Luminous Landscape Forum

Site & Board Matters => About This Site => Topic started by: dreed on March 30, 2014, 09:38:01 am

Title: A Redux Critique Of Contemporary Camera Design
Post by: dreed on March 30, 2014, 09:38:01 am
From the latest article:

"Zoom with your feet and when the limits of that technique have been reached, change the focal length of the lens. Some of the most interesting compositions occur when the photographer is dealing with a found, fleeting moment, and may not be in the perfect position or have the perfect lens mounted, so has to creatively find a way to use all the negative space around the primary subject in the composition. When time isn’t of the essence, the need to physically change lenses for compositional reasons, makes the process more deliberative. You really have to think for a moment about which lens is right for what you want to do. There are two common phenomena that occur when we see something we want to photograph. One is to take in everything the eye sees, which typically results in a photograph that lacks focus. And I’m not referring to something the lens does, but what the mind does. Not everything in front of us is a photograph, even though there is indeed a photograph there before us, somewhere. "

That paragraph, and particularly the last sentence, is the challenge that is photography. Of all the articles I've read here and elsewhere, none have captured it so well.
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: Isaac on March 30, 2014, 10:52:10 am
"Not everything in front of us is a photograph…"

Click! It is now (http://dictionary.cambridge.org/us/dictionary/british/photograph_1?q=photograph).
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: BBrandonScott on March 31, 2014, 08:46:12 am
I found Richard Sexton's piece very insightful and intriguing. But I would like to make two comments.
1) "An essential design attribute for any contemporary digital camera is dual composition and focusing technology. " I argued with this statement in his piece and re-read it several times, but I think he is absolutely correct. I am a long time Nikon user and also a Panasonic GX7 user with a selection of primes (agree about primes also). The focusing of the GX7, while very fast, does not compare with the Nikon. I cannot follow fast moving action, esp little birds. I love the GX7 for travel, but if I'm going after wildlife, it just does not get there. I think Richard might be right. The issue is in the human eye's construction and EVF will never get good enough. The great advantage of micro 4/3s is size and weight. What you give up is great focusing and the ability to follow action in the viewfinder. So you need a dual focusing technology or two camera systems!
2) It seems to me that photography is still in a state of shifting as regards what we do with the final outcome--do we print or do we post. To the largest print size I print to (13x19), micro 4/3s is fine. I'm very satisfied with the print quality. If I were printing to really large sizes, then a larger sensor would be needed.  But I'm not.  If posting to the web, micro 4/3s is fine. The question is, for how much longer will the print be standard for output? I still like to print and have invested both money and time in becoming a good printer. But I find I'm posting a lot more than printing.
Brandon Scott
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: Christoph C. Feldhaim on March 31, 2014, 09:09:24 am
"Zoom with your feet and when the limits of that technique have been reached, change the focal length of the lens."

I always found this sentence misleading when I saw it in one or the other form in various writings .
In my world it's not at all about zooming or getting lazy with your feet.
It's about getting the best position or viewpoint where the subject looks best,
has the best size relation to its surrounding objects or background and foreground,
then -and only then- chose the appropriate lens to get the desired framing.
That is why I (in the meantime) love zoom lenses.
Not because they spare me walking around.
It's because they allow me to always find the best position to take the shot and
get the optimal framing and use of film or sensor real estate after doing that.

Cheers
~Chris
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: Dale_Cotton2 on March 31, 2014, 11:17:19 am
Chris -- you explain it well. Diff between changing focal length and changing camera position is called a change in perspective.

My take is that Richard is asking the reader to meet him half way on details like this or the essay would have had to be multiple times its current length. I'm like you in preferring a moderate zoom; but I read Richard's approach, quoted above, as saying he can find a desirable composition often enough using a combination of primes and relocation that it's not an issue for him.

Richard's essay is half about manufactured-object design and half about why he's made the camera acquisition choices he has. I was frequently delighted with the insights into design history, most of which were new to me. Both his LL essays are 'way up there near the top of my list of favourites on this site. And their being typo-free is the rarest of treats.
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: BJL on March 31, 2014, 11:36:11 am
My take is that Richard is asking the reader to meet him half way on details like this or the essay would have had to be multiple times its current length. I'm like you in preferring a moderate zoom; but I read Richard's approach, quoted above, as saying he can find a desirable composition often enough using a combination of primes and relocation that it's not an issue for him.
I agree: this essay is one excellent photographer who has used a wide range of gear explaining his evolving gear choices in the context of his approach, not dictating that we must all emulate him.  Read that way, it is one of the most thoughtful essays I have read in some time about how camera technology is evolving, and how this can affect gear choices.


P. S. For reasons that others have stated, I also prefer zoom lenses for everything except extreme close-ups ("macro"), but do not begrudge anyone with different priorities.
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: mecrox on March 31, 2014, 12:05:25 pm
An interesting counterpoint would be to lend this Leica equipment to someone or several people of 18-24 for three weeks, say, and then ask how it works for them compared to what they normally get up to. What do they think of these classic but alas uber-expensive cameras and lenses? What do they look for from photography equipment? That might give an insight into where the camera-makers need to go.
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: Hans Kruse on March 31, 2014, 01:32:17 pm
"Zoom with your feet and when the limits of that technique have been reached, change the focal length of the lens."

I always found this sentence misleading when I saw it in one or the other form in various writings .
In my world it's not at all about zooming or getting lazy with your feet.
It's about getting the best position or viewpoint where the subject looks best,
has the best size relation to its surrounding objects or background and foreground,
then -and only then- chose the appropriate lens to get the desired framing.
That is why I (in the meantime) love zoom lenses.
Not because they spare me walking around.
It's because they allow me to always find the best position to take the shot and
get the optimal framing and use of film or sensor real estate after doing that.

Cheers
~Chris

Chris, well said!! That's exactly the point and couldn't agree more.
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: amolitor on March 31, 2014, 04:22:58 pm
It is certainly one of the results of using primes, that one is forced to:


Primes are inherently limiting. This can be a good or a bad thing. "zoom with your feet" is a terrible phrase that has, unfortunately, become something of a stock phrase which means almost nothing except "I prefer to use primes, and think that you should agree with me."

Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: Isaac on March 31, 2014, 04:53:26 pm
"Zoom with your feet and when the limits of that technique have been reached, change the focal length of the lens."

I was sure I'd seen that glib zoom versus prime sentiment expressed somewhere else but I was wrong.

I'd misremembered these (presumably Ernst Haas evading some "What is the best lens?" questions) --

Quote
“Best wide-angle lens? Two steps backward. Look for the ‘ah-ha’.” -- Ernst Haas

“The most important lens you have is your legs.” -- Ernst Haas
Title: Re:
Post by: Christoph C. Feldhaim on March 31, 2014, 05:35:18 pm
Primes educate you to walk. Once you have  grabbed that you can start using rubber glass.
Cheers
~ Chris
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: Telecaster on March 31, 2014, 05:55:39 pm
"Zoom with your feet and when the limits of that technique have been reached, change the focal length of the lens."

I was sure I'd seen that glib zoom versus prime sentiment expressed somewhere else but I was wrong.

Changing the focal length of the lens is not the same as changing the lens.

Being someone who learned photography with fixed focal length lenses, when using zooms I tend to do exactly as Richard's quote suggests. I use the zooming facility as a fine tuner.

"Primes." Zooms. Whatever works best in a given situation.

-Dave-
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: Isaac on March 31, 2014, 06:34:46 pm
Careful! Ray might explain perspective again.

Being someone who learned photography with zoom lens and SLR, I was baffled that with DSLR I seemed mentally stuck at the short-end or the long-end. So I switched to 35mm and 85mm (on APS-C).

(Well, and 100-200mm zoom but that was just the cheapest option for a longer lens.)

Primes. Zooms. Whatever works… (Isn't that what Ernst Haas was saying?)
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: Larry K on March 31, 2014, 08:28:51 pm
Great article; summary of where we are today.  I like primes more than zooms, but when I tried to zoom with my feet one time with a prime, I discovered that I had to do more than walk.  I had to swim a well!  Or fall off a path on a cliff.  That would make a nice cartoon, zooming with your feet into the ocean to photograph a whale.  Why is someone always making us choose one or the other, whether it's a lens or a camera.  Don't we all have 2 plus cameras now and more than one lens? 
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: telyt on April 01, 2014, 12:31:48 am
An interesting counterpoint would be to lend this Leica equipment to someone or several people of 18-24 for three weeks, say, and then ask how it works for them compared to what they normally get up to. What do they think of these classic but alas uber-expensive cameras and lenses? What do they look for from photography equipment? That might give an insight into where the camera-makers need to go.

When I was in that age demographic a kindly old gentleman lent his funky, quaint Leica IIIG with an impossibly slow f/3.5 lens to me and, being accustomed to TTL-viewing and -metering with macro and long lenses, I chose to humor the old guy and pretended I was thrilled.  Within a week the IIIG was my favorite camera.
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: Ray on April 01, 2014, 05:49:34 am
This article by Richard Sexton seems rather convoluted to me. From my perspective, there are just half a dozen main considerations when choosing a camera. They are noise, resolution, autofocusing accuracy, weight, price and lens quality.

There are lots of other secondary considerations, such as frame rate, camera LCD resolution, LiveView implementation, menu arrangement, auto-exposure bracketing range, etc, but all these are issues one can learn to adapt to, or are limitations one can learn to tolerate.

If the fundamental requirements that I've listed are lacking, then some of your images will unavoidably be lacking, which is not to say that it always matters. If your camera is limited in dynamic range, then you can specialize in images with large areas of solid black. Sometimes, that can enhance an image. However, I wouldn't like to be in the position of having to often disguise shadow noise in my images by rendering such areas completely black.

As regards zooms versus  primes, which may require a bit of walking, which may not be good for capturing the moment, but may be good for general health, the issue is also one of lens quality and price. The main reason to buy a prime is if it produces a noticeably better quality, sharper image, than a zoom of the same focal length.

Primes also tend to have a wider maximum aperture than the equivalent zoom, so that's an advantage for the prime, if one likes a very shallow DoF, and/or if one needs a fast shutter speed to freeze movement at a low ISO, which may also result in lower noise.
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: Alan Goldhammer on April 01, 2014, 09:00:13 am
This article by Richard Sexton seems rather convoluted to me. From my perspective, there are just half a dozen main considerations when choosing a camera. They are noise, resolution, autofocusing accuracy, weight, price and lens quality.
Weight is increasingly important to me as I progress in years.  When we travel to Europe I really don't take a full kit of lenses because of the weight and unwillingness to check baggage.  I have a nice shoulder bag that will hold the camera and three lenses.  On our last trip I took the 24mm f2.8, 35mm f2.8 and 16-85 f3.5-5.6 zoom for my Nikon D300.  I admit it was a strange choice given the two primes are pretty close in focal length and I ended up only using the 35mm a couple of times.  When I looked at lens usage, it was about equally split between the zoom and the 24mm.  For Croatia this May the debate is whether to take two primes 24 & 85mm or the 24mm and the 16-85mm zoom.  I can probably shave a little bit of weight off with the two primes but the zoom is a really nice lens for routine shooting.
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: Isaac on April 01, 2014, 10:34:04 am
When I was in that age demographic…

Perhaps a sufficient number of years have passed for that to no longer be considered a useful data point :-)
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: Isaac on April 01, 2014, 10:39:48 am
From my perspective, there are just half a dozen main considerations when choosing a camera.

But… What about the really important things? Upmanship, nostalgia, aesthetics, style…
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: JimAscher on April 01, 2014, 10:57:20 am
While Alan pertinently addresses the lens weight factor, of equal concern to me is the issue of the relative ease (or difficulty) in switching between the various prime lenses I carry with me.  I primarily use old 35-mm film lenses, which I have found cost-benefit analysis much favors -- very fine lenses at relatively low cost.  However, involved with my use of these lenses on various of my cameras involves employing several different lens adapters.  The principal cameras I am currently using are an M-Mount Ricoh GXR, a Sony NEX 5n converted for infrared, and a Foveon sensor Sigma SD15 modified for Nikon-mount lenses.  So, I'm dealing not only with M-mount, NEX-mount and Nikon-mount lenses, but in addition with a few wonderful Kodak Retina Schneider-Kreuznach lenses (and one M-39 Leica mount lens).  Now, admittedly, I do not carry all three cameras on my person at any one time, nor carry all the lenses (and lens adapters) with me at any one time, but I do carry enough to make the switching of these primary lenses in the field for a perceived photo a bit of a chore, both mentally and physically.  I have employed variously a photo backpack, a slingbag (which renders access to the lenses a mite easier), or a photographers vest, with pockets for the lenses.  (My current vest is a wonderful Vested Interest one, but unfortunately its ease of use also renders any incognito photography just about impossible.  Wearing it, one sticks out like a proverbial sore thumb.)  But at least, I can have more ready access to my lenses with it.  I recognize that a zoom lens eliminates many of these difficulties, weight included, but I do like, in the words of the famous advertisement, what my primes do for me (and my photos).
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: Telecaster on April 01, 2014, 04:14:53 pm
With the Sony A7r if I wanted to assemble a "best of" legacy lens lineup it would make for an awkward pic-taking experience. Some focusing rings would turn clockwise to focus closer, others counter-clockwise. Aperture rings would be in different places with different lenses and turn in different directions to stop down. Blecchh! So instead I've put together a number of single-brand/mount lineups, each with two, three or four fixed-focal-length lenses. This way I only need one adapter at a time too. If I want to use native FE lenses instead I go with the 24–70/4 and 55/1.8.

-Dave-
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: BJL on April 01, 2014, 08:51:32 pm
Careful! Ray might explain perspective again.
To Isaac: Yep!:
From my perspective ...
To Ray: about
... there are just half a dozen main considerations when choosing a camera. They are noise, resolution, autofocusing accuracy, weight, price and lens quality.
I have nothing against you adopting that perspective, but what has it for to do with Richard Sexton's article, since he clearly has a somewhat different set of main considerations.  Hopefully you are not suggesting that every photographer should share your apparently lower priority on considerations like accurate manual focusing, quick and convenient manual focus on an off-centre subject, dynamic range, or the types of lenses available (as opposed to just their quality).  But you do seem to imply that when you shift from speaking of your idiosyncratic perspective to the more general phrasing "issues one can learn to adapt to, or are limitations one can learn to tolerate".  This one is not interested in adapting to or tolerating weaknesses like poor manual focusing capabilities, a lack of zoom lenses covering my preferred combination of FOV range and weight, or the lack of good macro lenses.
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: Ray on April 02, 2014, 12:15:35 am
To Ray: about I have nothing against you adopting that perspective, but what has it for to do with Richard Sexton's article, since he clearly has a somewhat different set of main considerations.

Everyone has a different perspective. That's what being an individual means. Even inseparable twins, with two heads and one body, can have a different perspective, each from the other.

Quote
Hopefully you are not suggesting that every photographer should share your apparently lower priority on considerations like accurate manual focusing, quick and convenient manual focus on an off-centre subject, dynamic range, or the types of lenses available (as opposed to just their quality).

Of course not. I'm quite opposed to the herd mentality. The interesting aspects of discussions such as this, and I also include critiques of any photographs, are the reasons expressed for liking, disliking, agreeing, disagreeing, and so on.

Sometimes reasoning can be flawed, or not supported by the evidence. For example, in your above statement you imply that I consider dynamic range as one of the lower priorities, outside of the 6 main priorities that I listed. I think that is an oversight on your part, so I hope you are grateful for my correcting you.  ;)

One of the 6 main issues I mentioned was 'noise'. Dynamic range is all about noise, although specifically noise in the shadows rather than noise in the midtones, but noise nevertheless. So obviously I consider DR capability one of the main considerations.

Likewise, the term 'quality' as applied to lenses can refer to more than the MTF response. A telephoto lens has the quality of being able to render distant objects sharp, bearing in mind that sharp is a relative term. Even if the MTF response of the telephoto lens is poor, it will likely still do a far better job of rendering distant objects sharp, than the most expensive wide-angle lens available.

Quote
But you do seem to imply that when you shift from speaking of your idiosyncratic perspective to the more general phrasing "issues one can learn to adapt to, or are limitations one can learn to tolerate". This one is not interested in adapting to or tolerating weaknesses like poor manual focusing capabilities, a lack of zoom lenses covering my preferred combination of FOV range and weight, or the lack of good macro lenses.

As regards adapting to or tolerating weaknesses such as poor manual focusing capability, would you care to explain why you consider poor, or 'less than ideal', manual focusing capability to be one of the major concerns?
For example, do you frequently take shots with very wide apertures resulting in a very shallow DoF where accurate focusing is more critical and autofocusing is not up to the job?

If you were offered a choice from two, new, cutting-edge cameras which differed only in the respect that one camera boasted an unprecedented, accurate autofocusing system in all types of lighting and with all lenses designed for the camera, but had only an average manual focusing capability; and the other camera boasted an unprecedented manual focusing capability, but had only average autofocusing capability, which would you choose?

I would choose the camera with improved autofocusing. Perhaps at this point I should clarify what I understand as the distinction between manual focussing and autofocusing. If you think that focal length is not a 'quality', perhaps you think that manually moving an autofocusing square in the viewfinder, using one's thumb, constitutes manual focusing. I hope not.  ;)  But I agree there is a certain manual dexterity required in that process.
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: OldRoy on April 02, 2014, 06:22:21 am
"...smaller, more discrete camera bodies... quiet, discrete, take anywhere cameras..."
As opposed to conglomerated cameras, perhaps?

This must be the internet. Another one to add to the near-ubiquitous "loose/lose" confusion (which I find breathtakingly illiterate). I guess proof reading (or maybe the literate proof-reader) has become an anachronism. That's Progress folks...

Roy
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: amolitor on April 02, 2014, 06:57:20 am
There's also it's/its confusion. My pet peeve. But it is the internet, and the standard is 'is it clear?' not 'is it correct?'
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: BJL on April 02, 2014, 09:40:43 am
Ray, thank you for clarifying your list of priorities, such as your perspective on the meaning of "quality" and on the relationship between noise and dynamic range.  I note that noise and dynamic range are not usually considered to be exactly the same thing, due to the role of full well capacity in the latter.  In particular, in comparison between a larger CCD with larger photosites and a smaller active pixel CMOS sensor with smaller photosites, it is sometimes the case that the former has a greater dynamic range and better image quality at low exposure index while the latter has less dark noise and better image quality in low light, high exposure index situations.

I indicated that good manual focusing features are one major priority for me, including for example the ability to use an enlarged image through the eye-level electronic viewfinder because the definition of the image in an OVF is at best comparable to about 2MP, and not always good enough for shallow DOF situations like macro photography.  Since I did not declare that this or any one aspect dominates over all others in my weighing of the advantages and disadvantages of various systems, your "which camera" question based on hypothetical extremes of AF vs MF performance with no information on any other factors relevant to my system choice is a false dichotomy.  How about just accepting your final statement about which camera is preferable from your perspective?

It is in your word "confusing" that after acknowledging that we can have different perspectives, you set out to challenge my priority on manual focusing features (shared by a great many other photographers in this forum).

What is it going to be Ray: are we all individuals entitled to our diverse perspectives, or is anyone who does not share the Ray perspective "confused"?
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: Ray on April 02, 2014, 11:39:35 am
Since I did not declare that this or any one aspect dominates over all others in my weighing of the advantages and disadvantages of various systems, your "which camera" question based on hypothetical extremes of AF vs MF performance with no information on any other factors relevant to my system choice is a false dichotomy.  How about just accepting your final statement about which camera is preferable from your perspective?

It is in your word "confusing" that after acknowledging that we can have different perspectives, you set out to challenge my priority on manual focusing features (shared by a great many other photographers in this forum).

What is it going to be Ray: are we all individuals entitled to our diverse perspectives, or is anyone who does not share the Ray perspective "confused"?

Not at all, BJL. We are all confused to the extent that are we unable to provide a rational, convincing or plausible explanation for our views and perspectives.

My hypothetical question about 'which camera' seems quite straightforward to me. I was simply trying to elicit from you an explanation as to why you appear to give a higher priority to manual focusing capability than I do, and whether or not the reason is because you find the current autofocusing accuracy of your cameras inadequate.

Assuming you are in the market for a new camera, and assuming that all the features of both cameras offered in this hypothetical scenario meet your requirements, would you choose the model with unprecedented autofocusing accuracy, but average manual focussing convenience, or the one with improved manual focusing ease and convenience, but average autofocusing?
Title: if you really need an explanation of why manual focus is useful ...
Post by: BJL on April 02, 2014, 01:30:40 pm
I was simply trying to elicit from you an explanation as to why you appear to give a higher priority to manual focusing capability than I do, and whether or not the reason is because you find the current autofocusing accuracy of your cameras inadequate.
Since the subject is how people choose between actual systems, I do not see the point of asking me to to consider your imaginary hypothetical case of "unprecedented, accurate autofocusing system in all types of lighting and with all lenses designed for the camera", particularly since that description ignores one of the well-known limitations to AF systems: selection of the focus target.  So let me try to answer your latest version of the question:

Some of my favorite subjects are of wildlife using long lenses and of very small subjects at close range, where DOF can be quite shallow (even in smaller formats like 4/3"!), and the desirable focus target is often a feature like an eye or a certain part of a plant or insect that is not at dead-center and is not distinguished in any simple algorithmic way like being closest to the camera or having a clear high-contrast edge.  AFAIK, no AF software in any system will reliably detect my compositional choice of focus location in these cases, so AF requires at least either pushing buttons to select amongst a large array of AF points or focus and recompose, either of which I find far less convenient than using my eyes to choose and check focus.  So I switch to manual focus in these situations.

So no, the reason is not just the autofocusing accuracy of my particular current cameras.


Ray, may I ask you a question in return: are you genuinely in doubt that a good number of photographers have a valid rational and empirical basis for caring about manual focusing, and for judging that some systems support this significantly better than others?
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: Isaac on April 02, 2014, 01:52:59 pm
The interesting aspects of discussions such as this, and I also include critiques of any photographs, are the reasons expressed for liking, disliking, agreeing, disagreeing, and so on.

Yes! I'll take the opportunity to express strong agreement with you about that.
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: Isaac on April 02, 2014, 02:02:36 pm
As opposed to conglomerated cameras, perhaps?
That's mockery folks.

It would be more helpful to ask the web master to correct the  typo "discrete" to "discreet".

Has charitable reading become an anachronism?
Title: Re:
Post by: Christoph C. Feldhaim on April 02, 2014, 02:26:14 pm
Seems we nave a new pissing contest ...
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: telyt on April 02, 2014, 02:31:00 pm
If you were offered a choice from two, new, cutting-edge cameras which differed only in the respect that one camera boasted an unprecedented, accurate autofocusing system in all types of lighting and with all lenses designed for the camera, but had only an average manual focusing capability; and the other camera boasted an unprecedented manual focusing capability, but had only average autofocusing capability, which would you choose?

I would choose the camera with the better manual focussing capability, every time.  First, because I trust myself to focus on a particular point before I'd trust AF to focus on the point I want to be in focus (assuming that the hypothetical AF system is perfectly accurate once the point is selected - this IMHO is part of the 'unprecedented' part of the hypothetical AF system); second, because AF points are in fixed locations that do not cover the entire picture area and with a good manual focus capability the user can focus anywhere in the picture area, third because I want visual confirmation that the point I want is in focus, and fourth because I do not want to be limited to the lenses designed for the camera.

The DOF in my photos - of small critters at close range with little opportunity to use small apertures - is vanishingly small so focussing errors are unforgivable; the critters move, so focus-lock-recompose is out of the question (been there, done that, threw out lots of pictures) and the time wasted recognizing when the AF system isn't doing what I want then over-riding the thing means lost opportunities.  I won't waste forum bandwidth posting examples of images but those who are interested in seeing if my photos consist of small fast-moving subjects are welcome to browse my website.  The photos on my website are 100% manual focus.
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: Isaac on April 02, 2014, 03:27:08 pm
The DOF in my photos … is vanishingly small so focusing errors are unforgivable…

As usual in these discussions, the difference in the type of photograph we commonly take is the main factor.

Even I'll notice that AF locked on the tip-of-a-nose rather than the iris; but maybe I just need a new camera (http://sonyalphalab.com/forums/showthread.php?362-Sony-A7-and-Eye-AF-Feature-Help) :-)
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: BJL on April 02, 2014, 04:51:11 pm
As usual in these discussions, the difference in the type of photograph we commonly take is the main factor.
That is the point of course (For example, I think that at 16MP+, all current systems have enough resolution for all my needs, but I am not going to interrogate anyone who still makes "resolution" a factor in choosing between systems.)

Even I'll notice that AF locked on the tip-of-a-nose rather than the iris; but maybe I just need a new camera (http://sonyalphalab.com/forums/showthread.php?362-Sony-A7-and-Eye-AF-Feature-Help) :-)
That Sony A7 eye focus is insufficient: I want to decide which eye to focus on.  Actually, Olympus claims that the E-M5 can do this:
"Improved Eye detection lets you choose from left or right eye priority modes to get the correct focus point quickly for perfect portraits every time" -- http://www.getolympus.com/us/en/e-m5.html
but I doubt it works well on non-human eyes, especially insect eyes.
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: Isaac on April 02, 2014, 05:54:44 pm
That Sony A7 eye focus is insufficient: I want to decide which eye to focus on.

(I don't have an A7). We know the problem will arise when there are 2 eyes in the focus box, so maybe we could move the focus box; and for my purposes, having one eye tack-sharp would be better than having the tip-of-the-nose tack-sharp :-) Obviously, depending on face-orientation, a lot of the time, one eye in focus would mean both eyes in focus.

I'd probably still use MF out-of-habit but I can see that others might decide Eye AF was better for them.
Title: today's automation is impressive --- at least for common subjects like people
Post by: BJL on April 03, 2014, 10:11:44 am
We know the problem will arise when there are 2 eyes in the focus box, so maybe we could move the focus box; and for my purposes, having one eye tack-sharp would be better than having the tip-of-the-nose tack-sharp :-) Obviously, depending on face-orientation, a lot of the time, one eye in focus would mean both eyes in focus.

I'd probably still use MF out-of-habit but I can see that others might decide Eye AF was better for them.
Joking aside, the human face is such a dominant photographic subject that it has attracted an impressive technological effort at automation, and I have no basis for thinking that AF is inadequate for that case --- partly because of my ignorance: I rarely photograph people except as casual snapshots, and instead my main interest is nature photography.

Typically, it is the less common cases that the automation handle less well, and which require turning off the automation and making our own choices of exposure levels, shutter speeds, aperture, white balance, contrast level, amount of gain applied before raw conversion, focus point, etc.
Title: Re: today's automation is impressive --- at least for common subjects like people
Post by: telyt on April 03, 2014, 10:53:24 am
Joking aside, the human face is such a dominant photographic subject that it has attracted an impressive technological effort at automation, and I have no basis for thinking that AF is inadequate for that case --- partly because of my ignorance: I rarely photograph people except as casual snapshots, and instead my main interest is nature photography.

Typically, it is the less common cases that the automation handle less well, and which require turning off the automation and making our own choices of exposure levels, shutter speeds, aperture, white balance, contrast level, amount of gain applied before raw conversion, focus point, etc.

I agree completely.  With typical subjects and compositions automatic features are great, especially now that on-chip hybrid AF combined with facial recognition is becoming available.  It's the atypical photographs where automatic features leave much to be desired.
Title: Re: if you really need an explanation of why manual focus is useful ...
Post by: Ray on April 03, 2014, 05:56:33 pm
Since the subject is how people choose between actual systems, I do not see the point of asking me to to consider your imaginary hypothetical case of "unprecedented, accurate autofocusing system in all types of lighting and with all lenses designed for the camera", particularly since that description ignores one of the well-known limitations to AF systems: selection of the focus target.  So let me try to answer your latest version of the question:

Some of my favorite subjects are of wildlife using long lenses and of very small subjects at close range, where DOF can be quite shallow (even in smaller formats like 4/3"!), and the desirable focus target is often a feature like an eye or a certain part of a plant or insect that is not at dead-center and is not distinguished in any simple algorithmic way like being closest to the camera or having a clear high-contrast edge.  AFAIK, no AF software in any system will reliably detect my compositional choice of focus location in these cases, so AF requires at least either pushing buttons to select amongst a large array of AF points or focus and recompose, either of which I find far less convenient than using my eyes to choose and check focus.  So I switch to manual focus in these situations.

So no, the reason is not just the autofocusing accuracy of my particular current cameras.


BJL,
I understand that situation perfectly, and the Live View systems of some current models of cameras, particularly Canon models, allow for very easy manual focusing, provided one uses a tripod. I recall being amazed at the detail I could see on my Canon 50D LCD screen at 10x magnification, especially when using a telephoto lens.

What sort of improvements in manual focusing do you have in mind?

Quote
Ray, may I ask you a question in return: are you genuinely in doubt that a good number of photographers have a valid rational and empirical basis for caring about manual focusing, and for judging that some systems support this significantly better than others?

This is where the confusion lies. One cares about accurate focusing, period. I personally don't give a stuff whether it's manual or auto as long as the focusing system has the qualities that enable me to get the shot, with the parts in focus that I want to be in focus, and within the time-frame that's available for the shot; and, I should add, with as little fiddling as possible since that tends to distract the mind from the more important considerations such as composition. However, I will always try to use manual focusing in circumstance when I suspect or predict that autofocusing will be inadequate.

If someone were to devise a manual focusing system that was at least as good as current autofocusing systems in all situations, and better in certain specialised situations, such as macro photography, then I would be a great fan of such a manual system.

If someone were to device an autofocus system that matched the capabilities of manual focus in those specialised situations, but was far quicker and just as accurate in all situations, then that's the system I would prefer
However, I think it would be reasonable to claim that a manual focusing system cannot have the potential to equal the speed of autofocusing because it is limited by the speed of human manual dexterity.

Now, I don't claim to be an authority on what is technologically feasible regarding developments in autofocus technology, but there does appear to be some research taking place that could lead to even faster and more accurate autofocus systems in future cameras.

Here are some links. http://www.pnas.org/content/early/2011/09/16/1108491108.abstract
http://www.futurity.org/how-to-make-cameras-auto-focus-like-eyes/
http://www.theguardian.com/global/2012/jan/15/digital-auto-focus-defocus-technology

Also, I recall many years ago buying a Canon EOS film SLR which featured a system which could track the eyeball as one looked through the viewfinder, allowing one to select as a focusing point, that part in the scene that the eye had settled on.

It now occurs to me that further development of that technology could lead to a quicker method of  moving the focusing square that one sees in the viewfinder of modern DSLRs. If eyeball tracking was available 20 years ago, one would imagine it would not be too difficult to devise an improved system whereby a single focusing square could precisely follow the movement of the eye as it perused the scene through the viewfinder.

If one were to combine such an eye-tracking system with improved autofocus accuracy, in relation to whatever focusing point had been chosen, then I think one would have a system that would make manual focusing completely redundant. That's what I'd like to see. I'm a progressive.  ;)
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: Ray on April 03, 2014, 08:16:19 pm
I would choose the camera with the better manual focussing capability, every time.  First, because I trust myself to focus on a particular point before I'd trust AF to focus on the point I want to be in focus .....

The nature of the 'unprecedented' improvement in autofocusing includes both improved accuracy and improved 'efficiency of selection' of the precise focus point within the scene.

Oh! Ye person of little faith!  ;D
Title: Re: if you really need an explanation of why manual focus is useful ...
Post by: BJL on April 03, 2014, 09:25:59 pm
the Live View systems of some current models of cameras, particularly Canon models, allow for very easy manual focusing, provided one uses a tripod.
Good, so we agree that good live manual focusing can be an important factor in _some_ people's choice of camera.  Since that was my original point, perhaps I should stop here.

However, you have overlooked my statement that I like to be able to do this with my eye to the viewfinder, partly because I often want to do this hand-held and/or in lighting conditions where viewing the rear screen is difficult. And that of course becomes a differentiation between different systems, because some have live view eye level viewfinders (”EVFs") while others do not.
Title: Re: today's automation is impressive --- at least for common subjects like people
Post by: Isaac on April 03, 2014, 10:05:22 pm
It's the atypical photographs where automatic features leave much to be desired.

Which makes the transition between AF and MF, and between MF and AF an important factor; because we'll want to make best use of what both offer.

I like the sound of this: "Direct manual focus (DMF) You can make fine adjustments manually after the focus is locked. You can quickly focus on a subject rather than using the manual focus from the beginning."
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: telyt on April 04, 2014, 12:21:43 am
The nature of the 'unprecedented' improvement in autofocusing includes both improved accuracy and improved 'efficiency of selection' of the precise focus point within the scene.

Oh! Ye person of little faith!  ;D

Can't use a camera that doesn't exist  :D
Title: ergonomics: switching between AF and MF
Post by: BJL on April 04, 2014, 10:54:25 am
I like the sound of this: "Direct manual focus (DMF) You can make fine adjustments manually after the focus is locked. You can quickly focus on a subject rather than using the manual focus from the beginning."
Agreed: another of the considerations that many photographers would add to Ray's list is various aspects of ergonomics, and one good example is the ability to quickly and easily switch focus modes.  That "DMF" quote is from the description of a Sony camera, isn't it? It sounds similar to the various "AF+MF" modes of Olympus Micro Four Thirds cameras.  In fact, Olympus Four Thirds SLRs have also had it for a long time, but I prefer it with the option to have turning the focus ring also activate a magnified live view in the EVF.

I would like it even more if that magnified live view was in a window within a view that still shows the overall framing.  Which cameras, if any offer, that option?
Title: Re: ergonomics: switching between AF and MF
Post by: Isaac on April 04, 2014, 12:28:39 pm
That "DMF" quote is from the description of a Sony camera, isn't it?

I was trying to keep clear of the brand wars ;-) A7

Instead of AF or MF, how about a more dynamic focus control system that took MF actions as a directive to identify what AF should bring into focus: MF to AF to MF to AF…

I would like it even more if that magnified live view was in a window within a view that still shows the overall framing.  Which cameras, if any offer, that option?

Maybe… (http://www.google.com/glass/start/what-it-does/)
Title: blending AF with fine manual control of where AF is done, and ergonomics
Post by: BJL on April 06, 2014, 04:03:08 pm
I was trying to keep clear of the brand wars ;-) A7
A worthy goal, and for all I know, all ILC systems have such an AF+MF mode these days --- though it might be easier to implement in lenses that use focus-by-wire, to avoiding the clash of manually turning a mechanically coupled focus ring one way while the AF motor is trying to move in the opposite direction.

Instead of AF or MF, how about a more dynamic focus control system that took MF actions as a directive to identify what AF should bring into focus: MF to AF to MF to AF…
I look forward to camera makers experimenting with ideas like that. At risk of heresy, I can see enhanced touch-screen controls being useful even when using the eye-level VF so that the touch-screen cannot be seen.
For example, one current camera [brand name suppressed to keep the peace] offers some ingredients that have potential for "manual selection of AF point", but they could work together far better than they currently do:
1. Select focus region (by touch on the rear screen, which only works when looking at that screen rather than using the eye-level VF)
2. zoom the preview to the selected focus region (by a preselected magnification factor; done with a press on one of the programmable buttons)
3. AF on the zoomed region (done with a half-press and hold of the shutter release). This can gives precise selection of a very small AF target if the maximum preview magnification is selected, generally making MF unnecessary if just slight camera movement is allowed to get the focus target right in the bull's-eye.
4. unzoom the preview, to check and fine-tune framing (done with another press on that programmable button)

But I struggle to use the current implementation of this, so I would like:
a. to be able to do step 1 with my eye to the VF, by "tracking" my finger over the rear touch-screen or with a touch pad in place of the four-way arrows
b. to adjust the degree of magnification at step 2 quickly on the fly, maybe with a touch-screen gesture or a slider
c. to avoid step 4, by the option of a "window-in-window" preview.
Title: Re: blending AF with fine manual control of where AF is done, and ergonomics
Post by: telyt on April 07, 2014, 12:09:47 am
A worthy goal, and for all I know, all ILC systems have such an AF+MF mode these days --- though it might be easier to implement in lenses that use focus-by-wire, to avoiding the clash of manually turning a mechanically coupled focus ring one way while the AF motor is trying to move in the opposite direction.
I look forward to camera makers experimenting with ideas like that. At risk of heresy, I can see enhanced touch-screen controls being useful even when using the eye-level VF so that the touch-screen cannot be seen.
For example, one current camera [brand name suppressed to keep the peace] offers some ingredients that have potential for "manual selection of AF point", but they could work together far better than they currently do:
1. Select focus region (by touch on the rear screen, which only works when looking at that screen rather than using the eye-level VF)
2. zoom the preview to the selected focus region (by a preselected magnification factor; done with a press on one of the programmable buttons)
3. AF on the zoomed region (done with a half-press and hold of the shutter release). This can gives precise selection of a very small AF target if the maximum preview magnification is selected, generally making MF unnecessary if just slight camera movement is allowed to get the focus target right in the bull's-eye.
4. unzoom the preview, to check and fine-tune framing (done with another press on that programmable button)

But I struggle to use the current implementation of this, so I would like:
a. to be able to do step 1 with my eye to the VF, by "tracking" my finger over the rear touch-screen or with a touch pad in place of the four-way arrows
b. to adjust the degree of magnification at step 2 quickly on the fly, maybe with a touch-screen gesture or a slider
c. to avoid step 4, by the option of a "window-in-window" preview.

Rube Goldberg comes to mind.
Title: Re: blending AF with fine manual control of where AF is done, and ergonomics
Post by: BJL on April 07, 2014, 04:56:17 pm
Rube Goldberg comes to mind.
Maybe you were referring to this:
(http://www.dpreview.com/files/news/5952269437/VCT-55LH_XLR-K1M_CLM-V55_right_ILCE-7S_VX9124-1200.jpg)

It does seem lately that there is a rush of potentially useful new technology being put into cameras before the designers have thought out the user interfaces needed to let photographers use the new stuff as they want to.
Title: Re: blending AF with fine manual control of where AF is done, and ergonomics
Post by: telyt on April 07, 2014, 05:11:28 pm
Maybe you were referring to this:

 ;D

WRT focussing I don't grok why it must be so complicated.
Title: Re: blending AF with fine manual control of where AF is done, and ergonomics
Post by: BJL on April 07, 2014, 09:13:11 pm
WRT focussing I don't grok why it must be so complicated.
Most of the time it is not so complicated: the one tough case that I often encounter is when I want to focus on an off-center point (so away from any traditional manual focusing aids) in a shallow DOF situation like close-up photography, and focus-and-recompose is unsatisfactory, perhaps because I am hand-holding. Then the un-magnified off-center image of any VF, optical or electronic, is not large or detailed enough for precise focusing, and traditional manual focusing tools like split images of micro-prism collars are not available, so I want magnification at a chosen off-center location.

Anyway, I can envison an inteface that makes it not at all complicated; hopefully smarter people than I will eventually get a simple solution into a camera.

Any suggestions?
Title: Re: blending AF with fine manual control of where AF is done, and ergonomics
Post by: telyt on April 07, 2014, 11:14:20 pm
Most of the time it is not so complicated: the one tough case that I often encounter is when I want to focus on an off-center point (so away from any traditional manual focusing aids) in a shallow DOF situation like close-up photography, and focus-and-recompose is unsatisfactory, perhaps because I am hand-holding. Then the un-magnified off-center image of any VF, optical or electronic, is not large or detailed enough for precise focusing, and traditional manual focusing tools like split images of micro-prism collars are not available, so I want magnification at a chosen off-center location.

The situation you've described is one I also encounter often.  Before the multiple demands made of the light coming through the lens (i.e., light siphoned away from the viewing system for multi-pattern metering and AF systems) it was simple.  A large bright plain matte view screen like the Nikon F "E" screen or the Leicaflex SL's extremely fine micro prisms over the entire image area made this child's play.  This is 1960s technology.  Here's how the user focusses:

1) look through the viewfinder at the part of the image s/he wants to be in focus
2) turn the focussing ring until the desired image point has the highest contrast
3) there is no part three!

To put this kind of focussing system in a DSLR the camera maker would start by replacing the semi-silvered mirror with a fully-silvered mirror; this would make the viewfinder brighter so the viewscreen optimized for brightness (at the expense of focussing accuracy) could be replaced with one optimized for focussing accuracy.  The additional brightness resulting from the fully-silvered mirror also means the viewfinder magnification could be increased without reducing brightness to unacceptable levels.

This would mean no AF or multi-pattern metering.

I suggest that for many situations multi-pattern metering has outlived its peak usefulness because we can now review the histogram in near-real time.

I would further suggest that for many situations AF is a solution in search of a problem.  AF for landscapes?  Really?!?  It's not like the rocks & trees are going to run away.
Title: Re: blending AF with fine manual control of where AF is done, and ergonomics
Post by: Isaac on April 08, 2014, 04:18:04 pm
…a solution in search of a problem.

My guess is that cameras are made for the vast majority of people who don't practice focusing their camera.


AF for landscapes?  Really?!?  It's not like the rocks & trees are going to run away.

The light runs away.

(Last week a beautiful rainbow appeared in perfect position over the motif and faded-away; to quickly for me to grab a camera, let alone focus.)
Title: Re: blending AF with fine manual control of where AF is done, and ergonomics
Post by: ErikKaffehr on April 08, 2014, 05:37:37 pm
Hi,

My experience is that AF is pretty exact for stationary subjects. On the other hand I prefer to use manual focus in magnified live view as soon as I have the camera on tripod and the subject is stationary.

One problem with AF is to choose the point focused on. There may be many options.

It seems that some users can focus very well using MF, but it got harder with modern view finders.

Best regards
Erik

My guess is that cameras are made for the vast majority of people who don't practice focusing their camera.


The light runs away.

(Last week a beautiful rainbow appeared in perfect position over the motif and faded-away; to quickly for me to grab a camera, let alone focus.)
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: Ray on April 09, 2014, 12:49:06 pm
I can't see a major problem here when autofocusing with my Nikon D800E. There's an area outlined in the viewfinder, which is a bit smaller in size than a DX frame. Within that area one can choose a total of either 11 or 51 focus-point options which can be selected with one's thumb on the 'multiselector' dial on the back of the camera.

In other words, one can move a single focusing square to any of either 11 or 51 positions within a central area of the frame, as one looks through the viewfinder. Each press of the 'multiselector' dial moves the focusing square one position to the left or right, or to the top or bottom.

After moving the focusing square to the desired position within the composition, one can lock focusing with a single press of the AF-On button. No need even to keep one's thumb on the AF-On button. It's free to make other adjustments.

It should only be necessary to recompose (before pressing the shutter) if the desired focus point is close to an edge or corner of the frame, and even then, the degree of movement to recompose should not be significant, unless the desired focus point was in the very corner of the frame, which would be very unusual, especially considering that lenses are least sharp in the corner.
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: Telecaster on April 09, 2014, 03:42:15 pm
It should only be necessary to recompose (before pressing the shutter) if the desired focus point is close to an edge or corner of the frame, and even then, the degree of movement to recompose should not be significant, unless the desired focus point was in the very corner of the frame, which would be very unusual, especially considering that lenses are least sharp in the corner.

Once you've used a camera that actually can focus precisely in the extreme corners you may discover that what looks like a lack of resolving capability is often just field curvature.

-Dave-
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: ErikKaffehr on April 09, 2014, 04:05:30 pm
Hi,

I would agree on that point.

Best regards
Erik

Once you've used a camera that actually can focus precisely in the extreme corners you may discover that what looks like a lack of resolving capability is often just field curvature.

-Dave-
Title: Re: blending AF with fine manual control of where AF is done, and ergonomics
Post by: BJL on April 09, 2014, 05:28:36 pm
The situation you've described is one I also encounter often.  Before the multiple demands made of the light coming through the lens (i.e., light siphoned away from the viewing system for multi-pattern metering and AF systems) it was simple.  A large bright plain matte view screen like the Nikon F "E" screen or the Leicaflex SL's extremely fine micro prisms over the entire image area made this child's play.  This is 1960s technology.  Here's how the user focusses:

1) look through the viewfinder at the part of the image s/he wants to be in focus
2) turn the focussing ring until the desired image point has the highest contrast
3) there is no part three!
...
This would mean no AF or multi-pattern metering.
That sounds like a nice option for some photographers, at least those (1) working with a camera that has a large enough format and a lens of large enough maximum aperture to provide a sufficiently bright and detailed OVF image [I suspect 35mm and up, and either primes of fast zooms], and (2) who never need AF, or can carry another camera for that need.  (No comment on multi-pattern metering, since I haven't used that for years.)

I can see that working for you (as shown by your great bird photos!), but hopefully you can see that it is not for all of us. My priorities run to having a single system camera with a combination of compactness and telephoto reach (and price) that rules out formats of 35mm or up.

I have some concerns though:
1) The resolution of the scattered secondary image in an SLR's OVF is far less than that of the sensor (~2MP?) so when shooting at close to maximum aperture, I doubt that any OVF manual focusing could be very precise.
2) At large apertures, about f/2 and beyond, the  secondary image in an SLR's OVF image has more DOF than the actual recorded image, so reliably precise MF with an OVF seems impossible.
3) How good is this when focusing off-center with the extremely shallow DOF of extreme close-ups?
Title: UI progress: arrow keys are so 1980s
Post by: BJL on April 09, 2014, 05:41:11 pm
In other words, one can move a single focusing square to any of either 11 or 51 positions within a central area of the frame, as one looks through the viewfinder. Each press of the 'multiselector' dial moves the focusing square one position to the left or right, or to the top or bottom.
That sounds like up to about a dozen arrow button pushes to navigate between the "mere" 51 AF points that even state-of-the-art reflex system PDAF is limited to. For speed, I would prefer a single touch or quick glide across the rear touch-screen to select the AF point, along with the hundreds of AF points potentially possible with in-sensor AF and/or the complete flexibility in choosing the region to zoom in on in the EVF.  And I suspect that wildlightphoto would prefer the even faster focus point selection method of doing it with his eye and then watching that point as he turns the focus dial.


P. S. We have a lot of "this works well enough for my usage, so it should be good enough for every competent photographer" posts lately.
Title: Re: blending AF with fine manual control of where AF is done, and ergonomics
Post by: telyt on April 09, 2014, 06:06:33 pm
It should only be necessary to recompose (before pressing the shutter) if the desired focus point is close to an edge or corner of the frame, and even then, the degree of movement to recompose should not be significant, unless the desired focus point was in the very corner of the frame, which would be very unusual, especially considering that lenses are least sharp in the corner.

This is cropped from the corner of the image:

(http://www.wildlightphoto.com/birds/falconidae/webster02.jpg)

where my lens is plenty sharp at full aperture.  It's the Leica 280mm f/4 APO-Telyt-R.

According to Canon, focus-lock-recompose is unreliable at close range with shallow DOF, and my experience confirms it.  Along with the geometric inaccuracy caused by moving the image plane (or curved field, depending on the lens) which can be debated endlessly, subjects move in the time it takes to re-compose.



I have some concerns though:
1) The resolution of the scattered secondary image in an SLR's OVF is far less than that of the sensor (~2MP?) so when shooting at close to maximum aperture, I doubt that any OVF manual focusing could be very precise.

Accurate manual focus does not depend on the resolution of the screen.  Human eyes are much more sensitive to contrast than to detail, so the easiest way to focus is to maximize contrast at the point(s) you want to be in focus.  Nikon's "E" screen was great for this because the grid lines provided a reference point of maximum black.

Quote
2) At large apertures, about f/2 and beyond, the  secondary image in an SLR's OVF image has more DOF than the actual recorded image, so reliably precise MF with an OVF seems impossible.

That depends on the viewscreen.  The viewscreen you've described is typical of AF cameras, optimized for brightness at the expense of focussing accuracy.

Quote
3) How good is this when focusing off-center with the extremely shallow DOF of extreme close-ups?

See falcon photo above.  It would not have been easy keeping the eyes within a field of focus points.

Another benefit of being able to evaluate focus anywhere on the viewscreen is the ability to evaluate focus simultaneously in several regions of the picture.  While photographing this Mountain Bluebird I was able to maneuver so that not only is the eye in the plane of focus but much of the bird's back is also in focus - and I could see this in the viewfinder.

(http://www.wildlightphoto.com/birds/turdidae/mobl00.jpg)
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: BJL on April 09, 2014, 07:59:26 pm
Doug,
    I am definitely with you on the inadequacy of auto-focus, at least as it exists today, to choose focus on small off-center details in a scene like a bird's eye or a particular part of a botanical close-up.  With single point AF, it can be impractical to keep such a detail under a single selected AF point with a hand-held camera or a moving subject; with multi-point AF, the focus point is basically chosen as the part of the scene that is closest to the camera, which is not always where I want the focus to be!

Also, I had not thought of the fact that human focusing of an SLR (away from a central split image focusing aid) is basically contrast detect.  That motivates me to experiment more with how CDAF compares to manual focus on off-center subjects.
Title: Re: UI progress: arrow keys are so 1980s
Post by: Ray on April 09, 2014, 10:25:56 pm
That sounds like up to about a dozen arrow button pushes to navigate between the "mere" 51 AF points that even state-of-the-art reflex system PDAF is limited to. For speed, I would prefer a single touch or quick glide across the rear touch-screen to select the AF point, along with the hundreds of AF points potentially possible with in-sensor AF and/or the complete flexibility in choosing the region to zoom in on in the EVF.  And I suspect that wildlightphoto would prefer the even faster focus point selection method of doing it with his eye and then watching that point as he turns the focus dial.


P. S. We have a lot of "this works well enough for my usage, so it should be good enough for every competent photographer" posts lately.

I see that I should have provided more details. Whilst a single, brief press of the multiselector dial moves the focus square just one position, holding one's thumb on the dial causes the square to zip across the screen in a fraction of a second, but at a speed which is not too fast to avoid overshoot, with a bit of practice.

There is also a 'wrap around' option in the menu which, when enabled, allows one to move the focusing square in a single jump  from one extreme edge of the outlined frame to the opposite edge. Since the focusing square always remains in the same position it was last placed , this 'wrap around' feature can be useful when the following shot requires a focus point near the opposite edge to the previous shot.

Hope I've managed to clarify that.  ;)
Title: Re: blending AF with fine manual control of where AF is done, and ergonomics
Post by: Ray on April 09, 2014, 10:38:06 pm
This is cropped from the corner of the image: where my lens is plenty sharp at full aperture.  It's the Leica 280mm f/4 APO-Telyt-R.

Why is the image cropped?

My point is, it's very unusual to have the main focus of interest in the extreme corner. One can't expect manufacturers to cater to the requirements of a tiny fraction of their clientele whose speciality is breaking the rule-of-thirds. They'd go broke.

Your Leica 280mm f/4 APO-Telyt-R might well be impressively sharp in the corners, but by far the majority of lenses, including those primes with an excellent reputation for sharpness, exhibit significantly lower resolution in the extreme corners, especially if they are full-frame lenses used with a full-frame sensor. Whether or not that lower resolution is acceptable is another matter.

Quote
According to Canon, focus-lock-recompose is unreliable at close range with shallow DOF.

Of course it is. However, the degree of unreliability is proportional to the degree of recomposing, as well as the shallowness of DoF and the closeness of the subject.

With my D800E, I rarely need to recompose more than very slightly. 51 different focus positions within an area just a bit smaller than a DX frame within an FX viewfinder would seem plenty enough for most people, and more than enough for some, which is why Nikon offer an option to reduce that to 11 focus points. It's certainly enough for me.

However, I do see scope for improvement regarding the fundamental accuracy of current autofocus systems, which is why I placed 'autofocus accuracy' among my 6 major concerns, a few posts ago.

Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: Hans Kruse on April 10, 2014, 04:55:56 am
Once you've used a camera that actually can focus precisely in the extreme corners you may discover that what looks like a lack of resolving capability is often just field curvature.

-Dave-

That's easy to test/check. Just use live view and zoom in and move to the corner and focus in LV.
Title: Re: blending AF with fine manual control of where AF is done, and ergonomics
Post by: telyt on April 10, 2014, 07:37:12 am
Why is the image cropped?

It's not easy to keep a fast-flying, highly maneuverable bird anywhere near the center of the viewfinder.

Quote
My point is, it's very unusual to have the main focus of interest in the extreme corner. One can't expect manufacturers to cater to the requirements of a tiny fraction of their clientele whose speciality is breaking the rule-of-thirds. They'd go broke.

Yup, that's my point too.  They make cameras for typical situations.  I'm not interested in making typical photos.

Quote
... the degree of unreliability is proportional to the degree of recomposing, as well as the shallowness of DoF and the closeness of the subject.

And whether the subject moves or not.  Even if it just shifts it weight from one foot to the other the plane of optimum focus moves.  This grouse was preening, occasionally popping its head up out of its back feathers only for a fraction of a second.  Move a focus point to its eye, focus, recompose?  You've missed the picture (full frame, no crop):

(http://www.wildlightphoto.com/birds/phasianidae/sogr02.jpg)

Or catch the drip as the swan is raising its head from ground level to full height?  Focus-lock-recompose means you've missed the picture (cropped from horizontal, this is the left side of the full image):

(http://www.wildlightphoto.com/birds/anatidae/cygnus/trswan01.jpg)

Quote
However, I do see scope for improvement regarding the fundamental accuracy of current autofocus systems, which is why I placed 'autofocus accuracy' among my 6 major concerns, a few posts ago.

IMHO the advances in mirrorless cameras such as the Sony A7 will soon make phase-detect AF systems on DSLR cameras a thing of the past and with good reason.
Title: Re: blending AF with fine manual control of where AF is done, and ergonomics
Post by: Ray on April 10, 2014, 08:37:35 am
It's not easy to keep a fast-flying, highly maneuverable bird anywhere near the center of the viewfinder.

Surely it's much easier with  autofocus tracking

Quote
Yup, that's my point too. They make cameras for typical situations. I'm not interested in making typical photos.

That wasn't quite my point. They make cameras which output RAW images despite the fact that most people don't shoot in RAW mode. They make cameras with a high dynamic range (at least Nikon does), but BJL and Erik will tell you that most people don't shoot scenes which require a high dynamic range capability. In other words, some manufacturers are catering to the relatively few photographers who are striving for technical excellence.

I have difficulty in understanding how a capability to autofocus in the extreme corners of the frame where all lenses, without exception, are least sharp, constitutes technical excellence, or even artistic excellence. If you think it does, then manual focusing is always an option.

Quote
This grouse was preening, occasionally popping its head up out of its back feathers only for a fraction of a second. Move a focus point to its eye, focus, recompose? You've missed the picture (full frame, no crop):

Please explain how manual focusing would be quicker in such circumstances. Perhaps I'm out of practice, but my experience with manual focusing is that there's a greater risk of losing a shot when the subject is moving.

Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: BJL on April 10, 2014, 10:43:23 am
Given Douglass Herr's demonstrated excellence in this sort of photography (I highly recommend a tour of http://www.wildlightphoto.com) I think that both Ray and I could achieve more by learning from him than from a succession of skeptical challenges to his choices of technique and equipment.  Or Ray could show us his accurately autofocused images of falcons in flight or such, describing his technique. (No need for examples from me, because I struggle badly with BIF photography and am here to learn.)

Actually, I plan to start a discussion of techniques for BIF and related wildlife photography, maybe daring to post some of my far less than excellent BIF photos as a plea for help.  Where would be better, "Digital Cameras and Techniques" or "User Critiques"?


P. S.  Ray: I am familiar with using "key repeat" in moving the AF point first horizontally and then vertically (or vice versa) to the desired location; it is still far slower than a direct touch screen tap or moving one's eye to look at the desired part of the VF image. And slower than rotating the focus ring, at least for experienced users of MF: maybe you have got out of practice with fast MF because your particular focusing needs are now well-handled by AF.
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: Ray on April 10, 2014, 11:46:56 am
Given Douglass Herr's demonstrated excellence in this sort of photography (I highly recommend a tour of http://www.wildlightphoto.com) I think that both Ray and I could achieve more by learning from him than from a succession of skeptical challenges to his choices of technique and equipment.  Or Ray could show us his accurately autofocused images of falcons in flight or such, describing his technique. (No need for examples from me, because I struggle badly with BIF photography and am here to learn.)

Actually, I plan to start a discussion of techniques for BIF and related wildlife photography, maybe daring to post some of my far less than excellent BIF photos as a plea for help.  Where would be better, "Digital Cameras and Techniques" or "User Critiques"?


P. S.  Ray: I am familiar with using "key repeat" in moving the AF point first horizontally and then vertically (or vice versa) to the desired location; it is still far slower than a direct touch screen tap or moving one's eye to look at the desired part of the VF image. And slower than rotating the focus ring, at least for experienced users of MF: maybe you have got out of practice with fast MF because your particular focusing needs are now well-handled by AF.

Which is why I asked Douglas to explain how manual focusing would be quicker in such circumstances of fast movement.

If the issue is the rate of missed shots, due to either manual or autofocusing, posting one's best images doesn't necessarily shed any light on the issue.

I use manual focusing only when I can predict that autofocusing will prove inadequate. I also accept that different camera brands and models have different strengths and weaknesses. I find Canon's implementation of LiveView better than Nikon's, for example.

I have a few shots of birds in flight, somewhere, but it's not something I specialize in. If I were to use my Nikon D7100, with Nikkor 80-400, for BIF, I wouldn't be attempting to move focusing squares in the viewfinder. I'd have the camera set to autofocus-tracking and probably use a single focusing square in the centre, activated by the shutter button, and I'd be taking multiple shots.
Title: Re: blending AF with fine manual control of where AF is done, and ergonomics
Post by: Isaac on April 10, 2014, 11:57:07 am
…some manufacturers are catering to…

My impression is that wildlightphoto understands well-enough what is on offer, but feels that what is on offer does not cater to his specific needs.

Additionally, perhaps he feels that we'd all be better-off if we used the kind-of camera that would fulfil his specific needs -- and that's something more open to discussion.
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: BJL on April 10, 2014, 11:58:41 am
Somehow I took your repeated questioning of any arguments and evidence contradicting your apparent belief that AF can do it all as cynicism rather than genuine curiosity.  But if you are curios, let me try to help:
I have a few shots of birds in flight, somewhere, but it's not something I specialize in. ... I'd have the camera set to autofocus-tracking and probably use a single focusing square in the centre, activated by the shutter button, and I'd be taking multiple shots.
As already mentioned in this thread:
- using a single AF point (central or otherwise) can be unacceptable with fast and erratically moving subjects that are hard to get under the selected AF point, and
- autofocus-tracking and multiple AF point can get focus on the bird as a whole, but not reliably on a more specific target like the eye.  See the examples above where the eye is in focus but not all of the bird.
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: Ray on April 10, 2014, 07:14:02 pm
Somehow I took your repeated questioning of any arguments and evidence contradicting your apparent belief that AF can do it all as cynicism rather than genuine curiosity.  But if you are curios, let me try to help:As already mentioned in this thread:
- using a single AF point (central or otherwise) can be unacceptable with fast and erratically moving subjects that are hard to get under the selected AF point, and
- autofocus-tracking and multiple AF point can get focus on the bird as a whole, but not reliably on a more specific target like the eye.  See the examples above where the eye is in focus but not all of the bird.


BJL,
Do you really not see the illogical implications in what you write above, in connection with manual versus autofocus? Fast and erratically moving subjects certainly can be hard to get under a selected focus point. However, in my experience, using manual focus in such situations makes it even more difficult. In my experience, manual focusing is best for stationary subjects. I fail to see how one can manually and accurately focus on a bird's eye whilst the bird is moving fast and erratically. But I'm always willing to learn.  ;)

What I suspect has happened here is that certain photographers, including Douglas with his Leica 280mm f/4 APO-Telyt-R, have sacrificed the benefits of autofocus for the sake of a sharper lens. In such situations, of course one would prefer improved manual focusing features over improved autofocus features, because autofocus is never going to be an option with one's existing, and perhaps very expensive, manual focus lenses.
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: ErikKaffehr on April 11, 2014, 01:34:48 am
Hi,

These two images were shot with AF.

(http://echophoto.smugmug.com/Travel/Stuff/i-TnFPSSz/0/L/20080927-DSC00562-L.jpg)

(http://echophoto.smugmug.com/Travel/Stuff/i-6jJxWpN/0/L/20120711-_DSC2283-L.jpg)

Actually, neither is absolutely sharp, but I think they look good both at web size and projected on screen.

Best regards
Erik
Title: Re: blending AF with fine manual control of where AF is done, and ergonomics
Post by: telyt on April 11, 2014, 04:37:12 pm
My impression is that wildlightphoto understands well-enough what is on offer, but feels that what is on offer does not cater to his specific needs.

Additionally, perhaps he feels that we'd all be better-off if we used the kind-of camera that would fulfil his specific needs -- and that's something more open to discussion.

Please do not make assumptions about my beliefs and preferences.  I get more than enough of that from SWMBO.  Ask, don't assume.

What I'd like to see is more choices.  The equipment that meets my needs was discontinued several years ago, several critical repair parts are no longer available, supplies like replacement batteries and compatible memory cards are in short supply, and it can do 2 frames/sec with a good tailwind.

If I believed that a camera with AF best met my needs/wants/desires, I'd have lots of choices from several manufacturers.  Preferring to use a digial SLR optimized for manual focus, my choice is limited to a single discontinued model.  I have no problem with other's preferences, there's a wealth of AF cameras available.  

My primary mission is to educate y'all that there are ways to make properly-focussed images that don't involve focus 'points' (cross-type or otherwise, minimum aperture limits, etc), micro AF adjustment, USM or other motors, trying to figure out if left points are wonky and a host of other technical issues; that there are very valid reasons for preferring manual focus and manual focus need not be limited to stuff that doesn't move.  My secondary mission is to tell y'all you don't have to be so damned dependant on the technology.  The technology is fine in the right context but AF'ers don't need to be so afraid of trusting their own eyes and hands.
Title: Re: blending AF with fine manual control of where AF is done, and ergonomics
Post by: Ray on April 12, 2014, 01:32:58 am
If I believed that a camera with AF best met my needs/wants/desires, I'd have lots of choices from several manufacturers.  Preferring to use a digial SLR optimized for manual focus, my choice is limited to a single discontinued model.  I have no problem with other's preferences, there's a wealth of AF cameras available.  

As I mentioned before in  this thread, it is the reason why you prefer something, which is more interesting than simply the fact that you prefer it.

So far, you've mentioned just one lens that you use, and even boasted about its sharpness in the corners, the Leica 280mm f/4 APO-Telyt-R. As far as I'm aware, this lens could not autofocus even if you wanted it to.

I understand perfectly, if you are in the habit of buying lenses not designed for the camera, because they are the sharpest lenses available, and because maximum sharpness is a priority for you, then autofocus  is either not an option, or, if it is an option, it might not pass muster in its current state of development, especially if you frequently use telephoto lenses at wide apertures and/or insist on achieving the maximum sharpness that your lenses are capable of.

However, this is not my situation. Having taken photos with film cameras for a good many years, at least sporadically, as an amateur, I recall that some of the major improvements in camera technology which tended to renew my interest in photography, were (1) The built-in lightmeter, which inspired me to buy my first SLR, the Pentax Spotmatic.(2) Autofusing, as in the Minolta 7000 which I bought in the mid 1980's, and (3) Image Stabilisation, as in the Canon 100-400 IS zoom, which I bought with my first DSLR.

I value such features and would not trade them in for some potential, slight increase in resolution. That would be like 'throwing out the baby with the bathwater'.

Quote
My secondary mission is to tell y'all you don't have to be so damned dependant on the technology. The technology is fine in the right context but AF'ers don't need to be so afraid of trusting their own eyes and hands.


Being afraid of trusting my own hands and eyes is not the reason why I use autofocus. It's all about 'getting the shot', for me. There's a lot to be said for 'F8 and be there'.

There is also an issue that too many technical concerns and adjustments, required to be made during the framing of a composition, can get in the way of one's observation of the overall scene. In my experience, during manual focusing, one has to move between focus and out-of-focus on both sides of the target before one can be certain one has the sharpest setting. This not only takes more time than autofocusing, but detracts from one's awarenesss and appreciation of the composition as a whole. It's no wonder you sometimes end up with a bird in the extreme corner of the frame.  ;)

However, I freely confess that I do not specialize in bird shots, particularly birds in flight, or football matches or other sporting events where quick and erractic movement takes place. It might well be the case, if I were to test my equipment in such situations, I would find autofocus tracking inadequate.

Nevertheless, it would be very odd, after declaring the possession of a 100-400 zoom, if I had no shots of birds. Of course, I do have a number of shots of birds. Perhaps not in the thousands, but certainly in the hundreds. Most of them were probably taken with my first DSLR, the Canon D60, with its miserable 6mp.

Here's one such shot, attached, taken in February 2006. It's clear from the expression of the two Frogmouths on the right, that they are very skeptical of your claims for the benefits of manual focusing.
However, the two on the left clearly don't give a stuff.  ;D

Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: Ray on April 12, 2014, 02:32:11 am
However, the above shot in the previous post was not taken with my Canon 100-400, and is not the best test of autofocusing accuracy, because the DoF is quite extensive.

The following shot, was taken with the Canon 100-400 at 400 mm and F 5.6, with the 6 mp D60. No problems with autofocus here, I would suggest. Even BJL should be satisfied with the sharp eye, which, I assume, would have been my intended point of focus.  ;)

The shot was taken in October 2002.
Title: Re: blending AF with fine manual control of where AF is done, and ergonomics
Post by: Isaac on April 12, 2014, 12:53:05 pm
Please do not make assumptions about my beliefs and preferences…

I said what impression I had -- I did not claim they were in fact your beliefs and preferences.


My primary mission is to educate y'all… to tell y'all…

How presumptuous you are!
Title: Re: A Redux Critique Of Contemporary Camera Design
Post by: Ray on April 16, 2014, 06:28:43 am
One point that's very relevant to this issue of manual focusing, is shallow DoF.

Generally, I'm not a fan of shallow DoF if most parts of the main subject are OoF, as in a sharp eye but a fuzzy nose. However, a shallow DoF where a distracting background is OoF can be perfectly appropriate.

Macro shots, for example, of a grasshopper, where its eye is in sharp focus, but its antenna and wings are OoF, don't appeal to me much. I usually feel that I want to see the whole grasshopper, or ant, or spider, because I rarely get to see such detail with my naked eye, and it's interesting.

When searching for bird shots, recently, to demonstrate my point that autofocus is not too bad, one of the more appealing shots I came across I've now attached.

I like this shot because almost everything is in focus. It was taken with the Canon 6mp D60, at F8 and with a 180mm lens, which is about 288mm in full-frame terms.

I guess my taste is different from that of most people. Can't help it. That's how it is.

The birds, by the way, are described as Rainbow Lorikeets, for the obvious reason they are so colorful. However, the Latin name is 'Trichoglossus haematodus', which means 'hairy tongue'.
Title: Re: blending AF with fine manual control of where AF is done, and ergonomics
Post by: ErikKaffehr on April 17, 2014, 04:31:15 am
Hi,

The reason that the Leica R line was stopped that Leica did not make enough money on it and perhaps that they could not afford the development of new digital backs for the R-series.

I know a guy who shoots birds with manual focus on Nikon (I think), but it seems that most photographers work with AF. So lenses and viewfinders are developed for AF and may not be very suitable for manual focus.

I guess this is reality of life.

AF is still in development, so I guess AF will improve year for year.

Best regards
Erik




What I'd like to see is more choices.  The equipment that meets my needs was discontinued several years ago, several critical repair parts are no longer available, supplies like replacement batteries and compatible memory cards are in short supply, and it can do 2 frames/sec with a good tailwind.

If I believed that a camera with AF best met my needs/wants/desires, I'd have lots of choices from several manufacturers.  Preferring to use a digial SLR optimized for manual focus, my choice is limited to a single discontinued model.  I have no problem with other's preferences, there's a wealth of AF cameras available.  

My primary mission is to educate y'all that there are ways to make properly-focussed images that don't involve focus 'points' (cross-type or otherwise, minimum aperture limits, etc), micro AF adjustment, USM or other motors, trying to figure out if left points are wonky and a host of other technical issues; that there are very valid reasons for preferring manual focus and manual focus need not be limited to stuff that doesn't move.  My secondary mission is to tell y'all you don't have to be so damned dependant on the technology.  The technology is fine in the right context but AF'ers don't need to be so afraid of trusting their own eyes and hands.
Title: is good manual focus still a factor in camera choice? hand-held close-ups?
Post by: BJL on April 17, 2014, 11:00:26 am
I share the majority view that most photography, and in particular most photography of rapidly moving subjects like birds is best done these days with state-of-the-art autofocus, and this is intensified by the fact that most modern camera have viewfinders designed with AF in mind, at the cost of being worse for manual focusing than the best viewfinders of the manual focus era.

That said, I still find cases where MF is better than any current AF, including some live subjects that are moving, but not so quickly.  One example is a perching bird or other small animal where
(1) it can be still for a few seconds at a time, enough time to focus manually if the MF ergonomics are good, and
(2) my desired focus point is off-center, and has to be chosen by me, because it can not be reliably identified by the algorithms of any existing AF system.
Actually, the case that I encounter most often is hand-held nature close-ups, where DOF is shallow and the strategy that works best for me in some cases is to first set the focus of the lens roughly (maybe with AF-S) and then adjust the camera position slightly until the desired part of the scene comes sharp.

For this, my ideal would be a moderately enlarged live view image, maybe 2x: low enough that any "rule of thirds" focal point is still in view and yet high enough that the EVF image is bigger and more detailed than the OVF of any AF camera can give.  Unfortunately most cameras I know of do not offer this, instead jumping from 1x to 5x or more.  The only options I know of for this modest VF magnification are:
(a) using the 2x "digital teleconvertor mode" of models like the OMD EM5, but then using the raw file, which is the full uncropped image (only the JPEG is the 2x crop seen in the VF.)
(b) get a Pentax 645Z, with its live view magnification options of 2x, 4x, 8x, etc.
Obviously, each of these has significant drawbacks; can anyone suggest other options?

Title: Re: is good manual focus still a factor in camera choice? hand-held close-ups?
Post by: Isaac on April 17, 2014, 12:08:11 pm
…cases where MF is better than any current AF…

There are cases where I can MF "better" than the actual AF of my camera: