the f-stop of the human eye
August 31, 2015 6:31 PM Subscribe
Differences between eye and camera: practical implications, Ming Thien
The Camera Versus The Human Eye, Roger Cicala
How We See, Ken Rockwell
The Camera Versus The Human Eye, Roger Cicala
How We See, Ken Rockwell
I guess it's worthwhile to know things about the mechanics of vision, but I've always thought of photographers obsessed with reproducing "reality" as the equivalent of audiophiles and their expensive cables. I'm much more interested in the fringes of what's possible with the medium, and accepting that our brains do a whole lot of subconscious interpreting so "reality" can be defined many different ways.
posted by jeweled accumulation at 7:44 PM on August 31, 2015
posted by jeweled accumulation at 7:44 PM on August 31, 2015
Ming Thein is a goddamn treasure.
His reviews are hands-down the best camera ones out there. They mix beautiful photos with relevant shooting experience and extensive technical info.
The fact he does this as a hobby is extraordinary (IIRC he works for Olympus Imaging) EDIT: I was mistaken, this is his full time gig.
Ken Rockwell is the worst.
posted by lattiboy at 7:49 PM on August 31, 2015 [9 favorites]
His reviews are hands-down the best camera ones out there. They mix beautiful photos with relevant shooting experience and extensive technical info.
The fact he does this as a hobby is extraordinary (IIRC he works for Olympus Imaging) EDIT: I was mistaken, this is his full time gig.
Ken Rockwell is the worst.
posted by lattiboy at 7:49 PM on August 31, 2015 [9 favorites]
Spot focus is my preferred mode, then choose the f stop to either bring everything into focus more with the center of interest most focused; or with a lower f stop to de-cocus every thing but the center of my interest. The human eye is governed by interest and available light, the autonomous nervous system opens that pupil right up to the max in certain circumstances. So the photographers job is to duplicate the focus of interest. This is whether the light is bright or low.
Photographing sunsets well, high dynamic range photos look unreal. Our clever little eyes jump all over the place ultimately putting everything into focus glance by glance. Photographing the sunset is about photographing the sky, but not the sun, and finding a happy medium between the foreground, and the sky.
You can collect imagery artfully in a fashion that is hyperreal, without seeming weird. You have to know what the eye expects vs what the camera can collect. Then there is the pesky director who wants you to look where their intent lies.
posted by Oyéah at 7:51 PM on August 31, 2015
Photographing sunsets well, high dynamic range photos look unreal. Our clever little eyes jump all over the place ultimately putting everything into focus glance by glance. Photographing the sunset is about photographing the sky, but not the sun, and finding a happy medium between the foreground, and the sky.
You can collect imagery artfully in a fashion that is hyperreal, without seeming weird. You have to know what the eye expects vs what the camera can collect. Then there is the pesky director who wants you to look where their intent lies.
posted by Oyéah at 7:51 PM on August 31, 2015
One of the things I loved discovering, when doing an image processing course as part of my EE degree, was considering images as a two-dimensional version of an audio signal, with analogies for dynamic range, sampling rate, frequency filtering, and so on. Coming from a sound background and doing my honours thesis in audio DSP with zero experience in photography, it blew my mind a little. Thanks for posting this.
Also, colour theory is a whole lot more complicated than just RGB values !
posted by other barry at 8:02 PM on August 31, 2015 [4 favorites]
Also, colour theory is a whole lot more complicated than just RGB values !
posted by other barry at 8:02 PM on August 31, 2015 [4 favorites]
Re: seeing image processing as an audio signal - there's probably lots of beautiful ways to see it, I tend to think of photography as analogous to cooking, or the sensation of eating.
When choosing what goes into each dish, and what dishes go into the meal, there's consideration for contrast, saturation, and overall luminance of the experience. Depth of field would be analogous to how homogenous the dish is. There's different schools and styles in both, minimalist, impressionist, modern, traditional. Just like how no dish stands on its own and is meant to be consumed in the context of the meal, a photograph is made stronger by being part of a coherent series.
posted by xdvesper at 8:42 PM on August 31, 2015
When choosing what goes into each dish, and what dishes go into the meal, there's consideration for contrast, saturation, and overall luminance of the experience. Depth of field would be analogous to how homogenous the dish is. There's different schools and styles in both, minimalist, impressionist, modern, traditional. Just like how no dish stands on its own and is meant to be consumed in the context of the meal, a photograph is made stronger by being part of a coherent series.
posted by xdvesper at 8:42 PM on August 31, 2015
The second link missed the point a bit re: the design of the 35mm camera system and specifically how it pairs with a 50mm lens. An image shot with a 50mm lens on a 35mm camera (or "full frame" digital sensor) will, when printed uncropped at 8x10" and returned to the location the image was shot and held at a comforatable veiwing distance, just cover the scene exactly. Pretty simple, but took a bit of clever engineering nonetheless.
For more on the nature of color vision see the Feynman Lectures on Physics (free online ebook), specifically chapters 35 and 36 of volume one...after that it really really gets into the physics of it...
posted by sexyrobot at 8:54 PM on August 31, 2015
For more on the nature of color vision see the Feynman Lectures on Physics (free online ebook), specifically chapters 35 and 36 of volume one...after that it really really gets into the physics of it...
posted by sexyrobot at 8:54 PM on August 31, 2015
sexyrobot: "when printed uncropped at 8x10" and returned to the location the image was shot and held at a comforatable veiwing distance, just cover the scene exactly."
If you print it slightly larger or smaller you will have to hold it at a slightly different viewing distance to cover the scene exactly? It still seems kind of arbitrary to me. I look at things at lots of distances, and most of them are pretty comfortable.
When the photo from a fisheye lens is displayed on the inside of a sphere, that will match the scene for someone in the centre of that sphere. The fisheye lens only looks distorted when you display it on a flat surface. For a flat surface, the distance you view it is always arbitrary. Only a curved surface has a prescribed distance that follows logically from its shape.
posted by RobotHero at 10:31 PM on August 31, 2015
If you print it slightly larger or smaller you will have to hold it at a slightly different viewing distance to cover the scene exactly? It still seems kind of arbitrary to me. I look at things at lots of distances, and most of them are pretty comfortable.
When the photo from a fisheye lens is displayed on the inside of a sphere, that will match the scene for someone in the centre of that sphere. The fisheye lens only looks distorted when you display it on a flat surface. For a flat surface, the distance you view it is always arbitrary. Only a curved surface has a prescribed distance that follows logically from its shape.
posted by RobotHero at 10:31 PM on August 31, 2015
image processing as an audio signal
And here's some audio processing as applied to images. A pitch-shifted kitty!
posted by quinndexter at 11:08 PM on August 31, 2015 [6 favorites]
And here's some audio processing as applied to images. A pitch-shifted kitty!
posted by quinndexter at 11:08 PM on August 31, 2015 [6 favorites]
I look at things at lots of distances, and most of them are pretty comfortable.
You read books an inch from your eyes, or with your arms stretched as far as they will go? and that's comfortable? Most folks hold things about 2 feet from their eyes, most of the time, when looking at handheld objects. (BTW, I've done the trick of holding up the 8x10. It's like looking through a black and white window. It's neat.)
posted by sexyrobot at 11:58 PM on August 31, 2015
You read books an inch from your eyes, or with your arms stretched as far as they will go? and that's comfortable? Most folks hold things about 2 feet from their eyes, most of the time, when looking at handheld objects. (BTW, I've done the trick of holding up the 8x10. It's like looking through a black and white window. It's neat.)
posted by sexyrobot at 11:58 PM on August 31, 2015
He lost me at "but I have a feeling that the eyes are very highlight-biased. "
posted by blue_beetle at 5:54 AM on September 1, 2015
posted by blue_beetle at 5:54 AM on September 1, 2015
sexyrobot: "You read books an inch from your eyes, or with your arms stretched as far as they will go? and that's comfortable? Most folks hold things about 2 feet from their eyes, most of the time, when looking at handheld objects. (BTW, I've done the trick of holding up the 8x10. It's like looking through a black and white window. It's neat.)"
Sometimes I look at things that I'm not holding in my hand. Sometimes things I hold in my hand are not 8" by 10". What if you have a 5x7?
Maybe we should be praising the 8x10 print for being the perfect size for a 50mm lens, rather than the other way around.
posted by RobotHero at 6:31 AM on September 1, 2015
Sometimes I look at things that I'm not holding in my hand. Sometimes things I hold in my hand are not 8" by 10". What if you have a 5x7?
Maybe we should be praising the 8x10 print for being the perfect size for a 50mm lens, rather than the other way around.
posted by RobotHero at 6:31 AM on September 1, 2015
Ming Thein is one of my very favorite photographers. His work is beautiful and interesting. He also has put some of the most interesting essays on the subject of photography onto his blog. Highly recommended. It is funny that this essay should make its way to the blue today, as Ming has hinted in a recent post at another essay on dynamic range and the limitations of the media versus the power of our eyes and brain.
posted by caddis at 8:51 AM on September 1, 2015
posted by caddis at 8:51 AM on September 1, 2015
lattiboy: Ken Rockwell is the worst.
OK, I know Rockwell suffers from being popular, and therefore an easy target, but ... why is he always criticized so heavily? It's not like he's actually worse than Hitler, like Amanda Palmer or Cesar Millan are.
posted by IAmBroom at 8:54 AM on September 1, 2015
OK, I know Rockwell suffers from being popular, and therefore an easy target, but ... why is he always criticized so heavily? It's not like he's actually worse than Hitler, like Amanda Palmer or Cesar Millan are.
posted by IAmBroom at 8:54 AM on September 1, 2015
> Maybe we should be praising the 8x10 print for being the perfect size for a 50mm lens
Except it's not. An 8×10 is a 4:5 crop from the 2:3 frame that is the 135 standard. Perfection would be 8×12.
Real perfection would be circular, of course.
</pedantry>
posted by scruss at 9:39 AM on September 1, 2015
Except it's not. An 8×10 is a 4:5 crop from the 2:3 frame that is the 135 standard. Perfection would be 8×12.
Real perfection would be circular, of course.
</pedantry>
posted by scruss at 9:39 AM on September 1, 2015
*sigh* the 8x10 print is part of the design, just like the sprocket hole distance and etc. It's ergonomics, like tables and chairs and stairs being all about the same height. Maybe you print things all different sizes and look at them on the ceiling. Whatever. I'm telling you what the industry standards consideration process goals were. If I were to hazard a guess, I would assume it had something to do with army surveillance during WW1 and/or WW2.
posted by sexyrobot at 10:22 AM on September 1, 2015
posted by sexyrobot at 10:22 AM on September 1, 2015
sexyrobot: If I were to hazard a guess, I would assume it had something to do with army surveillance during WW1 and/or WW2.
While I don't know, and that seems reasonable, I'd bet the HUGE consumer market drove the design specs.
posted by IAmBroom at 10:48 AM on September 1, 2015
While I don't know, and that seems reasonable, I'd bet the HUGE consumer market drove the design specs.
posted by IAmBroom at 10:48 AM on September 1, 2015
Not sure about others, but Rockwell rubs me the wrong way because he is a perfect specimen of the Dunning-Kruger effect. He's constantly obsessing about brands and what pros use. Did he mention that he's a pro? He applies cargo cult logic to photography. His "award-winning" examples of his best work is just objectively bad.
He has a clear bias against cheap gear (it took me just 15 seconds of skimming to find a recent example, see the Fuji 16/1.4 review where he calls Sigma a "mud brand" that lacks Fuji's experience in expensive optics. Note that Sigma's lenses often outperform first party lenses and at a lower price to boot). He actually claims that investing in expensive lenses outperforms the stock market, and by "stock market" he means picking stocks that failed during the dot-com bubble, not a sane strategy like an index fund.
His site is notable for being long-running. I love vintage lenses but his site is not even useful there because he is incapable of being objective and ignoring price.
He lists actual scale weights of items he weighs himself, so I guess that's nice.
posted by ilikemefi at 11:16 AM on September 1, 2015 [2 favorites]
He has a clear bias against cheap gear (it took me just 15 seconds of skimming to find a recent example, see the Fuji 16/1.4 review where he calls Sigma a "mud brand" that lacks Fuji's experience in expensive optics. Note that Sigma's lenses often outperform first party lenses and at a lower price to boot). He actually claims that investing in expensive lenses outperforms the stock market, and by "stock market" he means picking stocks that failed during the dot-com bubble, not a sane strategy like an index fund.
His site is notable for being long-running. I love vintage lenses but his site is not even useful there because he is incapable of being objective and ignoring price.
He lists actual scale weights of items he weighs himself, so I guess that's nice.
posted by ilikemefi at 11:16 AM on September 1, 2015 [2 favorites]
Thanks, ilikemefi. I get tired of the pile-on "har har we all HATE this person AMIRITE" mentality on the web.
Those critiques are perfectly valid.
posted by IAmBroom at 11:25 AM on September 1, 2015
Those critiques are perfectly valid.
posted by IAmBroom at 11:25 AM on September 1, 2015
I honestly do not want to put any effort into talking about Ken Rockwell, as he is objectively a negative presence in photography and probably the earth itself.
Imagine the type of guy who would say "all lives matter" in a condescending tone to the BLM protesters.
I don't know his politics (and he could be incredibly tolerant), but if you can imagine the kind of smirky, know-it-all guy who would do that you will understand the hatred some have for Ken Rockwell and his writing...
posted by lattiboy at 11:31 PM on September 1, 2015
Imagine the type of guy who would say "all lives matter" in a condescending tone to the BLM protesters.
I don't know his politics (and he could be incredibly tolerant), but if you can imagine the kind of smirky, know-it-all guy who would do that you will understand the hatred some have for Ken Rockwell and his writing...
posted by lattiboy at 11:31 PM on September 1, 2015
→ the 8x10 print is part of the design, just like the sprocket hole distance and etc
I think you're attributing a bit much intelligent design to the process. Dickson and Edison slit Eastman's 70 mm film stock in half, joined it, and perforated it to make 35 mm movie film. “Miniature” camera enthusiasts in the early 20th century used the cheap and plentiful film stock, flipped it sideways to give landscape frames, kinda guessed the field of view to be the same as the eye in using “standard” lenses to make a camera of manageable size. The 8×10" print stock was originally for full plate contact printing from glass negatives.
Like evolution, there's a huge amount of make-do-and-mend in the process, and a colossal amount of historical seemed like a good idea at the time crud lurking about in the standards. Even (smaller) digital sensor sizes are often described in terms of an obsolete standard for video camera tubes.
posted by scruss at 9:30 AM on September 2, 2015 [2 favorites]
I think you're attributing a bit much intelligent design to the process. Dickson and Edison slit Eastman's 70 mm film stock in half, joined it, and perforated it to make 35 mm movie film. “Miniature” camera enthusiasts in the early 20th century used the cheap and plentiful film stock, flipped it sideways to give landscape frames, kinda guessed the field of view to be the same as the eye in using “standard” lenses to make a camera of manageable size. The 8×10" print stock was originally for full plate contact printing from glass negatives.
Like evolution, there's a huge amount of make-do-and-mend in the process, and a colossal amount of historical seemed like a good idea at the time crud lurking about in the standards. Even (smaller) digital sensor sizes are often described in terms of an obsolete standard for video camera tubes.
posted by scruss at 9:30 AM on September 2, 2015 [2 favorites]
Ming's latest blog entry: Thoughts on achieving natural tonality
posted by caddis at 8:28 AM on September 3, 2015
posted by caddis at 8:28 AM on September 3, 2015
What are the limits of human vision?
But let's not sell ourselves short. A million colours; single photons; galactic realms quintillions of miles distant – not bad for the blobs of jelly in our eye sockets, wired to a 1.4 kilogram sponge in our skulls.posted by the man of twists and turns at 9:50 AM on September 3, 2015 [1 favorite]
« Older How companies make millions off lead-poisoned... | ConspiraSea Newer »
This thread has been archived and is closed to new comments
http://journal.smpte.org/content/124/5/56.abstract
"One such study performed on a TV-sized display at Dolby Laboratories found that to satisfy 90% of viewers, a luminance dynamic range of 0.005 to 20,000 cd/m2 would be required. ... We have conducted a study on viewer preferences for the cinema environment isolating the maximum diffuse luminance, the minimum luminance, and the highlight luminance from these other factors. We used a 6 kW cinema projector and 13 ft, 2.8-gain screen to produce a luminance dynamic range of 0.002 to 2100 cd/m2, and concluded that the most demanding viewers prefer 22 f-stops of cinema dynamic range—double the number of f-stops of the existing dynamic range of the cinema."
posted by intermod at 7:26 PM on August 31, 2015 [3 favorites]