« Canons Take a Dunking and Keep on Clicking | Main | Out-of-This-World Camera Bag »

Wednesday, 15 July 2009

Comments

I understood hardly a word of that, but I disagree

I've spent an unfortunate amount of time having to argue with people who think arithmetic is a substitute for reality. Me too, except in the engineering world where I teach computer simulation.

I think what you said can only be considered "on average" as true, and in particular, sometimes false.

"If you're thinking about trying to write a rebuttal to the facts, don't bother. You will be wrong. This is my field of expertise! If you're a surgeon, I wouldn't argue the fine points of your surgical specialty with you. I know what I'm talking about"

Yes absolutely, wouldn't dare rebut such a well known professional as yourself.

As for your post, it must have been very interesting to people who might have understood the meaning of it... count me out. I'm just a lowly Ph.D. Mathematician, and what do they know hey :-)

Could you please go over to the HiFi forums and write a similar - but even more emphatic - rant?

Yahbut, yahbut, yahbut,

warden.

As someone said, for every complex problem there is a solution that is clear, simple and wrong.

Apart from my photography, I also work in the world of bicycles and bicycle component design. If you want ignorant, misinformed, superstitious, pseudo-scientific beliefs and assumptions check the cycling world! It's enough to make you, er, ah me, crazy. Great article.

The problem is not so much that arithmetic is no substitute for reality, but that there is no substitute for (real) knowledge.

The "arguers" are trying to do exactly what you are doing in your columns, except without your extensive knowledge of the topic and underlying physics.

One issue that is quite apparent in your post is that in the end it all boils down to an argument of authority -- and people who are willing to push simple arithmetic way past reason are unlikely to bow down to externally imposed authority.

And no, photography is not more prone to this than any other technical field!


Oh, for the love of every applicable deity, please do write something similar on audio, as suggested by Roberto M!

(If any of you think Leica lenses might cost a tad more than they should, take a look at audio cables. On a related note/rant, phone cable for interconnects and CAT-5 cable, unbraided, for speaker cables, with the teensiest, flimsiest possible plugs, or bare wire, if possible, sound fantastic.)

Huh? :-) Loved it. keep at it.

Amen, Ctein. Could we all stop arguing about how wavey particles move and go outside to take a few pictures?

"Mr. johnston.... I have been following your blog day-by-day for donkeys years now..."

Jan,
Your comment (not published) was quite enigmatic. What exactly do you mean?

Mike

And whose fault is this conundrum? You come to digital photography 5 or 6 years after your readers. You get a "gee whiz this is so cool" attitude as you experiment for the first time with a digital point and shoot camera which most of us did years ago, having long since moved on to better cameras. You make assertions with images that do not really show anything and say to trust your judgment for quality differences. Then, apparently, you get some disagreement with your sudden expertise, and you launch a near incomprehensible rant that boils down to "Shut up, don't argue with me". C'mon. You can do better than that.

I think there are two issues, both of which are prevalent around the world (especially around the internet):

1) People who have models in their head which are incorrect because they over-simplify the behaviour of the system that they are attempting to model.
2) People who have models in their head which are incorrect because they over-complicate the behaviour of the system which they are attempting to model.

The first category of errors is common in photography - mostly because many photographers learn (from books or teachers) rules of thumb to use in the field. In many cases these rules work well most of the time, but are very inaccurate under some circumstances. The second category of errors is more common in HiFi. Here, people believe that the electrical and acoustic properties of their equipment cannot control what they hear, and they try to add extra factors. This leads to cryo-treated power cables, for one example.

These myths often propagate on the internet because people counter them with "believe me, I'm an expert" and "I wouldn't come into your office and tell you about your job". For the newbie this is frustrating, and often leads to people leaving online communities, or just parroting a new set of false assumptions because they are compatible with the beliefs of the online community. I understand that Ctein's ending was supposed to be funny, but unfortunately many "experts" on the internet have exactly that attitude. "Your beliefs are wrong. I am not going to tell you what is right, because your wouldn't understand. You are dumb. I am an Expert!"

On a different note, no discussion of diffraction should pass without a recommendation of Feynman's excellent QED lectures/book.

It was my understanding that there would be no math...

In addition, some things are just plain hard to model.

For example, is it better to judge color accuracy with a colorimeter or with a jury of people? That depends on the application, but for images designed for viewing by people, who cares what the instrument says?

Mike gave a similar example with total harmonic distortion a few months ago. THD only matters if it accurately models sound quality as perceived by people.

Since subjective appraisals give rise to (computer simulated) food fights, one can understand a desire to reduce the comparison to an objective measurement.

It boggles my mind that people then battle over the objective measurements rather than question the model.

quantity vs quality

I sometimes wander how we ever got so far ~(as humans)... when quantity seems to win most of the time

simplification of reality... in photographic subjects it is somewhere between amusing and annoying, depends on the subject and the writer... in politics, economics, war and some other more "important" disciplines it is also dangerous, but there is no getting away from it... "the wall"... which we not only need to live with most of the time, but which rules more often than not.

So if a hen and a half lays an egg and a half in a day and a half, how far does a mouse turd have to fall to split a shingle?

Dear Roberto & Juze,

Oh, nonononono, dem peoples iz CWAAAAZY!

'Sides, I don't know enough about audio to lay down any kind of law. I've been pestering Mike with all sort of noob questions in private.

-------------------

Dear Cyril,

Absolutely! Most of what I know about photography (and science) comes from authority, taking the word of people who know demonstrably more than I do. In my column's list, the only topic on which I can claim primary knowledge is depth of field, where I have derived all the equations myself and examined the results experimentally.

Everything else, it's 'cause my betters told me so and I knew enough to know they were my betters.


pax / knows-it-some-but-not-all Ctein

How's that speech-to-text thingy working? I think you forgot to switch it off near the end...

Dear Winsor,

I hope you're not insisting that seniority=authority. I would certainly never make that assertion.

('sides, I am a VERY fast learner. [s])

But, a factual clarification does seem to be in order. I decided to buy a digital camera only a handful of years ago, but I've been involved, both personally and professionally, with digital/electronic photography since about 1970.

In any case, none of the falsehoods I trotted out are special to digital photography; film photographers were making closely similar ones for years before they even dreamed of silicon. I couched most of them in digital terms to make them more 'contemporary' (Mike even questioned, pre-publication, whether I should even bring up the old film ISO business).

The commonest mistake newbies (people with less than 38 years experience, y'see [vbg]) are likely to make is not realizing how LITTLE difference there is between film and digital photography...

... yet.

It's gonna get very different. Maybe next column's topic, in fact.


pax / Ctein

Dear Clayton,

I could do the math...

... what atmospheric density and what gravity?

...or did you want the *experimental* answer?! You gotta pay for the shingles, then. And some ISO-standard mice.

pax / penurious Ctein

------------------------

Der Chris,

Still mch faster and more accuarte than my typing! (tlak about damning with fint praise... intntionally left this uncorrected, no joke, I;'m really this bad)

pax / fumble-fingers

Nor is arithmetic reality.

Ctein, you rock.
cheers,
john

"Nor is arithmetic reality"

Which never ceases to amaze scientists:

The Unreasonable Effectiveness of Mathematics

All I want to know is which lens is sharpest.

Hey! This sound familiar to me. This happens a lot to me, just happens I work in the computer business.

A lot of time you have to argue with people with use articles or notes about performance, reliability or anything to show you why are you wrong.

The point is near every installation of computers is nearly unique. You have to know the problem to solve it. Nothing replaces the experience.

The computer aren't very arithmetic or logic, do you know?

Bravo!

Dear JCdeR,

For some reason, I'm reminded of an impolite joke from my Caltech days (warning- something in here to offend most techies):

Q: Are all odd integers prime?

Physicist's answer: Let's run an experiment:

3 is a prime, 5 is a prime, 7 is a prime, 9 isn't, 11 is a prime, 13 is a prime. OK let's analyze the data

9 is a statistical outlier and so likely experimental error. So, yes, all odd integers are prime.

Chemist's answer: What's a prime?

Biologist's answer: What's an odd integer?

Engineer's answer: 3 is a prime, 5 is a prime, 7 is a prime, 9 is a prime, 11 is a prime, 13 is a prime, 15 is a prime...

Mathematician's answer: I'm sorry, that's about numbers. REAL mathematicians don't concern themselves with NUMBERS.

[g,d,&r]

pax / Ctein


Dear Ctein,

just one quote:

"Never worry about theory as long as the machinery does what it's supposed to do."

".. a properly ISO-standard-calibrated light meter's going to give them an exposure recommendation based on an assumed 12% equivalent reflectance."

Just curious, where does the 18% come from then ? Why are there 18% grey cards and not 12% (or, at least, 18% is more popular than 12) ? [It is quite possible that I did not understand your point]

...people who think arithmetic is a substitute for reality.

...just a lowly Ph.D. Mathematician, and what do they know...

Why does the word "derivatives" come to mind when I read a photography blog?

Whenever I think that people and the world they inhabit makes any sense I try to remember that Sarah Palin could have become the most powerful individual on the planet.

So without further ado:
My Olympus DSLR has a perfect meter and shallower DOF than any Canon DSLR because it's friggin awesome! Diffraction is just some crazy bullsh*t liberals came up with after they "invented" evolution.

And the moon landing was fake, you can like totally see the wind moving the flag.

;-)

//Anton

It's a bit like some musicians who think that reading music and clinically playing it exactly as written is better than improvistation and feeling.

So what....?!

Confusing food for thought. Convincing maybe for camera shoppers. Certainly not for photographers. Thanks anyway for sharing it.

As every child knows, reality is, of course, logarithmic: http://www.sciencedaily.com/releases/2008/05/080529141344.htm

What we need is flypaper. A site somewhere called "DOF-everything-you-wanted-to-know.com" or something. Then all the people who like to endlessly argue about how many angels can be sharply resolved at f8 on the head of a pixel would go there and chat among themselves. Every once in a while Ctein could drop in, leave a thread bomb that would send everyone off for another 3 weeks. Think of all the disk space it would free up on all the other servers.

I think it should be an internet rule that photos be posted without any EXIF data. When people asked what camera, lens, and f-stop you used, you could answer, "Can't remember", and the yammering would end.

In thirty years, it could have been hoped that you'd have found a more graceful and charming way of making your point (which is an important one.)

I've found a couple of tools to be effective in persuading people to open their eyes, drop some prejudices and prepare to work hard in educating themselves in some technical matters.

One is humour. Another is curiousity. Provocation, used carefully, is a third.

"I'm clever, well-educated, accomplished, worked with smart people and done lots of this stuff. Many of you are not. Many of you keep making these mistakes and it bores me to repeat myself pointing it out."

Is that a fair paraphrase ?

Well, Ctein, I am substantially younger than you with a considerably less prestigious education and fewer accomplishments. Nonetheless: your approach stinks. I'd like to suggest you try one of mine.

However, what interests me more is to know what brought you to such a lazy and off-putting style of discourse, and whether, in fifteen years or so time, I run the risk of following the same path ?

We need curmudgeons, more than ever with Bill Jay's passing. This article wasn't it.

Y

I love it when Ctein gets going. Great stuff, especially your comment on "average".

I had a boss who used to mock the "average" users by saying, "My ass is in the freezer and my head is in the oven, but on average, I feel pretty good!"

John Camp wrote: "Nor is arithmetic reality."

My father-in-law is a mathematics professor, and he always says "Reality is highly overrated."

Yanchik. Agreed.

Much of the furor over technical matters is aggravated by "emotional noise," people getting irritated over charmless aggressive rhetoric. The same is true in politics. There are practitioners of this noisy rhetoric everywhere. It's difficult for people to set aside or work through their differences when they are being (even indirectly or subtly) insulted. We're emotional critters no matter how technically savvy.

Dear Francesco,

Hmmm, I know what I've READ; I don't know if it's true. I have read that the gray card is 18% gray because that's a visual middle gray on the Munsell perceptual scale. Whether that is in fact the reason it is middle gray or it is one of those back-construction rationalizations I rail against, I don't know. Sometimes in the real world of manufacturing, it's nothing more complicated than a design engineer at Kodak who says, "OK, I've got this nice, stable, visually-neutral gray dye, and it kind of falls near the middle of the visual tonal scale, and it will work reproducibly on the manufacturing line, so that's what we use."

Anybody confidently know the REAL answer?

Even more uncertainly, I have read that Ansel Adams was unhappy with Kodak's plan to produce an 18% gray card because he predicted it would confuse people doing metering who would think it was supposed to be the exposure meter calibration point. Sounds too good to be true. But it might be.

----------------

Dear Robert,

Oh no, you couldn't pay me enough!

Well, actually you could, but it would be an awful lot. There are people out there who actually like to rant to make a living. I'm not one of them; it's just occasionally good (and useful) to blow off.

---------------

Dear Yanchik,

Over 30 years, I have made all these points repeatedly, cumulatively nearing 100,000 words. Almost all those words were polite, thoughtful, and explanatory (except for the very rare occasions when I really had to smack someone upside the head before they did themselves or others harm). Doesn't mean I won't rant on occasion. Doesn't mean I won't continue to make these points repeatedly, cumulatively, and thoughtfully in the future.

'Member, a single data point is not a trend.

If you can decide that your style is to never, ever do a rant and stick to it, more power to you.

---------------

Dear David,

I may very well be remembering wrong, but in all fairness to the mathematicians I believe it was a couple of physicists who came up with the idea of derivatives (note: we're talking financial instruments, not calculus).


~ pax \ Ctein
[ Please excuse any word-salad. MacSpeech in training! ]
======================================
-- Ctein's Online Gallery http://ctein.com 
-- Digital Restorations http://photo-repair.com 
======================================

Dear Spiny,

There is a naive and wrong rule of thumb that says it's the lens you paid the most money for.

The correct answer is it's the one the salesperson gets the largest commission on.

On average, of course.

Trust me; would I lie to you? I'll even throw in a UV filter for free...


---------------------------

Dear Cyril,

This one never ceases to amaze me, too! I alluded to that in some of the comments following "A Grand Democracy" -- http://tinyurl.com/d6rgqj

Personally, I think it is the most interesting question in all of science: why the hell does the Scientific Process work so fabulously well? Which it most self-evidently does. This whole super-majority, consensus approach is not inherently a path to accuracy (as many realpolitik uses of the same decision-making mechanism demonstrate), nor is it obvious why the scientific rules of evidence make that approach so accurate in this application. Yet, it works!

This is one of those situations where I strongly suspect that anyone who is sure they know the answer to the question has not yet appreciated the scope and depth of the problem.


~ pax \ Ctein
[ Please excuse any word-salad. MacSpeech in training! ]
======================================
-- Ctein's Online Gallery http://ctein.com 
-- Digital Restorations http://photo-repair.com 
======================================

One possible addition to consider for the next iteration of the article: The failure by some to recognize that perspective is determined by distance, not by focal length.

A few years back there was a media flurry in the US over "Saddam look-alikes." The news outlets could have simply asked any competent photographer why Saddam looked sometimes flatter-faced, sometimes rounder-faced.

The (doubtless honorable) perpetrators of the hoax used mathematical facial measurements to justify their claims, but failed to note differences in angle and distance of the photographers involved.

One of my favorite quotes sums it up for me:

I don’t understand why photographic artists have such a marked tendency to speak mainly of the methods they use instead of the emotions they are able to communicate through their negatives.

Jose Ortiz Echague

A couple of things come to mind:

- On tone of voice: Ctein's grumpiness is entirely acceptable because it's not ad hominem. The problems people have alluded to in internet discourse usually create issues when they are directed at individuals. Ctein's replies to comments here seem pretty civil.

- What I liked a lot about his rant was that it picked up a theme that matters more and more to me--an increasing desire on the part of individual human beings as well as bodies politic, administrative and commercial to make decisions based on rules or metrics (by which I think I mean roughly the same as Ctein's "arithmetic"), rather than mature and reflective judgement. Numbers are often very important (I really, really don't want nuclear power installations run on a lets-suck-it-and-see basis)--but at least equally often they are not.

My working hypothesis is that people are drawn to using metrics inappropriately because using metrics is easier than exercising judgement and because it appears to be more "certain" or "provable". And much of that ease comes from avoiding what Ctein points out towards the end of his piece: "It depends on the two cameras/lenses you're considering. The variance is not ignorable." Balancing the interaction of a set of variables not only makes peoples' brains hurt, it also takes time: metricists love quick and easy (and apparently "provable") answers; everyone else knows it takes patience. For photographers, that probably means taking a lot of photographs and judging the results carefully to see if the equipment being used is good enough to realise the desired result and not worrying about some metric of perfection.

I am confused by the concern over the "tone" of the writing. Ctein stated in the very first sentence to prepare for a "rant". That's pretty declarative. After reading that, I do not think it is reasonable to complain that the tone was not balanced or serious enough.

Hughlook,

On tone of voice: no, Ctein doesn't get off my hook that easily. Whether "ad hominem" can be distinct between plural and singular is by-the-by. That's not my beef anyway.

Sometimes at air shows or motor shows one sees a magnificent old vehicle, full of horse-power, thrust, what-have-you, which for some reason or another can't be run up to full throttle, can't be let rip. While it's lovely to see such a beast, there's always a nagging disappointment - "Couldn't she look great, if they'd give her her head..."

That's my feeling on reading Ctein's column. He's not backward in coming forward to express his credentials; that's fine, I'm a "modest" Brit. But actually, I've spent half a generation mediating very difficult technical/commercial explanations between volatile, expressive Russians and quieter cultures. I know a few things that are effective, and a few things that polarise or alienate.

So my hope was to use a little shame to get Ctein to raise his game, because I'm sure he can do better.

To Ctein's "My style": Yes. I'll take that challenge for myself. In the public platforms where I'm to write in cold, black-and-white text for a readership that has scope to reflect on my words, I think I will be able to avoid the simple rant.

Y

Politicians often think that arithmetic is reality! "Only ten people will get killed by my decision which is within the budget/guidelines/whatever so that is acceptable!" I think not.

Personally, for most photographers I'm not sure what this much maths has to do with photography anyway except as in the bad old days of the darkroom to lock out those with lesser knowledge and mystify the process making the photographer look cleverer than they actually are! Perhaps dof, but then there are plenty of pda calculators or pocket reference cards so there is no excuse to get this wrong. Really, leave the maths at home and go out photographing, it's much more fun!

clayton… 33 and 1/3 egg lengths (standard ISO eggs and turds of course and current imported shingles the old ones would throw everything off!).
lol! great rant, lovely conversation that ensued.

Ctein,

If you take requests -
Would the use of a gaussian apodization function rather than the "top hat" function produced by the apertures in most lenses get around the diffraction limit? I'm thinking of something more along the lines of the Minolta STF than a photon sieve.

I assume the answer is no, otherwise someone would be selling it, but there are plenty of good ideas no one sells.

Dear Hugh,

Oh sure, ask me the easy ones [g ].

I am no expert on apodization, so no promises that the following is correct... but here's what I recall about it.

My memory says that apodization can gain you a good stop beyond the normal diffraction limit but that it's not a very information-efficient process. That is, the area under the MTF curve shrinks substantially and you have to shift a marked amount of low-spatial-frequency energy into the high spatial frequencies.

The two problems I see coming from this are that first overall image quality is going to drop visibly, since those lower spatial frequencies have much more of a visual impact. The second problem I see is that if you're suppressing low spatial frequencies and enhancing high ones, the boke is probably going to truly suck.

There may very well be other problems with the idea, but those two are what comes to mind.

Just because someone isn't selling it doesn't mean it isn't a good idea. I've had some very good ideas for darkroom instruments and enlarger modifications, that neither I nor anyone else put into production because it wasn't obvious there was enough of a market to justify the initial costs.


~ pax \ Ctein
[ Please excuse any word-salad. MacSpeech in training! ]
======================================
-- Ctein's Online Gallery http://ctein.com 
-- Digital Restorations http://photo-repair.com 
======================================


I also agree with Yanchik's point about the tone of the writing. Nor does it apply only to this one of ctein's posts. I think it would make for greatly improved reading on this blog if instead of dissmissing the issue as the commenter's error, ctein would try to see what he writes from another point of view.

A full frame camera does not necessarily provide better image quality than a smaller-sensor camera.
And yet, in the real world, every full frame camera does provide better image quality than its half frame contemporary.

An ƒ/2 lens does not necessarily provide usefully shallower DoF than an ƒ/2.8 lens
Quite honestly, if you're going to dismantle myths you better have some substance. "Usefully shallower" means zit.

This is my field of expertise!
This may come as a shock, but you can be wrong even in your field of expertise. It's a realization that comes to most people at some point in their lives.

"I know that I know nothing". Smart fellah this Socrates, eh?
What is in fact edifying is the exercise of humbleness. Arrogance is stagnating.

The comments to this entry are closed.