• Please use real names.

    Greetings to all who have registered to OPF and those guests taking a look around. Please use real names. Registrations with fictitious names will not be processed. REAL NAMES ONLY will be processed

    Firstname Lastname

    Register

    We are a courteous and supportive community. No need to hide behind an alia. If you have a genuine need for privacy/secrecy then let me know!
  • Welcome to the new site. Here's a thread about the update where you can post your feedback, ask questions or spot those nasty bugs!

Pre-planning Photography I: A look at "Output Quality"

Asher Kelman

OPF Owner/Editor-in-Chief
We see some especially beautiful, well-composed and exposed images presented here, mostly from digital cameras. I often wonder how large a picture from a camera with a 5 micron sensel size in 35 mm full frame format gets printed when the exposure was at f16.

Yes the DOF increases but the value of having 24 MP might be decreased to about half or more. Perhaps in doing architecture, it doesn't matter where there's little fine granular detail to be conserved and shown. Of course a 6MP digicam could be used for a giant billboard viewed 100 meters across a freeway. But are we really organized to plan our shots with presentation output in mind? I'm seeing more than a few highly competent colleagues using their Hassleblads or DSLRs as if the f stops didn't cost anything in quality as the DOF increases.

So what limits do you give yourself for the pictures you shoot? I generally try to keep below f8.0 for a DSLR and generally less than 5.6 so I can print as large as possible if I so choose. Of course we're still limited, but how often is DOF so much more important than quality?

Incidentally, we accepted the notion of stitching adjacent overlapping images to gain more pixel "real estate" for a panoramic view. But why not deal with depth too? Why don't we do more focus stacking, especially in architecture?

Asher
 

Nicolas Claris

Administrator/Moderator
Incidentally, we accepted the notion of stitching adjacent overlapping images to gain more pixel "real estate" for a panoramic view. But why not deal with depth too? Why don't we do more focus stacking, especially in architecture?

Asher
Simply because we need to win our food and cannot spend hours in PP that most of clients are not ready to pay for.
If a large sensor camera is able to deliver large size image and details, one will prefer to spend some more thousands Euros/Dollars to get the result needed in one go.
If our client request large prints to be seen able from affect or two, they're ready to pay for a MF shoot, but you just can't talk stitching or stacking, this is not their business…
Also, this is just not credible to imagine sticking or stacking a shoot of a moving subject like a model, a car, a boat…
 

Doug Kerr

Active member
Hi, Asher,

Thank you for framing this important issue for discussion here.

We see some especially beautiful, well-composed and exposed images presented here, mostly from digital cameras. I often wonder how large a picture from a camera with a 5 micron sensel size in 35 mm full frame format gets printed when the exposure was at f16.
Indeed. At f/16, the "diameter" of the Airy circle from diffraction is about 21 microns (about 4 pixel pitches for your example). Note that the format size is not involved in that. (That of course does influence what the pixel dimensions of the image are.)

Yes the DOF increases but the value of having 24 MP might be decreased to about half or more. Perhaps in doing architecture, it doesn't matter where there's little fine granular detail to be conserved and shown. Of course a 6MP digicam could be used for a giant billboard viewed 100 years[Meters} across a freeway.
100 years? That's a pretty wide freeway.
But are we really organized to plan our shots with presentation output in mind? I'm seeing more than a few highly competent colleagues using their Hassleblads or DSLRs as if the f stops didn't cost anything in quality as the DOF increases. ...
I am especially interested in seeing the responses from the members to the question you posed that I have highlighted above in blue.

Best regards,

Doug
 

Asher Kelman

OPF Owner/Editor-in-Chief
Nicolas,

Let me state that I've seen one "guru" who used to use 4x5 film now apply the f stops common for that format to MF digital work and f16 is applied without any reference to its effect on resolution. Actually any point of focus is now spread in "ripples" of photon spread across neighboring pixels. The tiny pixels size is able to record this spread of one point to others and so the image is degraded. Certainly, for landscapes or architecture, focus stacking should be comfortable, routine and de-rigueur‎ for those of us wanting to make super large prints and wanting DOF.

Of course, one cannot easily use focus-stacking when people are involved. Also one needs to be economical or no client will pay for hours of work. However perhaps one could batch "focus-stack" each position in a scene first. Then we'd use the new version of Autopano Giga, where people can be either added or ignored preferentially.

Asher
 

Asher Kelman

OPF Owner/Editor-in-Chief
If needed, would you stack such image ?


Nicolas,

How fair is it to show a landlubber such a startlingly impressive and successful shot? But that's your work! A picture with such breadth and drama is hard to improve on. The asymmetrical composition shows the live moving water, a serene skyline, sailors busy and relaxing as the sleek modern boat, barrels towards us, so close by and barely misses us as it skims past on its way!

In such a picture, the action and spectacle make it so powerful and attractive.

We don't need to look for fine details or exact proportions as this is happening at great speed and that's all part of the experience. Even the anamorphic distortion of the human heads is not at all important, where the action, rare sights and movement dominate so! This is, after all, an almost "imperial" picture, the pinnacle of the rich man's nautical achievements over the sea and the rest of us!

But imagine for the moment a new important "wind-speed, force and navigation" instrument that has to be mounted very close to the deck and to your left. The builders want to show off this device which helps to trim/adjust/set the new wing-sail. Of course this would be a challenge to most other photographers. But you, uniquely are not any ordinary professional photographer. Few have so much experience, skill and reflexes to be so "at home" on the boat, responding to feel of the sea beneath the boat and knowing instinctively what you'd want to add, in perfect focus.

If this was indeed something detail-rich and so critically important, it's just a matter of grabbing more real estate of the scene with a new focus point. So you'd take two successive second and third shot focussed to the left, overlapping and progressively covering the new important area of interest. Before stitching, you'd mask out everything distant and to the right, like the sea and the other sailors in one broad sweep. Very simple, as the deck and windows are so well defined with a lot of corners, edges and contrast, would be easy for Autopano Giga to align and stitch.

Whenever depth changes are so asymmetrical like this with the distant see and skyline bg. above or to one side, "focus-stacking" by simple panorama-stitching is a "cheat" one can often do, even handheld "on the fly".

At home, one would pick which sets work best together. But it does require the extra work of masking out the moving sea and sky.

Sometimes, for sure, you'd need 30 minutes of fine work, repairing some double cables, close by, where stitching was faulty because of parallax errors. However, nothing as challenging as taking your original picture in the first place. I would have great difficulty wrangling a heavy camera as the Pentax 645D in such a dancing and swaying boat. However, the stitching and repair would be much less of a challenge.

Your resulting image, if successful, would be even more magnificent and I'm sure you'd deliver that with aplomb and not a bead of sweat!

Asher
 

Asher Kelman

OPF Owner/Editor-in-Chief
Nicolas,

I myself, don't always remember update my ISO to optimize image quality. Sometimes coming from a dim interior, I'll grab shots with the Ricoh GR at an unnecessarily high ISO sacrificing color depth, adding unneeded noise and reducing DR.

Here, I believe, use of a (6x6 micron sensel)- MF camera at f11 and 1/320 at ISO 100 results in having essentially a camera with effectively towards 25MP approx down from the nominal 40MP of the sensor array! With the 6x6 micron size pixel, a setting of the aperture opened a stop to f8 is needed to use all of the Pentax 645D's 40 or more MP. At f11 a sensor of 40-80 MP can show up only as merely about 25 MP. The question is does one need the extra resolution and is f11 required to get the DOF one wishes? As the sensel size gets smaller, we do get a tad more leeway, but would still loose ~ 5% of effective pixels at f11 for the 645Z which has pixels at 0.51 x .51 microns. These are approximations. No doubt, Bart could provide more accurate figures.

Likely as not, you use f11 with the Pentax 645D as it seems to work well for you. However, at f8, you might also have sufficient DOF to your liking, get a higher shutter speed, (and still be able to grab something close with an overlapping coverage, if that was important enough).

Have you actually tried f8 and found it came short on DOF? Have you also discovered that you really don't need all those pixels for output in a book as opposed to a 3 meter wall image? In that case, sacrificing effective pixels for DOF is a fair bargain!

I suspect that for your books, the f11 shot gives you the tad extra DOF and printing it to a page at 300 DPI means that you are generally not short of pixels anyway! However, if the successful image is also going to be used in a trade show 3 meters high and viewed up close, then it might be a strong case for trying to work at f 8.0.

Just some thoughts prior to my own jumping into the luxury world of MF digital. Seeing what choices others make is helpful!

Asher
 
We see some especially beautiful, well-composed and exposed images presented here, mostly from digital cameras. I often wonder how large a picture from a camera with a 5 micron sensel size in 35 mm full frame format gets printed when the exposure was at f16.

Yes the DOF increases but the value of having 24 MP might be decreased to about half or more. Perhaps in doing architecture, it doesn't matter where there's little fine granular detail to be conserved and shown. Of course a 6MP digicam could be used for a giant billboard viewed 100 meters across a freeway. But are we really organized to plan our shots with presentation output in mind? I'm seeing more than a few highly competent colleagues using their Hassleblads or DSLRs as if the f stops didn't cost anything in quality as the DOF increases.

So what limits do you give yourself for the pictures you shoot? I generally try to keep below f8.0 for a DSLR and generally less than 5.6 so I can print as large as possible if I so choose. Of course we're still limited, but how often is DOF so much more important than quality?
Hi Asher,

It's a hard question to answer in general. It probably boils down to; do we know in advance how our images might need to be repurposed for other sizes than the ones we initially envisioned?

Things get pretty easy, pretty fast, when we only produce images for web sized output. In fact, we'd probably have to struggle to have a limited depth of field at small output sizes, unless we used a 'small' sensor with a short focal length. Small is a relative term, which requires specification to put it into context. However, practical experience learns that clients may want to repurpose images at a larger than initially agreed size. Whether that also requires a larger input file size (in pixels, and/or per pixel quality) is undetermined, as yet. For all we know things will not get anything more challenging than a billboard, to be viewed from a hundred feet away, or on the side of a truck (while potentially viewed from a closer distance, not allowing the same time to gulp in the detail).

So, my vision is that we may be confronted with 'terra incognita', a situation outside of our comfort zone (i.e. the requirements for input quality we already know intimately from experience). In that case we could use a bit of help in translating the experience into new types of expression (with a required level of input quality). It's mostly in those types of situations that my DOF output quality tool might be helpful. It will also allow to 'previsuallise' the chance of pulling off the job as required with a given set of equipment. And if the equipment at hand won't suffice, what will be needed (as in new investments or hired equipment), to achieve what's requested. and create a longer term commercial relationship with our customer/audience.

Incidentally, we accepted the notion of stitching adjacent overlapping images to gain more pixel "real estate" for a panoramic view. But why not deal with depth too? Why don't we do more focus stacking, especially in architecture?
This is part of an arsenal of available solutions, including tilted lens configurations, to stacked images. The reasonable (and practically achievable) approaches can be estimated with my free tool. Focus stacking will require some experience (and tools) to achieve successfully though, but what's new with that? Good quality photography requires practice ..., and hard work for most of us, except for the insanely talented amongst us ...

Cheers,
Bart
 

Nicolas Claris

Administrator/Moderator
Asher
On this particular images I wished to have the blocks (rolling blocks for ropes) on the left almost as sharp as the skyline.
This may be possible @ ƒ8 but would need the focus point to be particularly precisely placed.
In this condition of shooting, I cannot take my iPad or iPhone (or Android item) to calculate the focus point distance, then go from my place with a ruler to set the required point, then go back to the shooting place.
Adjust focus point with the camera and then move the camera with shutter half pressed to the desired composition.
In that the 645Z has a major improvement with more focus point than the D.
It is much easier and safer to shoot @ ƒ11
Same applies for food photography with the difference that one is shooting at a very short distance and one would even need more precise more precise focus point.
Precision needs time, and if you take too much time, the cream gets flat!
So in this case (using a tripod) I shot at ƒ16.
Show me the "broken" pixels, I can enlarge these as much the client would ask for…

Sometimes one have to hear its experience to make the right decision instead of calculating the size of the pixel and the related DOF…

Also, I NEVER know in advance which shot the clients will choose to have printed at 4 meters wide (the new standard since I use the Pentax 645D)…
 
Last edited:

Andrew Stannard

pro member
Hi,

Firstly, Nicolas - your shots always make me envious! What a fantastic experience being on a large sailing boat seems to be.

But to answer Asher's questions.....

If I'm taking family snapshots then I rarely worry about exactly what setting I'm using. Normal use is for either web or small prints, and so it doesn't matter so much. Often I'll shoot almost wide-open, but for artistic rather than quality reasons.

If I'm doing more serious landscape work then I like to think I do try to optimise as much as I can. Where possible I'll be tripod mounted, mirror lock, low iso, graduated filters if required and for full frame normally I'm happy to go up to around f11 before I think I can start to detect the effects of diffraction (5D MkII). In truth I haven't experimented much with DoF stacking. No real reason apart from the fact I get a certain satisfaction from capturing everything in one shot.

Often though I find some sort of other limiting factor comes into play. If I'm up in the mountains then perhaps I don't have a tripod, or the light is such that there is no time to set it up. In those circumstances maybe I need to compromise to be able to hand-hold. So what to do? Increase ISO and introduce some noise, or maybe open the aperture and compromise DoF? Of course technology helps us here - modern sensors have excellent ISO performance, modern lenses have great stabilisation.

I actually enjoy this part of photography - taking into account all the variables and deciding upon the optimum approach to maintain quality without compromising the artistic intent.

All that said, if I were on that boat then I suspect I'd forget all that- would be too busy holding on and trying to enjoy the view!


Regards,
Andrew.
 

Andrew Stannard

pro member
Just to add to my previous reply....

I always consider one of the best photographers for juggling iso, f-stop, shutter etc etc to be the late Galen Rowell. His book 'Mountain Light' has always been one of my sources of inspiration.

Most of his picture's aren't what many people would call 'technically perfect', but the artistic vision behind them is superb. Often shooting in extreme conditions, and without the aid of modern technology, it was a real balancing act to get the shot he was looking for.

If you've never read the book it's well worth getting hold of a copy.


Regards,
Andrew.
 

Doug Kerr

Active member
We have seen a better hammer, but everything is not a nail.

Over the years, different photographers have had differing awareness (if any) of the implications of diffraction on the final image result, and take differing account of that (if any) in their shot planning.

In this forum, that matter is from time to time called to our attention in various useful ways. For one example, Bart van der Wolf's excellent Depth of Field Output Planning Calculator helps us to take that into account, and we have seen several essays on the how the implications of diffraction (and perhaps their tension with our depth-of-field aspirations) may differ as we go to larger format sizes.

Recently our host, Asher, in part apparently in the course of looking into how he can best optimize the process of generating "life-size" printed images of people, has had his attention to the matter of diffraction heightened, and in turn has called our attention to it.

Asher and I are very much alike in at least this way: we learn something, and we want to share it.​
One aspect of this is an urging to pay more attention to depth of field in shot planning than may have been our custom.

But we have heard of many situations where various factors may make it impractical for us to take advantage of our possibly-heightened awareness of the impact of diffraction.

And thus it is with all tools. As a "new" tool comes to our attention, offering in some cases the prospect of better results, we do not need to look urgently at every turn for a way to apply it. I may be gratified to receive for Christmas an adjustable open-end wrench with motor-driven jaws (yes, this wretched excess is true), but I still work brass fittings with a classical open-end wrench (or maybe even a fitting wrench, if needed, and if it will "fit" in the workspace).

Let's be glad that we have the chance to understand more about the available tools, or to be reminded of what we already know. But everything is not a nail. And that's not a problem.

Best regards,

Doug
 

Andrew Stannard

pro member
We have seen a better hammer, but everything is not a nail.
I like this analogy. Perhaps I can be excused for taking it further...

If we're trying to build the best possible shed that we can, then we really do need to know about our hammers and nails, but of course this is only part of the story. Only by knowing all the tools available to us will we get the best result.

We need to know which bits of the shed to nail together, and in which order (composition?).
We can choose to take our time over each nail, or bang them, in wily-nilly (tripod vs spray and pray?).
We need some reasonably decent wood to start with (equipment?)
Once our shed is built we still need to paint it and decorate it (post-processing?).

Of course some of us might decide to just go out and buy a ready-made shed (point and shoot?).

Building it ourselves offers much more creative potential (DSLR?): http://www.cliveelsdon.co.uk/wp-content/uploads/2012/11/garden-shed.jpg

But of course it could all go horribly wrong... http://www.raiseatoast.ca/wp-content/uploads/2013/10/Cabin-Shed.jpg
 

Nicolas Claris

Administrator/Moderator
Let's be glad that we have the chance to understand more about the available tools, or to be reminded of what we already know. But everything is not a nail. And that's not a problem.
Of course Doug, if one wants to break the rules, you better know these rules!
That comes for composition but for technology as well.
The latter known (or been more or less understood for a guy in the filed like me), then one can chose among the different compromises…
Theories and techs must be understood as necessary, but one needs to take into account the realness of shooting in various conditions.
In the seconds before pressing the shutter, one has to have in mind the wished result and applies the necessary settings taking in account what is controllable and what is not.
At the end, you have to be able to have the shot accepted by the client… And more preferably tp make him glad and happy to have assigned you and no other!
 

Robert Watcher

Active member
Only by knowing all the tools available to us will we get the best result.
If this is being implied in the strictest sense, then it is not an ideology that I subscribe to. "Best results" are totally subjective and can be accomplished without full knowledge of tools and sometimes even technique. As well - from my experience in this field - - - technical best or excellence at the highest standard is seldom the criteria for great photography. But then I may just be comprehending what you are saying, incorrectly or out of context.



In the seconds before pressing the shutter, one has to have in mind the wished result and applies the necessary settings taking in account what is controllable and what is not.
At the end, you have to be able to have the shot accepted by the client… And more preferably tp make him glad and happy to have assigned you and no other!
That is so true. . . even for someone like me who just goes around shooting everything that I find interesting with my travel photography - I am fully aware of where I want to go and what I want out of the photograph before I shoot it. Sometimes there are pleasant surprises where I have something even better than I anticipated. But even my casual fly-by snapshots are thoughtfully produced. With my professional portrait and event work, I take a lot of images, but that is so I have a lot of variety to tell a story - - - all variances are fully planned and executed as I want them. The variations that are thrown away still suit my standard and may be saleable on their own but just don't fit into the final project requirements.

As for output - - - I am far more concerned that my image will provide a usable print that how it will look on the web. I never worry too much about how web images look as there is no standard that the majority are going to be using to view the images. The same photograph will be too light or too dark or have subtle differences in colour and tonality than what I might want to display. I have no control over peoples monitors, but I do have full control over my printing.
 

Asher Kelman

OPF Owner/Editor-in-Chief
Nicolas,

We photographers are both hunters and artists and live in both worlds. Works of art are made of an iterative "gestalt" from all the artist's senses and mind guiding his/her hands. I'd not tell Michelangelo that he could make a better and taller David if he'd only been 3mm thinner!

A hunter, however, is not creating the deer he seeks. Instead he uses guile, rules of nature, and surprise positioning for his aim in making the kill he needs. That's a beast, small enough to carry home to feed enough people, sufficiently well! The same with a photographer who provides images to satisfy the client's needs!

Your pictures, Nicolas? I'm in awe of them! I don't ask you to take away one parameter over another to make something I'd value more. I've made that mistake already! When my youngest son, Emile, was just 10, we did a "father and son" art "Dragon" project. I fashioned the wire fame far more then he and he mixed the paints far more than I did ............and it won first prize. He said to me, "But Daddy, in the end, it's not really my dragon!!!" So I'm aware of distorting other folks intentions to meet my own!

Rather, my point here is to make us more aware of the surprising cost of smaller apertures once the resolution becomes limited by diffraction. I was taken back realizing that the Pentax 645D, at f16 with a 6.1 mm sensel is losing up to 23 of the 40 MP "worth" and potential resolution the camera's sensor offers. I'd wager that more than a few here had no idea of such a high price to pay for the gain in DOF from 8.0 to just f11. (With the new 645Z, however, f11 can be chosen with almost no such loss as the pixels are smaller). Of course, when a rare butterfly is flitting past a beautiful flower, I'd not hesitate to trade enlargement potential for just "getting the shot". After all, that's the sine qua non of photography. The rarer the appearance, the more it's all that matters! So we don't really disagree, :)

I do not criticize completed work on choice of f-stop. What's the point for images that are already outstanding and meet the client's needs? None at all! My cautions on tiny apertures (past the onset of diffraction limits), is for the future. Hopefully it might help with our new artistic choices.........within the always limits of that moment!

Asher
 

Wolfgang Plattner

Active member
Hi
Thanks for all these thoughts here ... Good to know and worth to keep it in mind, even though I think, that I, as a hobbyist, will not be touched by the resulting problems on printing output up to DIN A2 ...
 

Nicolas Claris

Administrator/Moderator
Plenty words…
Please, explain me.
Where is the loss of pixels, where is diffraction on theses ƒ16 shots (with a 645D)

All are crops at 100%, sharpened for the web





I need to know in real!
 

Asher Kelman

OPF Owner/Editor-in-Chief
Plenty words…
Please, explain me.
Where is the loss of pixels, where is diffraction on theses ƒ16 shots (with a 645D)

All are crops at 100%, sharpened for the web





I need to know in real!
Nicolas,

These are the most beautiful and exquisite pictures ever! All are superbly illuminated and in perfect focus. The processing is mouthwateringly top rank! The colors are gorgeous and the contrast stunning! No one could do better.

Nevertheless, that doesn't speak at all to the actual resolution quality captured! These pictures hardly have the fine grain detail to easily lose impact printed large. Do this with a newspaper article! At an enlargement where the f16 image will have lost its bite, the f8 image will be high contrast and intact for each letter!

No loss for these stunning food prints shot at f16 for your likely print sizes? It only matters if you wish to print more than a great 12 MP image could give you. [I didn't take into account magnification, so this grim prediction "12MP", as Bart points out, (below in post #26), are in the right direction but over-pessimistic!]

Asher
 

Nicolas Claris

Administrator/Moderator
Asher
I'm on location shooting some interiors of a large catamaran, waiting for the rain to stop!
When I'll be back i'll post a portion of enlargement.
For now please remind that what you see above are crops at 100 % the file being about 45x62 cm @ 300dpi (camera output)...
 

Asher Kelman

OPF Owner/Editor-in-Chief
Asher
I'm on location shooting some interiors of a large catamaran, waiting for the rain to stop!
When I'll be back i'll post a portion of enlargement.
For now please remind that what you see above are crops at 100 % the file being about 45x62 cm @ 300dpi (camera output)...
Well then, Nicolas, I'd better eat my hat! Still, do it, just for me at f16 and f 8 with a newspaper. I'd love to see the results. Meanwhile, I am humbled by your 100% cut outs! I didn't realize that fact!

Amitiés,

Asher
 
Plenty words…
Please, explain me.
Where is the loss of pixels, where is diffraction on theses ƒ16 shots (with a 645D)
Hi Nicolas,

Diffraction is a physical fact of life which we cannot avoid. However, it may have just a limited influence, or an influence that can be somewhat reversed by clever post-processing. Sometimes it is unavoidable to get enough diffraction that will deteriorate image quality, but as a trade-off (e.g. for required DOF) we must accept it.

One of those scenarios is when we shoot close-up images. We will almost always have very limited DOF due to magnification, so we must find a trade-off between diffraction and DOF. As I said in another thread, especially when we have to shoot outside of our comfort zone, it can be useful to do some experimentation/simulation before we engage in the actual shoot.

Compared to wider apertures, f/16 will give additional DOF, but it will also hurt image quality (reduce resolution and micro-detail contrast), it's a physical certainty. However, it may be more tolerable than an image that is too blurry with too little DOF. It's a trade-off, and it also depends on how large the output must be displayed.

You ask where the loss of pixels is, the question can only be answered by comparing to another image that was shot with a wider aperture. However, we can predict how severe the loss is by using a tool like my DOF output quality planner. It can be predicted that at f/16 the larger sensels of the 645D will not suffer too much, but at f/18 you will lose effective resolution (you might as well have used a lower resolution camera, or you need to reduce the maximum output size). So you happened to use the smallest aperture that did not damage resolution beyond repair, but the larger sensels of the 645D helped because they were relatively large compared to the diffraction pattern of f/16.

When you start using the 645Z, you'll find that f/14 is the smallest aperture for a similar shooting situation with a similar per pixel quality as the 645D, because of the smaller sensels of the 645Z compared to the size of the diffraction pattern. However, you may not need the additional pixels for a certain smaller output size, so if you limit the output size, the real loss of capture resolution may still be acceptable.

Of course, when we really need to have even higher image quality, or even more DOF, we can prepare for focus stacking. By being prepared we will not need much time to do it in practice. Especially the Live View capability of the 645Z will allow to do that much easier than with the 645D.

As always, practice makes perfect, but sometimes we do not have the time to practice, and a little planning ahead will then go a long way.

Cheers,
Bart
 
To give a bit more of a practical idea about the effect of diffraction on image quality, here is a small animation which shows the increase of DOF, but also the decrease of mirro-detail contrast and thus pixel resolution at f/5.6 and narrower apertures:



The example comes from another thread about the effects of diffraction in the context of photomicrography.

The exact tipping point where pixel quality starts to deteriorate, depends on many factors, but sensel pitch (and thus sampling density), Aperture, and magnification, are the dominant parameters. Of course when we do not need maximum output dimensions, we can down-sample the image and claim back some effective resolution.

Hope that helps.

Cheers,
Bart
 
No loss for these stunning food prints shot at f16 for your likely print sizes? It only matters if you wish to print more than a great 12 MP image could give you.
Hi Asher,

That's a bit pessimistic...

Simple physics dictates that the 6 micron sensel pitch of the 645D will start to be visibly affected by diffraction at f/6.3 and narrower apertures. At first the contrast of micro-detail will start to deteriorate, but it will still be recoverable to a certain degree, but it will become increasingly more lossy as we continue to stop down further.

Close-up shooting will magnify the negative optical effects at the pixel level. That also means that there will be some leeway if we do not have to produce full size output, or will be viewing from a bit larger distance. My planning tool will tell what to expect.

The large number of pixels from the 645D, and even more from the 645Z, will often allow to even down-sample a bit to achieve the required output dimensions.

There is also the possibility to salvage some of the losses by proper Capture sharpening (e.g. with FocusMagic), and to recover some of the diffraction induced overall lower contrast.

An IMHO perfect tool to regain some structural detail that was dulled by diffraction, is Topaz Labs Detail, currently available with a 50% discount till the end of the month.

Cheers,
Bart
 

Jerome Marot

Active member
We see some especially beautiful, well-composed and exposed images presented here, mostly from digital cameras. I often wonder how large a picture from a camera with a 5 micron sensel size in 35 mm full frame format gets printed when the exposure was at f16.

Yes the DOF increases but the value of having 24 MP might be decreased to about half or more. Perhaps in doing architecture, it doesn't matter where there's little fine granular detail to be conserved and shown. Of course a 6MP digicam could be used for a giant billboard viewed 100 meters across a freeway. But are we really organized to plan our shots with presentation output in mind? I'm seeing more than a few highly competent colleagues using their Hassleblads or DSLRs as if the f stops didn't cost anything in quality as the DOF increases.

So what limits do you give yourself for the pictures you shoot? I generally try to keep below f8.0 for a DSLR and generally less than 5.6 so I can print as large as possible if I so choose. Of course we're still limited, but how often is DOF so much more important than quality?

Incidentally, we accepted the notion of stitching adjacent overlapping images to gain more pixel "real estate" for a panoramic view. But why not deal with depth too? Why don't we do more focus stacking, especially in architecture?

I am not really sure what to make of your questions. You seem to be only concerned with getting everything sharp: as many pixels as possible and maximum depth of field. But in practice and if we have a tridimensional subject (the only ones for which depth of field matters), a limited depth of field is often more an advantage than a nuisance.

You are also taking a 24 mpix 24x36 DSLR as an example. This implies a pixel size of about 6 µm. In practice, the effects of diffraction will not be noted on prints of any size in a camera with that pixel size at f/11 and will be barely noticeable at f/16.
 

Jerome Marot

Active member
Simple physics dictates that the 6 micron sensel pitch of the 645D will start to be visibly affected by diffraction at f/6.3 and narrower apertures.
I think that you are neglecting the effect of the Bayer array, which dictates that the final information is reconstructed from more than a single pixel.
 

Asher Kelman

OPF Owner/Editor-in-Chief
I am not really sure what to make of your questions. You seem to be only concerned with getting everything sharp: as many pixels as possible and maximum depth of field. But in practice and if we have a tridimensional subject (the only ones for which depth of field matters), a limited depth of field is often more an advantage than a nuisance.

You are also taking a 24 mpix 24x36 DSLR as an example. This implies a pixel size of about 6 µm. In practice, the effects of diffraction will not be noted on prints of any size in a camera with that pixel size at f/11 and will be barely noticeable at f/16.
Jerome,

A fair question! Well, there are my on needs ........and then there's my general observation of camera use by us all:

1. In the first instance, except for stitching panoramas, I love having more limited DOF and enjoy peripheral vignetting. But maximum detail retention in chosen portions is paramount.

2. In the latter instance, I just like to remind us all that sometimes, choice of tiny apertures can leading to surprising losses of the quality of an image for reproducing fine detail with good contrast when one needs it for larger prints.

My reference for my discussion are currently as follows:

1. Bart's free Depth of Field output quality planner tool addresses that handily, allowing one to chose the right camera and settings for end use considerations.

2. A discussion of effective pixels versus physical pixels, based on Imatest calculations, especially Table 2. in The Luminous-Landscape,here




I hope this better clarifies the scope of my intent. :)

Asher
 

Doug Kerr

Active member
Hi, Bart,

What a valuable illustration. It so clearly shows the tradeoffs as we change aperture.

Grasping that is our real tool.

Thanks.

Best regards,

Doug
 
Top