I don't necessarily agree that camera manufacturers are using MP numbers as a sales gimmick. Well, some may be (maybe point- & shoot cameras, cell phones, etc.)
That was a large part of the statement, that the bulk of camera users are consumers printing no larger than 8x10 and those consumer devices are marketed largely by megapixel, as a marketing tool along the lines of computer manufacturers pushing clock speeds. There were other caveats in the article regarding those who do print larger and the need for higher megapixel cameras.
but those who crop their images benefit from high MP counts.
Again even a doubling of megapixels does not give that much more room to crop, and I can tell you from experience, the more you shoot professionally the more you find you have less time to tinker with photos which aren't "right" or "near right" out of the camera. If one finds themselves routinely needing to significantly crop photos, perhaps their process should be reconsidered because taking the time to correct what should be taken right in the first place (and let's be honest, one of the advantages of digital is the ability to shoot several variations at no real cost) or shooting at sizes twice what one really needs only "in case" they need to crop, and dealing with the resultant processing on such files is a wasteful approach to one's time and resources.
In my view, high MP sensor production will continue along high ISO noise reduction technology, and so smaller hard drives of larger capacity and speed. Perhaps a few of us don't ever print images larger than 8" x 10", but a lot do, specially advertising agencies that print murals and display images in giant screens.
Billboards, murals, movie posters, etc. are generally not shot significantly higher resolution than any other photo because they aren't intended to be view inches away. Take a close look at one of the 4 foot x 5 or 6 foot movie posters the next time you go to the movie theater. You may be surprised how low resolution these actually are.
I am going to say it once here, then repeat myself a little further down
there is a difference between "in theory", "on paper", "technically" versus what is actually real-world practical. These articles are intended in part to avoid going into esoteric discussions which have 1% relevance and rather focus on a practical, real-world approach based on my current experience and understanding (which is always subject to change mind you).
first, Daniel Browning on the POTN forums dispels the myth that only a few megapixels are needed for a good print here:
The understated utility of smaller pixels
I promised I would repeat myself, but I will phrase it a bit differently. This series of articles is titled, "The Art of Photography". I am trying to give a broad overview of things with a practical bent rather than wallow around in technical esotericism which has little to no benefit in practical use. I am trying to give people an overview of the state of things without implying they should get wrapped up in, or invest in, how many megapixels their camera is, or how fast the continuous shooting rate is, because that is not photography. Too many people spend their time tinkering with their gear, waiting for the next best thing, or thinking if they can squeeze that extra percent of performance of theoretical quality out of their images, it will make their images better...
...and that is what we are talking about here, "theory". It is an interesting read "in theory", "on paper" and "technically", but the reality is the first several points of that post are moot (i.e. megapixels are already well beyond screen resolution concerns, etc.) and the latter points act as though there is nothing else involved in the camera system (the poster does remark there are other factors), as though magically increasing a sensor to the "ideal" 42 MP wouldn't outpace the resolving power of the lens (the poster would need essentially a "perfect" lens), the noise capability of the sensor (obviously there is debate on this), camera shake (none whatsoever would be required which is impractical), pixel "bleeding", whether it would really be any noticeably different than, say, 24 MP etc.
EDIT: Here's a practical example. The post talks about how many megapixels would be needed to print 360ppi. Now, my distance vision is slightly blurry but my near vision is excellent. Several years ago I ran some print tests to see if I could notice significant differences between 240ppi and 300ppi as well as whether there was a noticeable difference between JPEG compression at "High" and "Normal" on Nikon cameras at the time. It turned out I could see a very slight difference...but only with the photo so close to my eye I had a hard time even focusing at that distance..a couple inches or less...I actually ended up damaging my eye slightly trying to discern these differences over the course of a few hours of test prints. I decided while I could see a very slight distance when focusing within an inch or two of the page, there was no practical difference from a regular viewing distance, and even then unless you had the same photo at both resolutions side by side and were specifically looking for the difference, it would be imperceptible...and that's with 240ppi, a full 33% less that the 360ppi the poster proposes. That's how silly the "theory" gets in all this. Who goes into a gallery and views photos from 2 inches away?
Again the poster does mention there are other factors but you can't abstract one part of a system, talk about its theoretical maximum isolated from the rest of the system and present it as practical, because "practical" and "real" involves the rest of the system and the other detriments that such a thing would bring. In reality, might a 42 MP deliver better results than a 8 MP sensor? Sure, but not on any scale worth fussing about. Digital has a lower dynamic range than film, and yet we can easily take digital photos indistinguishable from film if we, as artists and masters of our tools (and any good artist can do great work with mediocre tools) know how to use them. I've seen any number of gorgeous "perfect", "professional" digital photos with todays tools and they would not benefit perceptibly from 3x the megapixels.
Again, the point is to focus on the art and realize there is a diminishing point of return in that 1% of technical discussion, which do not produce results anywhere near the trouble to get there.
Years ago when I was taking music theory, my classmates and peers wanted to sit around all day talking theory, comparing esoteric composers, etc. without actually composing. That is not "The Art of Music". There is a time and a place for it, but it quickly loses relevance if that is all one is doing with it. It isn't "practical" or "real". People in the arts town my photos are shown in, one of the top rated in the country, sit around in cafes discussing the wonders of art...of one photo or painting over another. Meanwhile I am out hiking in Glacier National Park and shooting. Which one is "The Art of Photography"?
second, from a quick glance at your "Aperture" article, you need to get something clear: f-stop (f-ratio) is the ratio of the diaphragm opening to the focal length of the lens. The aperture of a lens is the size of the diaphragm opening, e.g. a 50mm lens at f/2 has an aperture of 25mm.
Yes but the 50mm lens at f/4 also has an aperture of 12.5mm. That was part of my point, that aperture is relative and the actual mm of the aperture is not important, rather the ratio to the focal length is. I was expressing aperture differently, i.e. instead of saying Density = Mass / Volume, I was rewording it as Volume = Mass / Density, which is the same thing. The aperture is, as you say, the size of the diaphragm opening, and because an f-stop is a ratio of that opening to the length of the focal length, an f-stop naturally implies an aperture. I don't know if I am explaining that clearly.
Let me try a slightly different approach, again I am writing these articles with regard to practical understanding, for practical shooting, for practical practice of the art of photography, not merely discussing photography, so when we talk aperture, we have no practical concern for the actual measured mm dimension of the aperture. It is inconsequential to our work. What we care about are the effects of aperture and to achieve those effects and understand aperture, we need only to understand the relative ratio of the aperture to focal length, the f-stop, and that ratio itself is an aperture in the sense that an f-stop of f/2 on a 50mm lens
is an aperture...an aperture of 25mm, but that last bit we aren't concerned about.
Perhaps I should redirect the article to an emphasis on the word "f-stop" rather than aperture. I will have a look at it. I am bound to write things which stray from the conventional way of viewing things and some people aren't going to like it. For instance, I referred to the difference in sensor sizes as relative to the camera's optical system rather than the common analogy of smaller sensors being equivalent to what "full frame" would "see" if "cropped" because that verbage is flawed.
All the best,
Jesse Widener
Art and Structure
