If there was a 100 MP DSLR announced tomorrow, I would pre-order it, then spend many sleepless nights waiting for it to arrive. I’d suffer nightmares where Imatest monsters would kick Zeiss Otus’s around and laugh at their feeble attempts at keeping up with my high resolution camera. It would get even worse when I put a second mortgage on my house so I could afford the new supercomputer to crunch those images. I’d pace the halls of my house, past my favorite matted and framed 24”x36” lens chart prints, while wondering if I needed a car with a bigger trunk that could carry enough CF cards for a day’s shooting.
But it doesn’t need to be that way. Even if they plopped the Lincoln Monument in the middle of Yosemite Valley I could take the architecture in the grand landscape shot with only a 12-16 MP camera and never know the difference between it and my 100 MP dream machine. How can I say this?
Here’s proof. I took these nearly identical images of these trees on Nasim’s recent fall color workshop. One was shot on the full frame 36 MP Nikon D810 with Zeiss Distagon T* 35mm f/2 ZF.2 – a match made in Mansurov Heaven. The other was the duo from hell – the discontinued 16 MP DX Nikon D7000 sinfully saddled with a Nikkor 18-300mm superzoom, Photography Life’s worst reviewed lens ever. I set up my tripod, framed the shots as equally as I could (zooming to 35mm equivalent on the loser lens), then snapped two frames. When I loaded them into Lightroom, the superzoom shot looked horrid – oops, I forgot to remove the split ND filter. Oh well, I dodged it back into decency then exported both files using Photography Life’s recommended web resolution. I then put the images up side-by-side on my laptop monitor (MacBook Pro Retina) and asked Nasim to tell me which was which. After much squinting and head scratching he pointed to one and said “That’s probably the Zeiss”. He was wrong – it was the superzoom.
“That’s not fair,” Nasim groused, “I need to see these at 100 percent.”
And there’s the rub. You don’t need to see images at 100% because that’s not the way final output is viewed. When you look at a Herb Ritts print in a gallery are you obsessing over seeing a 1:1 cropped scan of the original negative to make sure it’s a good photo? Of course not. All that counts in the end is how good the final output looks.
So my 16 MP shot fooled Nasim at final output (2048×1638 pixels) and it looks just fine on the web. But what about 16 MP vs 36 MP for gallery prints or magazine work? Again 16 MP is just fine because even printed at 300 PPI, for the cover of an 8.5”x11” magazine the file only needs to be 8.4 MP. Oooh, but what about a two-page spread? That would take 16.8 MP at 300 PPI. I hate to break it to you, but 16 MP is still enough as the human eye (assuming perfect 20/20 vision) when viewing a magazine just short of arm’s length can only resolve about 150-220 PPI depending on who’s figure you use for the resolution of the human eye (0.6 to 1 arc-second). That is why Apple calls their laptop display a “Retina” display because we view laptops at the same distance as we read magazines and the Retina display has a resolution of 220 PPI (for us geeks out there the new Mac 5K display is 217 PPI which unless you view it at laptop distance is overkill).
Here’s a recent shot of mine as it appeared in the table of contents in February’s Arizona Highways. It was shot with the 16 MP Nikon D4s and cropped to about 9.5 MP and looks tack sharp (I crunched the dimensions and got 218 PPI as final resolution as published). All those two-page spreads that impressed us five years ago were probably shot with 8-12 MP cameras. They look just as good as those today.
The key to this all of this is the concept of “normal viewing distance”. We view shots on our phones closer than we view magazine pages. We view huge gallery prints from across a room, not tucked in a narrow hallway (my lens chart pin-ups excepted). We view billboards from a block away.
The further away your final output is viewed, the less PPI it needs to be printed at. This is why billboards are printed at ~10 PPI. From half-a-block away you can’t tell the difference between 10 PPI or 1000 PPI. A 16 MP image printed at 10 PPI will be 41 feet wide. I’ve only had one photo published on a billboard and it was taken with an 8 MP point-and-shoot and looked fine at 30 mph.
Even though camera resolution has grown rapidly in the last decade, the resolution of the human eye has remained the same. Unless you’re a cropoholic, a forensic specialist, or working for over-demanding clients, anything more than a 16 MP camera is overkill. With the exception of one of the first two shots (I can’t tell which), all of these were shot on 16 MP DSLR bodies.
I hope I’ve saved you a load of money with this article. When it comes to sharp photos, you don’t need to invest in a high-resolution camera or matching state-of-the-art lenses because your eye will never know the difference. Go ahead and laugh at those fools with their 50 MP cameras and 5K displays. Then you can take all that money you just saved and go over to my website and buy some delicious gallery prints. Goodness knows I could use the money. How else will I be able to pay back for my Nikon D810?
All Content © John Sherman
On larger screen very high resolution you will see the difference.
Funny… clever… and educational! :D
Thank you😊
Great article, beautifully explained with lashings of humour. Thank you
Perfect article for me right now. I’ve been thinking about purchasing a used Nikon D4. You helped me make a decision!
My 16.7 Mp 1Ds mark ii failed a few weeks ago. As I am a bit strapped for cash I could not replace it right away so I am using my backup camera, a 1D mark ii N, an 8.2 Mp camera. How does it compare? For a start it is nicer to use, despite being over 10 years old (and older still in design nd technology). But it is more responsive, has a faster frame rate, is quicker to clear the buffer and I would swear that it focusses faster and more accurately too. And of course it is as tough as a very tough thing.
But what about the results?
On-screen … no difference
A4 print … no difference
A3 print … you can just see a difference if you get up close with a magnifying glass
A2 print … The 16.7 Mp image has more visible detail … from 6″ or closer. Further away than that … no visible difference
I shoot mostly at ISO 100 and find 11 stops of DR is enough.
Basically I agree with author, if your pictures used only for web or viewing on display, no need more than 12-16mp resolution. After downsampling to web-size images become perfectly sharp. Even if someone have a 4K or 5K monitor, 8 or 10mp image size will be enough.
But printing is other thing: if we want max possible quality (true 300 dpi for sample) our files must be sharp in pixel-level. That meant resizing original camera files from 2 to 4 times in number of pixels. After my experience with D800e I realized that real resolution this camera is around 16-20mp because Bayer sensors can not resolve each pixel as Foveon sensors made by Sigma. So that, 36mp files of D800e suitable just for A3 printing in 300 dpi quality (downsampled from 36 to 19 mp)
And finally, 300 dpi (or ppi at display) look not enough sharp for many people, so we need updated printing machines for 500-700 dpi quality and more megapixels of modern cameras. For sample: A3 size in 300 dpi transforms to A5 in 600 dpi, and if we want to print 600 dpi on A3, we will need 80 mp files! Personally I think that most professionals in future will be use 150-200mp medium format bodies because 36-50mp is upper limit of the FF sensor in the lens resolution sense.
Very interesting, is anyone printed over 300dpi quality? What impressions? In my country such printing quality not available.
I can’t begin to understand how this degenerated into a war over whether film or digital is superior, and if so, which, why and whatever else. Film is not the subject matter of John’s article.
And for those tearing in either direction over the relative merits of the two media, may I just intrude to mention that anyone and everyone is entitled to have an “opinion” – and also, dare I risk encouraging an outbreak of “opinion-itis”, they’re all entitled to express their opinions. Unfortunately that’s where it stops. “Opinions” are a peculiar obsession that humans develop. They (“opinions”, I mean) are incapable of being either “right” OR “wrong”. All they can do is “agree/coincide” or “differ”. Even more unfortunately, so many sufferers use them to bash other people over the head with – to argue about – and this instantly descends into utter nonsense.
To argue over whether film is better than digital or vice versa is a bit like arguing over whether this horse is a better horse, than that car is a better car. The analogy is a fair one. They are two different means of transport or whatever. Just as film and digital are two different media for reproducing images by using a camera. I used film for over 50 years and stopped. Why? Because I thought that was enough, I’m not the least bit likely to live for another 50 years, and before I expire I want to see what this digital stuff is all about – try it, explore it, and see what I can do with it. I leave a whole blank page on which others can scribble why they prefer one or other mediums. They are simply different. And I wanted to explore and exploit that difference.
If anyone really cares about pixel counts, they should buy a medium format or larger, with zillions more pixels. The guy who invented the sensors used in these cameras has just reinvented them, and come up with one which completely nukes all this discussion about pixels, by capturing individual photons. Sorry film lovers – you can’t ever hope to match that! Anyway all film has grain – and that’s not much different from pixels – so can we all blow our noses, shake hands and start over?
And one for you, Tony Cifani – a 4×5 digital back is due for releases next month with only 6.4MP. Why so few? – because they are larger – and being larger, they perform better than smaller ones. The advance publicity says “Being fat is lovely. People like to reminisce about medium format fat pixel digital backs. Fat pixels are sharp, have great tones and range. The . . . pixels [on this back] are beyond fat, they are humongous. Diffraction is not much of a worry with big pixels.” That amuses me – because it is precisely what I get with one of my cams, with only about 12 or 13MP.
John I was much impressed with the reasoning in your article. You might care to start saving, if you’re serious about pursuing more pixels in an MF format, because a sneaky individual in the trade snitched and told me one major manufacturer is currently thinking of introducing a new MF cam with a huge jump in pixel count – let’s just wait and see if it eventuates! But any student of art knows that viewing distance is just as important as image sharpness. Ansel Adams didn’t have access to the gear you can get today – nobody’s out there complaining about lack of visual acuity in his [film based] photos, or suggesting they should be junked as obsolete. Why? Because of viewing distance. Oh – and his genius as a photographer.
So – there are the choices. Stump up and be a genius – and viewing distance will resolve pretty much all the remaining issues, related to this topic.
Yep. And this too is a true story. I overheard recently a young art director talking to a staff photog at very very large studio: “OH COOL! 5D with the RED “R” on it – YEAH! 50 megapixels!!! We all love megapixels!!! Don’t have to shoot so many versions! Just crop in over here, crop in over there!” Save some money!
(yeh buddy, and I bet your photographer’s shooting at f/5.6 and that little thing you love so much way back there in the corner just ain’t gonna be in focus after you get done cropping down on it).
For decades and decades, photographers mastered the art of “filling the frame” – make use of as much of that film space as possible, even when shooting larger formats. Sacrificing just 5 or 10 percent of the film’s “space” could have a big impact on resulting print quality. And, filling the frame was an important exercise in composition for photography students. Cropping (in the darkroom) was actually frowned upon at school. Some of the Greats practiced this idea almost to the point of religious fervor: Henri Cartier-Bresson, Robert Frank, Sebastião Salgado. I highly recommend viewing their work, preferably in a gallery or museum.
I hope the kids these days are learning some of the cool classic stuff but I really really doubt it. Once they open an image in Lightroom or Capture One Pro or Photoshop, the immediate instinct is start cropping. Don’t get me wrong, I love digital cameras and I love digital image editing too. But, I think film trained the photographer on how to make the most from very little. Digital encourages a very different way of thinking, one where the end result will almost always be achievable because we have that wonderful safety net called Photoshop to fix our mistakes. And, we will be always convinced that there will soon be a better, faster, smarter, slimmer, lighter, multifunctional camera to solve all of our problems.
Megapixels? I like a camera that doesn’t have too many pixels crammed into a small space. (yes, I’m probably a little too concerned about microns and well depth). For a typical “full frame” dSLR, 16 – 21 MP is plenty.
Remember, more and more of your photos are going to be used for web only projects. Companies do still print stuff but every year the amount of
printed materials goes down. Don’t believe me? Talk to someone running large 4 color offset printing presses at a commercial printing firm. My job sends me on “Press Checks” for catalogs. Twenty years ago, I went on a LOT of press checks. Ten years ago, I went on some. Now, very very few if none at all because what we are printing is very low page count. Difficult to justify flights and hotel for an 8 pager.
OK – one more super important point and then I’ll shut up for good. Promise. When your photo gets to experience that rare moment and goes to print – say a catalog cover for a large national retailer, first a graphic designer places your image into layout software. When the layout is all perfect and tidy, graphic designer (or in some cases a graphic artist exports to a highly specialized PDF file and ends it over to the ‘prepress’ department (usually in same building as printer, where they do a lot of complicated stuff to make sure everything’s lined up to go on the correct plates). Now days, this is almost always a digital process – Computer-to-plate (CTP).
Almost all offset printers will tell you the same thing about “resolution” – line screen, how many lines per inch. JUST about all of the time, I hear this”150 to 175 linescreen. VERY simple rule of thumb, divide camera image DPI by 2 = Line screen. 300 dpi / 2 = 150 Line screen (roughly).
BUT, 4 color printing is a messy and sometimes sloppy process, but it can be quiet forgiving too. LPI and DPI are NOT the same. Most pressmen would like 150-175 line screen for high end jobs, but I’ve seen 100-125 line screen that look pretty good (approx. 200-250 DPI). Also, once you factor in dot gain (ink droplets dry out and spread out on paper, producing a very slight darkening effect. Also any slight plate registration issues that aren’t resolved will cover lower quality image issues.
In an ideal world, that catalog cover shot would be fine shot with a high end 16 MB Dslr.
Total number of pixels on sensor (millions): 16.2 MP
Sensor Size, in pixels: 4928 x 3280
Output: @300 dpi – 16.4″ x 10.9″
Catalog cover size: 9” x 12” (on the upper end – mainly fashion and style stuff).
Your in luck! Your super awesome 16.4″ x 10.9″ (300 dpi) file easily works for the 9″ x 12″ catalog cover, which if given a megapixel value would only be 9.72 MP!!! You mean to tell me I could have used a 10MP camera to shoot that 9×12 catalog cover??? If you were extremely careful you might be able to pull it off, especially since four color printing is messy.
OH, and by the way, offset printing technology has not changed that much. The single most important development was Computer-to-plate (CTP) in the mid 90’s. Since then, not much in the way of new stuff to spend money on. AND, the web has advanced a lot recently, but really don’t need more than 6 MP to shoot or the web (assuming a lot of cropping). So, why is it that we’ve seen such enormous jumps in megapixels???????? Who decided this was so important???
I don’t guess you evern need to zoom in on and crop a photo!
The more megapixils the better!
I am still using a 16.7Mp Canon 1Ds Mark II, and I get a lot of ribbing about it, because it has so “few” Megapixels. What is it with these armchair critics? Not only can I easily print to A2 and beyond, but I can hardly tell the difference between photos from this camera and photos from my backup … a “mere” 8.2 Mp 1D Mark II N.
But I recently upgraded my lenses from “prosumer” lenses to some of Canon’s best ‘L’ zooms and that has made a huge difference to image quality in every way, sharper (right into the corners), more detailed, more saturated, and virtually aberration free.