Larry, you wonder why more megapixels? I appreciate your analysis of pixel size, diffraction issues, etc. (really, I do), but I think you missed the key underlying element: it is all a conspiracy between the camera vendors and manufacturers of hard drive storage units.
You also say "...I also can't understand why suburban housewives in Coastal California where it never snows, need a 5,000 pound four wheel drive SUV, that seats eight, has a 6 liter motor that puts out 300 hp." This is an example of great forethought and a demonstration of the purchasing acumen of American consumers; after all, someday during the 3-5 year lifespan of the vehicle they may have a bunch of kids and may want to drive to Lake Tahoe for a ski weekend and they may need the 4WD. Buying for that possibility means that they avoid regret later. And they avoid the expense of a weekend rental vehicle. And besides, it looks cool. stan On Dec 9, 2012, at 6:07 AM, Larry Colen wrote: > > On Dec 8, 2012, at 2:39 PM, Tom C wrote: > >>> From: Alexandru-Cristian Sarbu <[email protected]> >> >>> Now everywhere I can hear them talking about stuff like growth or >>> doubling the sales in 2013. I'm hearing they hired back R&D engineers, >>> to make up for those fired by Hoya. They have 2-years roadmaps >>> totaling 8 K-mount lenses + 1 TC, 4 645D ones and 3 for Q; and, most >>> important, they're on track with the execution. >>> >> >> Alex, respectfully, I don't have that kind of insider information on >> Pentax, so where do you obtain it? If their road map doesn't contain a >> 24 MP APS-C body, a FF body and lenses, then they're missing the boat. >> I'm sure the market for either of those two cameras would dwarf the >> market for a 645D or a Q. > > > Tom, > > You've mentioned 24 MP a few times. How large are you printing that you need > 24MP? > > Also, what APS-C camera has that resolution? 4000x6000 pixels, on a 16x24mm > sensor, that gives 4um on a side for a pixel. > Using the diffraction calculator at: > http://www.cambridgeincolour.com/tutorials/diffraction-photography.htm > > f/2.8 has an Airy Diameter of 3.7 um or about 1 pixel > f/5.6 has an Airy Diameter of 7.5 um or about 2 pixel > f/8 has an Airy Diameter of 10.7 um or about 2.5 pixel > > The page says: > As a result of the sensor's anti-aliasing filter (and the Rayleigh criterion > above), an airy disk can have a diameter of about 2-3 pixels before > diffraction limits resolution (assuming an otherwise perfect lens). However, > diffraction will likely have a visual impact prior to reaching this diameter. > > In other words, with an AA filter, a 24MP APS-C lens will definitely be > diffraction limited by f/8.0. Without an AA filter, assuming that the lens > is sharp enough, diffraction is probably having an effect by f/4 or f/5.6. > > 4um is 250/mm, which I understand works out to 125 lp/mm. > > Looking at DxOmark > http://www.dxomark.com/index.php/Lenses/Camera-Lens-Database/Nikon/Nikon-AF-S-Nikkor-35mm-f14G/(camera)/485 > Nikkor 35/1.4 52 lp/mm > http://www.dxomark.com/index.php/Lenses/Camera-Lens-Database/Nikon/Nikon-AF-S-Nikkor-50mm-F18G/(camera)/485 > Nikkor 50/1.8 53lp/mm > http://www.dxomark.com/index.php/Lenses/Camera-Lens-Database/Nikon/Nikon-AF-S-DX-Micro-NIKKOR-40mm-F28G/(camera)/680 > Nikkor 40/2.8 is 49lp/mm > > http://www.dxomark.com/index.php/en.../Lenses/Camera-Lens-Database/Pentax/smc-D-FA-MACRO-100mm-F2.8-WR/(camera)/676 > Pentax DFA macro 100/2.8 is 46 lp/mm > http://www.dxomark.com/index.php/en.../Lenses/Camera-Lens-Database/Pentax/Pentax-smc-DA-35mm-F28-Macro-Limited/(camera)/676 > Pentax DA 35/2.8 macro is 46 lp/mm > > It seems as if you want a sensor that has approximately twice the resolution > of a very good prime lens, which correlates roughly with the Nyquist rate, > which would mean that you would never need an anti-aliasing filter, because > even without getting into diffraction limiting, you're already past the > Nyquist rate. > > Is this why you feel that 24 MP is so necessary in an APS sensor camera? > > Bay photo seems to use 250 dpi as a print resolution: > http://www.bayphoto.com/bayweb/pro_fileprep.htm > > Which, interestingly works out to 16x24 inches. However using the scaling of > the graph they use for Minimum Pixel size to yield good prints, they seem to > work with about five printer dots per pixel, which means that with a > 4000x6000 sensor, you should be able to print at 80x100 inches. How often do > you make prints that large? > > On the other hand, lower resolution would give more surface area per pixel, > which would increase both dynamic range, and signal to noise ratio. In my > photography, I find myself running up against these two far more often than I > run out of resolution in printing. By Bay Photo's chart, my K-5's 14 MP is > good for something like a 60x80 inch print. A 5x7 for people who measure > their prints in feet, rather than inches. I might be missing something here, > because I've never printed much larger than 18x24. > > Although you implied market forces require that sort of resolution, which > means that the average photographer who uses their SLR as an expensive point > and shoot is printing that large. Or, are you just saying that it is one of > those marketing things, where people aren't satisfied with less than 300 > cubic inches of displacement in their car's engine, whether or not the > customer would ever use the power, or whether or not a car with a much > smaller engine would outperform it? > > As I see it, I'm either not technically inclined enough, and I am totally > missing the analysis of why16MP, or even 12, is not enough, and why a camera > needs 24 MP in order to be marketable. Or, it could be that I'm far too much > of a geek, and am missing something critical by only looking at the physics > of the situation. The latter is quite possible, because I also can't > understand why suburban housewives in Coastal California where it never > snows, need a 5,000 pound four wheel drive SUV, that seats eight, has a 6 > liter motor that puts out 300 hp. > > > > -- > Larry Colen [email protected] sent from i4est > > > > > > -- > PDML Pentax-Discuss Mail List > [email protected] > http://pdml.net/mailman/listinfo/pdml_pdml.net > to UNSUBSCRIBE from the PDML, please visit the link directly above and follow > the directions. -- PDML Pentax-Discuss Mail List [email protected] http://pdml.net/mailman/listinfo/pdml_pdml.net to UNSUBSCRIBE from the PDML, please visit the link directly above and follow the directions.
