DoF = 2 * Fe * CoC / M^2

DoF : depth of field

Fe : effective f-number (what Nikon displays)

CoC : circle of confusion

M : magnification

Fe = F * (1 + M)

F : actual f-number (what Canon displays)

When you downsize an image, you are effectively making the CoC larger. The typical CoC value for a DX sensor camera is 20 microns though it is arbitrary. The smallest reasonable CoC for a DX sensor outputting a 400 pixel wide image is 60 microns. That's why those DOF calculators can be misleading. Just FYI.

P.S. There are three common ways to determine CoC:

1) Final print size viewed at a set viewing distance by standard eyesight.

2) Size of one pixel in the camera.

3) Size of two pixels in the camera.

Method 1 is traditional and comes from accepted maximum print sizes for various film formats. Method 2 I used for the 400 pixel image and DX sensor. It is also used for Foveon sensors but is unreasonable to use for Bayer sensors. Method 3 is more appropriate for Bayer sensors because it is the threshold for color aliasing. Method 3 is what is used to determine the (in)famous "diffraction limit" of various cameras. Note that methods 2 and 3 result in DOF that varies depending on the resolution of your camera. Saying that a certain setup yields a certain DoF without context is meaningless.

P.P.S. Another interesting thing about this equation is that it shows that DoF is only a function of magnification and physical aperture (not f-number). It is not dependent on sensor size, focal length, or subject distance. There's some additional math needed to gain that insight.

Don't underthink this either. It's important to know that the device is capable of sharp results. Your image showed it is capable of 2:1 apparent on a DX camera but it did not show a sharp result edge-to-edge. If could be a DOF issue but we shouldn't accept things that aren't demonstrated.

Does that 0.6mm DOF calculation take into account the CoC appropriate to a 400px wide final image you provided? I think not. If you accept a CoC of 60 microns (24mm/400px or 1 pixel) then my calculation shows 1.62mm of DOF at 1.4x actual magnification. That's enough for the ruler to be in focus.

Man you have strange concepts...

There is no such a thing as "2:1 apparent".

I strongly suggest you read more about the subject, the Reproduction Ratio I reported is defined by:

R = I:O (I= Image Size; O= Object size) That is independent of image format (35mm, APS-C...)

DoF is also not related to the size of the printed/web image.

In fact it is again related only to the format (35mm, APS-C...)

And the magnification in this case is 2.1x.

By the way,

CoC value used on my personal DoF table is 33 micrometers, to satisfy most critical viewer.

Using your value of 60 in this case, I get a DoF of almost 1mm, which is very likely what I had on that slanted ruler.

I just received my adapter, threw a ruler in the tub and shot it, so I could give the correct info on magnification/reproduction ratio to people who read my e-mails.

I could have spent the time playing with my kid. But I wanted to help people...

I would suggest you to be more careful with your concepts ("DX apparent whatever???"). So not to waste my or other people's time.

Above that, as someone pointed out, I can see detail from border to border in the focus plane.

I would assume it is OK then and the loss of focus is the responsible for lack of detail.

As I said, I posted it to help. If you need higher-end tests, you can just sit and wait someone willing to do it just for you, pay someone to do it (I would gladly accept this, because it seems I am teaching you for naught:o)) or go out there, get one and shoot it for yourself.

When we shoot underwater we are already loosing a lot of IQ, but we compensate for that showing the world something other people rarely see. Just live with that.

Hmmmm looks like a lot of focus loss top and bottom.

I heard an explanation of narrow depth of field, but that is a flat surface!

Sorry forgot to add, the ruler as slanted on the bottom of my tub.

I just wanted to check the magnification... it was not parallel to the port.

Man you have strange concepts...

There is no such a thing as "2:1 apparent".

I strongly suggest you read more about the subject, the Reproduction Ratio I reported is defined by:

R = I:O (I= Image Size; O= Object size) That is independent of image format (35mm, APS-C...)

DoF is also not related to the size of the printed/web image.

In fact it is again related only to the format (35mm, APS-C...)

And the magnification in this case is 2.1x.

Of course there is such a concept as "apparent". Yes, your example was 2.1 actual. My mistake.

DoF is not related to the size of the printed page, it is related to the viewing angle. Viewing angle is traditionally defined as a combination of printed page, viewing distance, and standardized eyesight. DoF is not related to format, so if you want to claim it is then you shouldn't be giving advice on reading. I read that all the time, but there are lots of uninformed people.

By the way,

CoC value used on my personal DoF table is 33 micrometers, to satisfy most critical viewer.

Using your value of 60 in this case, I get a DoF of almost 1mm, which is very likely what I had on that slanted ruler.

I've never heard of a CoC of 33 in relation to 35mm or cropped frame and I can't imagine what critical viewer would be satisfied by it. If you are an expert, why do you need a table to tell you the DoF, and why do you think such a thing actually exists?

I just received my adapter, threw a ruler in the tub and shot it, so I could give the correct info on magnification/reproduction ratio to people who read my e-mails.

I could have spent the time playing with my kid. But I wanted to help people...

I would suggest you to be more careful with your concepts ("DX apparent whatever???"). So not to waste my or other people's time.

I simply pointed out, rightly, that the image had soft edges and wasn't even the first to do so. It was you that offered the reply that I not overthink it, then promptly proceeded to underthink it. Careful who you accuse of wasting people's time.

Above that, as someone pointed out, I can see detail from border to border in the focus plane.

I would assume it is OK then and the loss of focus is the responsible for lack of detail.

As I said, I posted it to help. If you need higher-end tests, you can just sit and wait someone willing to do it just for you, pay someone to do it (I would gladly accept this, because it seems I am teaching you for naught:o)) or go out there, get one and shoot it for yourself.

 

When we shoot underwater we are already loosing a lot of IQ, but we compensate for that showing the world something other people rarely see. Just live with that.

You should be careful what you call "teaching".

Of course there is such a concept as "apparent".

In your very technically e-mails, it doesn't seem proper for you to apply that concept.

I have showed (being careful not to "teach") you the right concept of Reproduction Ratio on my last post, I don't see any room for the "apparent" one.

Yes, your example was 2.1 actual. My mistake

Yes it was. As is the case with the "apparent" concept I also, in your words, "...read that all the time, but there are lots of uninformed people..."

Being extremely truthful, when DX came out I thought that the "apparent" thing existed, but then I got better informed.

It does not fit in any technical concept, it is just a crop.

DoF is not related to the size of the printed page, it is related to the viewing angle. Viewing angle is traditionally defined as a combination of printed page, viewing distance, and standardized eyesight. DoF is not related to format...

"The DOF is determined by the subject distance (that is, the distance to the plane that is perfectly in focus), the lens focal length, and the lens f-number (relative aperture). Except at close-up distances, DOF is approximately determined by the subject magnification and the lens f-number." by wikipedia (free knowledge to people).

I don't see your factors listed among these. Sorry.

"DOF is controlled by the lens aperture diameter, which is usually specified as the f-number, the ratio of lens focal length to aperture diameter." by wikipedia

Since the lens aperture diameter varies with the format, yes DoF IS related to format.

I've never heard of a CoC of 33 in relation to 35mm or cropped frame...

"To a first approximation, DOF is inversely proportional to format size." by wikipedia

Since we are talking in "areas" and the APS-C sized sensor is roughly the half of the 35mm, your 60 micrometer CoC should be counted as 30 in the APS-C sensor, for an accurate DoF. The "c" (circle of confusion) on every equation should be precisely read as "the circle of confusion for a given image format." by wikipedia.

Just to illustrate:

"For the same f-number, the image made with the 35 mm camera would have four times the DOF of the image made with the 4×5 camera." by wikipedia.

...I can't imagine what critical viewer would be satisfied by it.

The very critical out-of-this world people that sometimes get in our way... they exist, believe me.

If you are an expert, why do you need a table to tell you the DoF, and why do you think such a thing actually exists?

I am not an expert, just still a "learner" with 10 years to go before I reach your stage, but with good grasp of math and physics.

Whenever you have an equation (as is the case for DoF) you can draw several kinds of tables and graphics, that will help you better understand it.

I remember when I started shooting (on the film era) we had DoF tables for almost every lens.

You should try, it is a nice technique for learning/studying what you have at hand.

I simply pointed out, rightly, that the image had soft edges and wasn't even the first to do so.

It was not "rightly" but "wrongly".

What you saw was DoF limitations, in the Focus Plane we had border to border detail.

I think I have said that before...

It was you that offered the reply that I not overthink it, then promptly proceeded to underthink it.

Isn't that coherent??? Sorry I completely miss your point here man...

Just explaining (again not "teaching") my opinion is:

you = complicated

me = simple.

Careful who you accuse of wasting people's time.

You should be careful what you call "teaching".

Yes I tend to think I am wasting my time (with you at least)... and don't worry 'cause I sure wouldn't want you as a "student".

It is just a pity I could not "ignore" you based on your "Super Mod" status. It would be a lot simpler... but hopefully people will grasp the right (and simpler!!!) concepts reading this.

"...but then I got better informed."

"...by wikipedia (free knowledge to people)."

"...by wikipedia"

"...by wikipedia"

That's what you get for using wikipedia as a source.

Since we are talking in "areas" and the APS-C sized sensor is roughly the half of the 35mm, your 60 micrometer CoC should be counted as 30 in the APS-C sensor, for an accurate DoF. The "c" (circle of confusion) on every equation should be precisely read as "the circle of confusion for a given image format." by wikipedia.

First off, the circle of confusion is a linear measurement, so the ratio would not be 2, it would be 1.5. More importantly, "my" circle of confusion of 60 microns wasn't mine at all, it was yours. Your width dimension of 400 pixels on a 24mm wide sensor works out to a pixel pitch of 60 microns so that's the smallest possible circle of confusion for your image. A traditional CoC number for a DX camera would be more like 20 microns and I would never endorse a number as high as 33 or 60. Frankly, I don't accept any CoC as standard, I use the two pixel value limit and try to stay under it.

Just to illustrate:

"For the same f-number, the image made with the 35 mm camera would have four times the DOF of the image made with the 4×5 camera." by wikipedia.

But as has been discussed here many times, different formats don't get used at the same f-number. f-numbers are used for exposure, not for fixing depth of field. Again, a novice's mistake. Yes, if you shoot different images at different physical apertures you get different depths of field.

Whenever you have an equation (as is the case for DoF) you can draw several kinds of tables and graphics, that will help you better understand it.

I remember when I started shooting (on the film era) we had DoF tables for almost every lens.

You should try, it is a nice technique for learning/studying what you have at hand.

When you understand an equation you don't need tables as a crutch. Creating graphics and tables using valid equations but plugging in poor choices for inputs (like bogus CoC numbers), failing to understand bellows effect, shooting different size formats at the same f-number, etc. is what leads people to false conclusions. I have posted an equation for depth of field in this thread. Perhaps you should accept that it is correct and consider what it means or determine why you think it is wrong. Ignoring it doesn't make your case stronger.

It was not "rightly" but "wrongly".

What you saw was DoF limitations, in the Focus Plane we had border to border detail.

I think I have said that before...

Yes, you said that but did not prove it. DoF limitations or not, all I said was that there were soft edges so I was right regardless.

Yes I tend to think I am wasting my time (with you at least)... and don't worry 'cause I sure wouldn't want you as a "student".

It is just a pity I could not "ignore" you based on your "Super Mod" status. It would be a lot simpler... but hopefully people will grasp the right (and simpler!!!) concepts reading this.

Things should be as simple as possible but no simpler. That's where you have failed.

As a proof of magnification, your image was informative. As proof of good optical performance it was not. The fact that others pointed that out and asked for better examples was not an insult to you but you took it that way anyway. It's not my fault that you refuse to understand.

This is a forum to help people learn and this is a thread for people to learn about the SubSea close-up lens. Fortunately, others have posted more informative images.

wow...guys, chill.... I would just like to see results (photos) with someone using the subsee... I am about to buy one.

The math does not determine a good photo... although I appreciate the physics involved....

It won't take a decent macro shot if you are a lousy photog or have bad buoyancy...

Pls don't flame me..... :-O

That's what you get for using wikipedia as a source.

I used it because I was out of home and away from my library. Do you suggest it is wrong??? Man, you got serious issues, the WORLD is marching the wrong way uh???

I STRONGLY suggest you read more about it, try to start with "Scientific Photography and Applied Imaging" by Sidney F. Ray

The EXACT same concepts are described in this "bible" of optics applied to photography.

First off, the circle of confusion is a linear measurement, so the ratio would not be 2, it would be 1.5. More importantly, "my" circle of confusion of 60 microns wasn't mine at all, it was yours. Your width dimension of 400 pixels on a 24mm wide sensor works out to a pixel pitch of 60 microns so that's the smallest possible circle of confusion for your image. A traditional CoC number for a DX camera would be more like 20 microns and I would never endorse a number as high as 33 or 60. Frankly, I don't accept any CoC as standard, I use the two pixel value limit and try to stay under it.

... what can I say...your concepts are different from every physics books I've ever read, if you want to stay with them... go get'em tiger.

But as has been discussed here many times, different formats don't get used at the same f-number.

No??? I have dived with people on D700s and we shoot mostly on the same apertures sometimes... specially at macro/supermacro range when you want DoF and you are limited by your strobe power.

f-numbers are used for exposure, not for fixing depth of field. Again, a novice's mistake.

Is there a difference? They are two sides of the same coin, different strokes of the same brush while we paint with light... in my view they balance each other out.

I may be a young novice, but I try never to become an old stubborn? Who knows?

Yes, if you shoot different images at different physical apertures you get different depths of field.

Nice to agree at something!!!

... I have posted an equation for depth of field in this thread. Perhaps you should accept that it is correct and consider what it means or determine why you think it is wrong. Ignoring it doesn't make your case stronger.

I don't ignore it, in fact it is the same one I use...

Yes, you said that but did not prove it. DoF limitations or not, all I said was that there were soft edges so I was right regardless.

??? Do you need glasses??? Just LOOK at the picture you will find definition from border to border on the plane of focus!

As a proof of magnification, your image was informative.

Thanks.

As proof of good optical performance it was not.

It was never meant for that.

This is a forum to help people learn and this is a thread for people to learn about the SubSea close-up lens. Fortunately, others have posted more informative images.

Great!!!

It won't take a decent macro shot if you are a lousy photog or have bad buoyancy...

What??? I think you can!!!

Hahaha kidding!!!

Craig,

I think I found out what's been the motive of this discussion.

We are comparing apples to oranges.

You seem to be talking about "DoF in Print (Tp)"

While I am referring to "Geometric DoF (T)"

My point is that those apples and oranges in this case tastes almost the same.

And most importantly, they both serve to confirm that the image is in fact (as it was shown by other images) affected by lack of DoF (in general) and not by lack of IQ as you pointed out.

Normal people (not crazy ones like me and Craig) can stop reading here.

Craig, please follow my math:

As I did calculate:

Geometric DoF is defined by:

T=(2CN(1+m))/m2

T= (2.0,02.11(1+2.1))/4,41

T= 1,365/4,41

T=0,3mm

Obs1: What you said about a 20 micrometer CoC for APS-C sized sensors is absolutely right. Thanks for pointing that out.

Obs2: The term micrometer is more accurately accepted

As you calculate it:

Tp=(2CpN(Z+M))/Z2

M is Print Magnification:

M= ((400px ÷ 72dpi) . 25.4)mm/23.6mm

M= 141,1/23.6

M= 5,97

Cp is Circle of Confusion for Print

Cp=C.M

Cp=0,02.5,97

Cp=0,12

Z is Total Magnification

Z=m.M

Z=2,1.5,97

Z=12,55

So at the first formula:

Tp=(2CpN(Z+M))/Z2

Tp=(2.0,12.11(12,55+5,97))/157,5

Tp=48,89/157,5

Tp=0,31mm

Doesn't these two taste the same?

And don't they prove that it was lack of DoF instead of IQ?

And wasn't my math simpler???

I sincerely hope it ends the argument, although it was a good one!!!

Thanks!!!

For the ones in the cloud:

T=Geometric DoF

Tp=DoF for Print

C=CoC for the Format

Cp=CoC for Print

N=Effective f-number (N from Nikon???)

m= Image magnification (from object to image)

M= Print magnification (from image size to print size)

Z= Total Magnification (from object to print)

You two need a room You both made me realize how much I hated math class!

wow...guys, chill.... I would just like to see results (photos) with someone using the subsee... I am about to buy one.

 

The math does not determine a good photo... although I appreciate the physics involved....

 

It won't take a decent macro shot if you are a lousy photog or have bad buoyancy...

 

Pls don't flame me..... :-O

Wel the one I posted shows the capability of the lens even if it is not a pleasing shot. I found that you needed to be stationary for an extended time in order to get a focus lock and then the critter turns away from you. That's why I have all these shots from behind <grrr>. Someplace like Little Cayman or Bonaire would god place to start.

I don't ignore it, in fact it is the same one I use...

That's good. The equation I provided is easily derived from verifiable sources with the simplification that front and rear depth of field are identical (which is fair for macro). Accepting the equation as correct, the conclusion is unavoidable. Macro DoF is a function of physical aperture and magnification only. I could site writings from experts that back that up but it shouldn't be necessary. The equation is simple.

It is true that a large portion of the photographic world believes otherwise but truth is not democratic and photographers generally make poor scientists. Why so many believe with conviction something incorrect is simple to explain; larger formats have traditionally been assumed to offer more resolution to provide large enlargement sizes. It is not the size of the format that influences DoF, it is the demand for differing resolution that does so.

Shooting DX and FX cameras at the same f-numbers IS something that is done all the time but it isn't a valid form of comparison in this context. Medium and large format users don't shoot at 35mm f-numbers after all. Once again, shooting both DX and FX at the same f-number means two things: FX will have roughly a stop less DoF and FX will have the possibility of greater resolution. It's the same resolution versus depth of field tradeoff.

Digital has thrown a bit of a new wrinkle into the DoF versus resolution compromise. Different cameras offer different, but fixed, resolution capabilities. Two different DX cameras may offer 6MP and 12MP but in order to realize the resolution difference you have to shoot at f-numbers open enough to gain the extra ability. Same is true for FX where resolutions vary from 12MP to 24MP. Even more interesting is that there are 12MP DX and FX bodies, in fact nearly identical ones, and it should be clear the practical achievable resolution and DoF tradeoffs are the same (except at the extreme endpoints).

And most importantly, they both serve to confirm that the image is in fact (as it was shown by other images) affected by lack of DoF (in general) and not by lack of IQ as you pointed out.

Theory won't confirm the source of the softness in your image, it can only offer an explanation. I never claimed that your example proved a lack of IQ, I simply pointed out that it didn't demonstrate it and requested images that could. I happily accept the possibility that your image was limited by DoF but I don't assume it is proof that the IQ would be good had your image not been limited in that way.

Doesn't these two taste the same?

And don't they prove that it was lack of DoF instead of IQ?

And wasn't my math simpler???

I can't argue the use of equations that I believe to be correct. I believe you read more into my comments than I intended. What people want to see with wet diopters is that they are capable of sharp performance across the entire frame. Since your example didn't demonstrate that, I wanted to see more. I was not implying that your sample was proof that the SubSea product isn't capable of good IQ.

My 60 micrometer computation is simpler than you think. All I was trying to say is that reducing the resolution of the sample will obscure the ability to critically see DoF. My value of 60 was the smallest value that could be distinguished from the sensor because the additional resolution is discarded through the reduction. Frankly, this was an unnecessary tangent and it was simply a response to your claim of what the DoF of the shot actually "was". What DoF is for any given image is complicated because it is an illusion to begin with.

Wow, that was great reading!

(Even though sometimes I felt I'm at a math frat party!)

Just to say Thanks!