MicrostockGroup
Microstock Photography Forum - General => Off Topic => Topic started by: CJPhoto on November 23, 2006, 06:35
-
Stolen from another forum:
I thought it was interesting so here is the article from NY times tech writer Pogue: (link, [url]http://pogue.blogs.nytimes.com/[/url] ) BTW, test were all done with same camera but printed at different resolutons (down rezzed from native 13 MP) - Will
The Truth About Digital Cameras
. . . Actually, I’ll tell you about one thing right now. We did an episode on digital cameras. Part of the fun involved visiting a couple of big electronics stores, posing as somebody who didn’t know much about cameras, and, later, commenting on what they told me.
The clerks at one store recognized me. The guy at the other store had no clue that I’m a tech writer. Both of them were surprisingly frank, pointing out, for example, that five megapixels is plenty for prints up to smallish poster size.
Now, every time I write that, I hear from furious or baffled readers. “I don’t get it,” wrote one. “A ten-megapixel camera produces photos about 3640 pixels wide–enough to make a 12-inch print at 300 dpi (dots per inch) on a good printer. Sure, you can go lower, but quality is sacrificed; you can’t make an 11×14 print, let alone anything bigger.”
I have to say, the math sounds right. But I also have to say that he’s wrong.
On the show, we did a test. We blew up a photograph to 16 x 24 inches at a professional photo lab. One print had 13-megapixel resolution; one had 8; the third had 5. Same exact photo, down-rezzed twice, all three printed at the same poster size. I wanted to hang them all on a wall in Times Square and challenge passersby to see if they could tell the difference.
Even the technician at the photo lab told me that I was crazy, that there’d be a huge difference between 5 megapixels and 13.
Anyway, we ran the test for about 45 minutes. Dozens of people stopped to take the test; a little crowd gathered. About 95 percent of the volunteers gave up, announcing that there was no possible way to tell the difference, even when mashing their faces right up against the prints. A handful of them attempted guesses—but were wrong. Only one person correctly ranked the prints in megapixel order, although (a) she was a photography professor, and (b) I believe she just got lucky.
I’m telling you, there was NO DIFFERENCE.
This post is going to get a lot of people riled up, I know, because in THEORY, you should be able to see a difference. But you can’t.
And I’m hoping this little test can save you some bucks the next time you’re shopping for a camera.
-
His article is very non-scientific. There are too many unknowns. What type of camera was he using? What type of printer?
But the biggest problem with his "experiment" is that he is down-rezzing an image from 13 MP to 5 MP, when in fact he should be up-rezzing from 5 MP to 13 MP to prove his point. If he did the experiment correctly, he should have taken a photo at three different resolutions and then printed them at the same size.
-
Very non scientific but if the person on the street cannot see the difference, and they are the ones that will be viewing the photo, brochere, billboard etc, does it matter.
there are issues re whether his 5mp output is the same as what a 5mp camera would have delivered (or even a 13mp camera set to a lower resolution) but I dont think you need to up-rez. The up-rez is being done by the photo lab (via lower dpi I assume).
-
well i half agree with you geopappas.
taking a 13 mp image and downsizing it, is only going to make it sharper and have less noise. This is a common thing for many microstock submitters to do if their image doesn't pass the quality control at the stock sites.
However he still has a small point that 5mp is enough for an 11X14 print if the pixels are good quality. take the image at iso 800 with a 5mp and 13mp camera and i think you might notice a difference. HOWEVER. going back to the other side again. A camera with 13mp using the same sized sensor as a 5mp camera... perhaps will produce similar results since squishing more pixels into less space creates noise.
would be interesting non the less to see a more 'scientific' test.. or another test, using different constants.
-
for what its worth, i used to use the sony 707 in my studio, only 5mp. without any resizingon my part, I have made many 20x30 prints through shutterfly that came out great. i was shocked but then got used to it.
these were mainly portraits and landscapes. the only adjusting in PS was the levels. so I can attest to the fact that the sony 707 5mp files can be blown up huge and still look great, would i say this for all 5mp cameras? nope. quality of lens matters quite a bit, just like it used to with film.....
-
oh yes, quality of lens. Look at the review link on top and see the one :-[ review of the two lenses. There is significant deteriation of the image with a poorer lens.
Speaking of low MP's though. I know a few professional wedding photographers who have in the past, and a number are still, using a 4 mp camera (the canon 1d i believe). they make large prints and are happy with the results. I sure like the 11 mp's i have to play with and wouldn't risk shooting at 4mp, but whatever they are happy with i guess, and just goes to show you that it can be done.
it is also a good point that if you have a 4-6 mp camera, perhaps the thing you should be looking to upgrade is your lenses, and worry about the camera later.
-
Talk about misleading information. As a 20 year veteran of commercial fine printing, more will always be better. The med-res images we do let go are usually images that are more forgiving (softer focus, etc). I'd ike to see the same test with lots of details and straight lines in the image. (and interpolated from a 5mp camera as mentioned above)
Lora