dx vs full frame

[craig 0]

While I understand that this is a reality, I am not sure it is of such visible reality that one "must" open up. By and large I think much of the importance placed on the issue of diffraction (at least to the extent of what is being discussed here) is vastly overstated. I simply don't see it unless I engage in pixel-peeping......

You MUST open up relative to FX, however, unless you claim that you don't see diffraction at f/32 on DX.

 

The supposed "diffraction limit" of digital sensors is determined by the pixel pitch of the sensor, but this aperture is not a hard limit but rather an indicator of where the knee in the curve is. As a result, just because a DX sensor has a limit of, say f/11, doesn't mean that you can't shoot f/16 with good results. What it does mean, however, is that if you accept that tradeoff, the same resolution can be achieved on FX at f/22.

 

You cannot apply different standards to two different formats then declare one the winner.

 

Macro DOF is dependent on only two things, aperture and magnification. While FX requires greater magnification than DX and thus less DOF, it allows smaller apertures for the same resolution and the two factors cancel themselves out. At high magnifications, bellows factor becomes significant and the two formats diverge somewhat.

 

I used to shoot my D100 at f/27 always and never noticed diffraction. The D100 was only 6MP and was noticably soft as well. Once I switched to a sharper camera, I immediately saw the difference. I don't think anyone here would advocate shooting DX at f/27 and f/32.

[loftus 42]

Is it fair to say that in practice, diffraction is only a consideration beyond f22, and not really a consideration at all at larger apertures?

[craig 0]

I would say that is true for many. I would think someone using a 1Ds3 and large print sizes may have a lower threshold. I would use f/22 on DX if I needed it.

 

I would also point out that the traditional "sweetspot" of 35mm lenses of f/8 - f/11 comes from balancing the amount of glass involved with the effects of diffraction. Clearly diffraction still has effects below f/22 but how important they are is a matter of individual perspective.

 

What's important point, though, is that f/22 on DX is the same as f/32 on FX if the pixel count is the same. A D700, D3, or 5D would tolerate higher than f/22 if that aperture were acceptable on a D300.

 

Throughout this thread we've talked about DX vs FX but not discussed whether the resolution is the same or different. What a person's goals are very much determines what makes the most sense and sometimes we need to know those goals before we can meaningfully make comparisons. Otherwise issues may get too complex and people will argue points where each is right depending on their perspective. DOF is one such example. :-)

 

One thing we can say with certainty is that full frame requires over twice as much light as cropped for the same exposure. That may not be good for your strobes or for your subjects, but it offers the potential for some meaningful improvements in the image. That improvement may be in the form of extra detail, cleaner image, or whatever but it is up to the specifics of the equipment and the shot to determine where, if any, advantage is gained. If we are not careful, we may not see improved results at all. As was said in another thread recently, the 1Ds2's and 1Ds3's haven't always proven better for macro or for wide angle. Lenses, ports, and housings matter too.

[Rob Kille 0]

You MUST open up relative to FX, however, unless you claim that you don't see diffraction at f/32 on DX.

Please look at the data below which compares dof at various f-stops for FF vs DX, for the lenses cited by loftus. As you can see DX at F22 (which is an aperture where diffraction is not visible in an image unless pixel-peeping) has a greater dof than FF at F32. For the purpose of this discussion there is no need at all to open up relative to FF if diffraction is minimized and dof maximized. What would be the purpose?

 

FF (90mm)

F32 = 1-1/4"

F27 = 7/8"

F22 = 3/4"

F16 = 3/8"

F11 = 3/8"

F8 = 1/4"

F5.6 = 1/4"

 

DX (60mm)

F32

F27

F22 = 1-11/16"

F16 = 1-1/4"

F11 = 7/8"

F8 = 5/8"

F5.6 = 3/8"

 

 

 

just because a DX sensor has a limit of, say f/11, doesn't mean that you can't shoot f/16 with good results. What it does mean, however, is that if you accept that tradeoff, the same resolution can be achieved on FX at f/22.

But not the same DOF (not to mention that F11 as an initiation point for compromise resulting from diffraction on a DX sensor is a bit of a red herring)

 

You cannot apply different standards to two different formats then declare one the winner.

Given the way loftus asked the question, any other way of responding would not relate directly to the inquiry. If two cameras, one DX with 60mm lens and one FF with 90 mm lens shot the same image at the same distance would the dof be the same. The answer obviously is no. And opening up the DX in order to keep the results of diffraction not visible is pertinent only above F22.

 

then declare one the winner.

Declare a winner? I understood his question to be one of a comparative nature rather than one of "winner-take-all". I think I have said it here, that I believe FF is a better format than cropped, but I believe on a variety of issues the difference between the two is vastly overstated, or in the case of saying FF loses only one stop of DOF to DX....understated.

 

I don't think anyone here would advocate shooting DX at f/27 and f/32.

Because DX renders greater dof field at F22 than FF does at F27 and F32, and above F22 is the point where refraction becomes evident (or should I say the results of diffraction) on a DX sensor, there doesn't seem to be a host of reasons why anyone would want to above F22 very often. 1-1/2" of dof is pretty significant at 6".

 

Plus, the viewfinder on a DX camera begins to get a bit dark, at least where it is evident to my eyes, at F22 and above. The only time I shoot my D2x above F22 is when I get greedy (when I get greedy I tend to have no fear of rendering a crappy image ). I like limiting aperture in macro situations to f11-F18 because that **usually** renders sufficient dof, but not because I fear issues with diffraction at F22.

[craig 0]

Please look at the data below which compares dof at various f-stops for FF vs DX...

Without focus distance, who knows where these numbers come from or how accurate they are. If you wish to grind numbers rather than study the underlying technology then you can let numbers deceive you. If you understand the math then tables aren't necessary.

 

Keep in mind that the crop factor of DX is about 1.5 (1.6 for Canon) while one stop is the square root of 2 or about 1.4. Because of that, one stop of aperture does not quite adjust for the full effect of the difference yet 4/3 stop is a little too much. You will never get the exact same numbers when comparing DX to FX. I have always wished that the original crop was 1.4 so that these inconsequential differences would cease. As is, such slight differences would be difficult to detect in an image.

 

DOF for macro is a function of magnification and aperture only. Because of that, differences between DX and FX don't exist until you get to high magnifications. This has been discussed at WP at length with the relevant math. If WP isn't a sufficient source, go to your favorite photo site and study DOF. This is not controversial.

 

But not the same DOF (not to mention that F11 as an initiation point for compromise resulting from diffraction on a DX sensor is a bit of a red herring)

Yes, with the same DOF within the crude estimate of crop factor equaling 1 stop. f/11 is not a red herring for DX, it is simply a calculation of where certain DX sensors offer lines/mm that match diffraction. It is not a hard cutoff, it is simply an interesting number and useful when comparing cameras.

 

Given the way loftus asked the question, any other way of responding would not relate directly to the inquiry. If two cameras, one DX with 60mm lens and one FF with 90 mm lens shot the same image at the same distance would the dof be the same. The answer obviously is no. And opening up the DX in order to keep the results of diffraction not visible is pertinent only above F22.

No that's wrong. Diffraction doesn't only appear above f/22, it exists at all apertures. If a photographer arbitrarily chooses f/22 as his limit for DX then he could shoot FX under identical conditions up to f/32. If you are declaring f/22 the limit for both formats then you are applying different standards to the two formats.

 

Because DX renders greater dof field at F22 than FF does at F27 and F32, and above F22 is the point where refraction becomes evident (or should I say the results of diffraction) on a DX sensor, there doesn't seem to be a host of reasons why anyone would want to above F22 very often. 1-1/2" of dof is pretty significant at 6".

Rob, you really need to develop a better understanding of diffraction.

 

Plus, the viewfinder on a DX camera begins to get a bit dark, at least where it is evident to my eyes, at F22 and above.

Cameras don't work this way. The aperture you choose for a shot has no effect on the brightness of the finder.

[craig 0]

Less depth of field with large formats like 6 X7 is why these formats never were used much by U/W film photographers.

I don't agree with that statement because it is misleading. Large formats were not developed to produce small enlargements nor to shoot small subjects so the lenses made for those formats did not offer small enough apertures nor large enough magnifications. Furthermore, large format cameras would be enormous when housed and require enormous light. There are many reasons why such equipment was not of interest underwater and DOF with existing lenses was certainly one of them, but comparable DOF was certainly possible even if impractical. The same issues are true today with the H3D underwater. It lacks interesting macro lenses because the camera isn't intended for that use. It is suitable for UW within the limits of the lenses it does offer.

 

The attached charts show DOF for the three 24 degree lenses when focused at one foot. You will see that the 4/3 camera gains two F/stops over the 35 mm sensor, e.g. at F/11 the 4/3 camera has about the same DOF as the 35 mm camera has at F/22. I fail to see this as a bad thing because we all know that a lens will preform better near F/8 than near F/32.

I don't agree with your comment regarding the classic f/8 myth but it is irrelevant. You are right, though, about f/11 on 4/3 being comparable to f/22 on 35mm. It can also be said that f/11 is a practical upper bound for 4/3 if f/22 is practical for 35mm. Such reasoning would lead you to conclude that DX would be limited to f/16 roughly. This would be an arbitrary definition though a common one.

 

Curiously, f/22 was a common upper limit for shooters of 35mm film yet somehow that didn't translate to f/16 when those same shooters adopted DX. It sure should have.

 

I to have heard the diffraction argument and understand it, but like Alex and others it has little relevance in practice because I have no reason to need to shoot a such high F/numbers and when I do I'm not shooting test charts.

Hopefully you aren't suggesting that test charts cause diffraction to occur when otherwise it wouldn't? I'm not sure where you are headed when declaring that diffraction doesn't matter to you but I hope it's not toward concluding that somehow formats differ in that respect. I don't find it at all surprising that you or Alex or others find diffraction not of great concern, I don't either, but I would be shocked if that translated into a claim that diffraction issues are irrelevant when comparing formats. Diffraction is of equal concern to all formats.

 

The point is moot. If you are happy with a range of apertures beneath some arbitrary cutoff you will be equally happy with another range in another format assuming that such a range is offered. This is all a zero sum game. All you are doing is changing multipliers and effecting the underlying mechanisms not one bit.

 

As the format gets larger, the f-number must become greater to keep the DOF the same. When that occurs, more light is needed for the exposure and that is expected because the format has more area to expose. This is all precisely in proportion and it's really as simple as that. Depth of field is not in any way a function of sensor size.

[loftus 42]

I am going to change the subject, but ask another question that pertains to dx vs fx, now that I have been educated on DOF and diffraction.

I have ordered a D700, I am not sure yet whether I will even try to house it for underwater use; but can I expect my 17-35 lens to behave the same behind my 8" Subal dome as it does with my D200. It appears that trying to get the 14-24 to work well underwater may be difficult, but I am very pleased with the performance of my 17-35 on the D200. Can I expect similar results with FF, and if not, why not.

I'm on my 2nd Bunnahabn Single Malt and getting the fireworks ready. Happy 4th everyone!

[craig 0]

Some good reading on the subject: http://www.clarkvision.com/photoinfo/dof_myth/index.html

A commonly cited advantage of smaller digital cameras is their greater depth-of-field. This is incorrect.

 

The myth, simply stated, is: smaller digital cameras have a larger depth-of-field than larger digital cameras.

The simple reason why the myth is incorrect is that depth of field is set by aperture, focal length, and a criterion for spatial resolution, and if one keeps aperture of the larger camera the same as that in the smaller camera, the two cameras record the same image with the same signal-to-noise ratio and the same depth of field with the same exposure time.

Not only does Clark assert all the same points I have, though through different language, he also explains how exposure and quantity of light tie in as well.

[BottomTime 4]

I am going to change the subject, but ask another question that pertains to dx vs fx, now that I have been educated on DOF and diffraction.

I have ordered a D700, I am not sure yet whether I will even try to house it for underwater use; but can I expect my 17-35 lens to behave the same behind my 8" Subal dome as it does with my D200. It appears that trying to get the 14-24 to work well underwater may be difficult, but I am very pleased with the performance of my 17-35 on the D200. Can I expect similar results with FF, and if not, why not.

I'm on my 2nd Bunnahabn Single Malt and getting the fireworks ready. Happy 4th everyone!

Your 17-35 will perform identically on the D700 as is does on the D200 for the area that is coincidental to both sensors. However, the area covered by the D700 sensor will expand out further to the edge of the image circle and performance there is not as good as it is in the center of the image circle.

On my F90x, I had problems with corner sharpness on the Subal 8" dome/17-35mm combination. Corners were brutal at F4 and below, soft at 5.6 and pretty good by F8 (17mm end of the zoom with diopter). However, that was on film which has significantly less resolving power than the current crop of digitals so you might find you need even tighter apertures to get good corner performance, but that is very subjective.

On the plus side, getting smaller apertures with this sensor should be a lot easier than it is on film given the lowlight capabilities this sensor has.

[Rob Kille 0]

Without focus distance, who knows where these numbers come from or how accurate they are.

I am confident you understood from my previous comments that distance was set at 6". And that they are accurate. But if you question them please feel free to reference the website your clarkvision article cited http://www.dofmaster.com/dofjs.html , it's the same one I showed the link to in my first post earlier today

 

If you wish to grind numbers rather than study the underlying technology then you can let numbers deceive you. If you understand the math then tables aren't necessary.

I don't care about the underlying technology (which is not the same as saying I don't understand it), I only care about what the technology renders. And in the case of a DX with a 60mm lens vs. a FF with a 90mm lens, both at identical distance, aperture, shutter speed and ISO, the DX renders a greater dof. There is absolutely no legitimate argument to the contrary. None.

 

f/11 is not a red herring for DX, it is simply a calculation of where certain DX sensors offer lines/mm that match diffraction.

It certainly is a red herring, and not only given your use of it in the comment I replied to but also because diffraction can vary dependent upon the lens in question.

 

Diffraction doesn't only appear above f/22

As you know, I never said it did. I said above F22 is the point where on a DX sensor with 60mm lens it becomes evident enough in the image to be concerning.

 

Regarding your clarkvision article: it's quite the pedantic dissertation, don't you think? But it is no revelation. All he says is that there is a two stop disparity in dof between two cameras in his example "Another Way to Look at the Problem" and the bigger sensor camera has to decease light via shutter speed, aperture or iso, in order to make the images from each camera identical. And to offer support for his premise (that the 2-stop disparity exist before the adjustments to decease light) he cites the very same DOF calculator I provided a link to earlier today and you dismissed as the need to "grind numbers". And in most cases, forcing a decrease in light that is not necessary in the DX camera results in limiting the FF photographer's breadth of options.

 

The bottom line is this: same fov and same settings on lens/camera on both a DX and FF camera results in a 1 to 2 stop dof advantage for the DX. Usually 2-stop, particularly at minimum focusing distances.

 

Hey, despite your rudeness (why did you go there, by the way?) it's been fun discussing this with you and I wish you a happy and safe 4th of July.

[loftus 42]

Your 17-35 will perform identically on the D700 as is does on the D200 for the area that is coincidental to both sensors. However, the area covered by the D700 sensor will expand out further to the edge of the image circle and performance there is not as good as it is in the center of the image circle.

On my F90x, I had problems with corner sharpness on the Subal 8" dome/17-35mm combination. Corners were brutal at F4 and below, soft at 5.6 and pretty good by F8 (17mm end of the zoom with diopter). However, that was on film which has significantly less resolving power than the current crop of digitals so you might find you need even tighter apertures to get good corner performance, but that is very subjective.

On the plus side, getting smaller apertures with this sensor should be a lot easier than it is on film given the lowlight capabilities this sensor has.

Thanks. kind of what I figured, just hoping for better.

[Paul Kay 65]

but can I expect my 17-35 lens to behave the same behind my 8" Subal dome as it does with my D200. It appears that trying to get the 14-24 to work well underwater may be difficult, but I am very pleased with the performance of my 17-35 on the D200. Can I expect similar results with FF, and if not, why not.

I'm on my 2nd Bunnahabn Single Malt and getting the fireworks ready. Happy 4th everyone!

 

Hi loftus

 

You have to think about underwater optical systems in terms of their Field of View (FoV), NOT focal length. In this case you have an identical set up (dome, (diopter?), lens) operating on DX - with a reduced FoV - and on FX (or FF) with a greater FoV. As I have posted many times on wetpixel, when the field of view exceeds about 90 degrees (20mm on FX), the curved virtual image is of sufficient 'depth' to mean that it cannot be all retained in focus at wider apertures (ie anything wider than f/8 to f/11) when the in-camera image is projected onto the sensor. At the same time you are using a wide-angle lens to focus relatively closely - which has always been where many wide-angle's performance has deteriorated. Modern designs often incorporate floating element designs to improve performance at closer distances, but zooms are unlikely to be as well corrected as fixed focal length lenses. So in essence you are operating a lens where it is least well optimised, trying to photograph a curved virtual image with a system actually designed to work best with planar images and added to this either one (the dome) or two (dome and diopter) parts of your optical system are 'simple' lenses (ie they are a straightforward uncorrected lens which, whilst they do an adequate job, have never been optically corrected to achieve any optimisation with any specific lens).

 

So all this means that you are unlikely to achieve reasonable corners without using a small aperture.

 

On this note it might be worth saying that since we are using optical systems (whether dome or flat ports) which create problems anyway (ever noticed the chroma around high contrast subject matter in the corners of underwater macro images?), diffraction might in some cases be the straw that breaks the camel's back, and certainly I've noticed the effects of degraded images at very small apertures which I've put down to diffraction, but not all the time - it seems to depend on other factors perhaps as focussed distance and subject matter possibly too.

 

Go back to the Islay Scotch before thinking about it - I'm sure it will help!

[loftus 42]

Hi loftus

 

You have to think about underwater optical systems in terms of their Field of View (FoV), NOT focal length. In this case you have an identical set up (dome, (diopter?), lens) operating on DX - with a reduced FoV - and on FX (or FF) with a greater FoV. As I have posted many times on wetpixel, when the field of view exceeds about 90 degrees (20mm on FX), the curved virtual image is of sufficient 'depth' to mean that it cannot be all retained in focus at wider apertures (ie anything wider than f/8 to f/11) when the in-camera image is projected onto the sensor. At the same time you are using a wide-angle lens to focus relatively closely - which has always been where many wide-angle's performance has deteriorated. Modern designs often incorporate floating element designs to improve performance at closer distances, but zooms are unlikely to be as well corrected as fixed focal length lenses. So in essence you are operating a lens where it is least well optimised, trying to photograph a curved virtual image with a system actually designed to work best with planar images and added to this either one (the dome) or two (dome and diopter) parts of your optical system are 'simple' lenses (ie they are a straightforward uncorrected lens which, whilst they do an adequate job, have never been optically corrected to achieve any optimisation with any specific lens).

 

So all this means that you are unlikely to achieve reasonable corners without using a small aperture.

 

On this note it might be worth saying that since we are using optical systems (whether dome or flat ports) which create problems anyway (ever noticed the chroma around high contrast subject matter in the corners of underwater macro images?), diffraction might in some cases be the straw that breaks the camel's back, and certainly I've noticed the effects of degraded images at very small apertures which I've put down to diffraction, but not all the time - it seems to depend on other factors perhaps as focussed distance and subject matter possibly too.

 

Go back to the Islay Scotch before thinking about it - I'm sure it will help!

I know this is probably a very dumb question, but if we have so many problems with domeports and wide rectilinears, why not construct a larger flatport that fits very close to the lens? Obviously it will only work with lenses with internal zooming and minimal change of length with focusing.

[Paul Kay 65]

I know this is probably a very dumb question, but if we have so many problems with domeports and wide rectilinears, why not construct a larger flatport that fits very close to the lens? Obviously it will only work with lenses with internal zooming and minimal change of length with focusing.

 

Because of refraction. A flatport has to have sufficiently thick glass/whatever to withstand the pressure. A thick flat port introduces chromatic aberration (colour fringing), astigmatism and distortion. The chromatic aberration affects resolution so you've swapped one set of problems for another - flat ports are ok down to about 35mm focal lenght on FX/FF. I have read that a two element achromatic flat port might improve results but I'm not sure that anyone has tried building one!

 

Also and probably far more importantly, you lose FoV with a flat port so you'd need even wider lenses!!!

 

What we really need are aspheric domes but these might have to be designed for each lens/focal length and would be very expensive to design and no one has thought it cost effective so far.

[craig 0]

And in the case of a DX with a 60mm lens vs. a FF with a 90mm lens, both at identical distance, aperture, shutter speed and ISO, the DX renders a greater dof. There is absolutely no legitimate argument to the contrary. None.

Nonsense. The problem here is that you are talk-only, no listen.

 

If by "same aperture" you mean f-stop then the comparison is meaningless because you are applying different standards to the two formats. If you mean physical aperture then the reality is that you cannot set real 60mm and 90mm lenses to matching physical apertures. Lenses have discreet aperture values and those focal lengths don't correspond to 1/3 or 1/2 stop increments.

 

You cannot compare different format sizes using identical f-numbers and conclude anything useful regarding depth of field. Different formats require different f-number ranges. That's the whole point. Yes, f/16 produces more DOF on DX than it does on FX. So what?

 

It certainly is a red herring, and not only given your use of it in the comment I replied to but also because diffraction can vary dependent upon the lens in question.

And how is that? Diffraction is lens dependent? If so, then why does it consistently appear only at f/22?

 

As you know, I never said it did. I said above F22 is the point where on a DX sensor with 60mm lens it becomes evident enough in the image to be concerning.

Concerning to whom and under what circumstances? To you diffraction is like an on/off switch. If only it were that easy.

 

Why do you think point and shoot cameras don't offer f/22? Because their lenses are diffraction-limited far below that, that's why.

 

Regarding your clarkvision article: it's quite the pedantic dissertation, don't you think? But it is no revelation.

If it is no revelation then you then why are you vehemently arguing against it? I would say that the article would be a great revelation to many people if they took the time to understand it.

 

All he says is that there is a two stop disparity in dof between two cameras in his example "Another Way to Look at the Problem" and the bigger sensor camera has to decease light via shutter speed, aperture or iso, in order to make the images from each camera identical. And to offer support for his premise (that the 2-stop disparity exist before the adjustments to decease light) he cites the very same DOF calculator I provided a link to earlier today and you dismissed as the need to "grind numbers". And in most cases, forcing a decrease in light that is not necessary in the DX camera results in limiting the FF photographer's breadth of options.

I'm not surprised the article went over your head but your summary is grossly flawed. As for dismissing his DOF calculator, I did no such thing. I dismissed the ridiculous need to compute specific numbers, and to quote them without context, when it is entirely unnecessary. I don't need a few random numbers that explain a small piece of the story when I can apply a little thought and some 9th grade algebra and reach the destination all at once.

 

What Clark said is what I quoted the first time. It is unambiguous:

 

"A commonly cited advantage of smaller digital cameras is their greater depth-of-field. This is incorrect. ... depth of field is set by [physical] aperture, focal length, and a criterion for spatial resolution"

 

Got it? You are incorrect. The Clark article explains this.

 

The bottom line is this: same fov and same settings on lens/camera on both a DX and FF camera results in a 1 to 2 stop dof advantage for the DX. Usually 2-stop, particularly at minimum focusing distances.

Perhaps you need to read the Clark article again only this time actually seek the revelation. You are free to believe anything you like but don't go teaching it.

 

Rob, when you say things like "F22 is the point where on a DX sensor with 60mm lens it becomes evident enough in the image to be concerning", "diffraction can vary dependent upon the lens in question", and "the viewfinder on a DX camera begins to get a bit dark, at least where it is evident to my eyes, at F22 and above" it calls into question whether you really understand what's going on at all. I don't care whether you think I know anything or not, but you should care that you are arguing a point in direct conflict with a recognized expert in the field and one who's explanation you've labeled "pedantic". If the article were so laboriously detailed, why have you failed to understand it?

[craig 0]

Also and probably far more importantly, you lose FoV with a flat port so you'd need even wider lenses!!!

And that's the real killer. Flat ports are limited to about 100 degrees of FOV and they cause increasing pincushion distortion as the FOV widens. They are kind of like a reverse fisheye lens.

[vazuw 1]

This may be a stupid question, but are the technical differences reviewed in this discussion visible on print of say 16x20, or file viewed on an average monitor at an average viewing distances.

[Paul Kay 65]

This may be a stupid question, but are the technical differences reviewed in this discussion visible on print of say 16x20, or file viewed on an average monitor at an average viewing distances.

 

Depends on a lot of things. Sometimes but not often IMHO. You have to be producing a pretty good print for a start! There are so many variables that unless you are working to achieve the best quality throughout your entire workflow its probably difficult to determine where any visible problems arise. Suffice it to say that if you are operating to achieve the best image quality throughout (including not using apertures where diffraction might be a problem) then you might notice a difference if you shoot like this and further stopped down. But you MUST operate 'best practice' throughout. Somewhere on wetpixel there has been discussion of webimage vs. print image before. Producing high quality prints means being painstaking throughout the entire photographic process especially if images are to be printed at high enlargement. Bottom line is (as I've commented before) that photography is a practical process. Go and shoot images at different apertures and compare prints - at the end of the day its the only way that you will appreciate the REAL differences in IQ - pixel peeping won't help greatly here!

[vazuw 1]

Thanks. I think thats the bottom line to this discussion. Maybe in the near future we can see some real practice comparisons by you techies

[Paul Kay 65]

vazuw

 

I have prints (20" x 16") here in North Wales, but I can't post them on the web and unless you are close by........

 

And that is the problem. Showing details of images on the web is useful - up to a point. But there are so many variables and what is acceptable to one person may not be to another, and arguing over nuance when it is viewed on an unknow computer set up is not viable.

 

I'd say that we have better underwater photographic tools than ever before, but equally we have to use them to their full potential and this is not as easy as it sounds to do so consistently and under varying conditions. We are probably starting to discuss relative nuances with regard to topics such as diffraction limitation to a great extent. There are areas which we can and no doubt will improve and which are real world problems (rectilinear wide angle corners!) but for the most part we are pretty well off. Its time to start upping image content quality!!!

[james 0]

OK - I get to actually participate in a meaningful way on this one! Craig and I both used the Nikkor 17-35 underwater with the Kodak ProSLRn - the first FF f-mount camera.

 

I was QUITE happy with the topsides performance of the 17-35 and don't see any reason to go for the 14-24 - but that's just me. The 17-35 does have pretty strong vignetting but that's easily corrected in post.

 

Underwater the 17-35 was pretty decent behind the Aquatica 8" acrylic dome port. I have a few pix online somewhere here - from way back in 2005 when I took it diving.

 

Cheers

James

[vazuw 1]

vazuw

 

I have prints (20" x 16") here in North Wales, but I can't post them on the web and unless you are close by........

 

And that is the problem. Showing details of images on the web is useful - up to a point. But there are so many variables and what is acceptable to one person may not be to another, and arguing over nuance when it is viewed on an unknow computer set up is not viable.

 

I'd say that we have better underwater photographic tools than ever before, but equally we have to use them to their full potential and this is not as easy as it sounds to do so consistently and under varying conditions. We are probably starting to discuss relative nuances with regard to topics such as diffraction limitation to a great extent. There are areas which we can and no doubt will improve and which are real world problems (rectilinear wide angle corners!) but for the most part we are pretty well off. Its time to start upping image content quality!!!

Edited July 4, 2008 by vazuw