Baldy addresses megapixel wars

Pupator

Gee, I wonder who's correct on this one?? :rofl

http://news.com.com/2100-1041_3-6156398.html?part=rss&tag=2547-1_3-0-5&subj=news

What's funny is that CNET - a "leader" in technology information - is just now asking this question?

DJ-S1

Baldy addresses megapixel wars
See the article here. It's always nice to see members of the Smugmug team recognized as experts in the field to be interviewed for stuff like this.

Pupator

I posted this here: http://www.dgrin.com/showthread.php?t=53434 but didn't get any response. Maybe your title is more inviting.

ivar

I merged them. I do agree that DJ's title is more inviting and appropriate

Pupator

ivar wrote:

I merged them. I do agree that DJ's title is more inviting and appropriate

Check out Ivar, master of compromise! (I wouldn't have had my feelings hurt if mine had just been deleted though - but I appreciate sharing the "glory" with DJ!)

DJ-S1

Funny, I saw your title and almost read it but didn't get around to it. Then when I was looking to see if anyone posted this article, I didn't realize that's what your thread was about! Sorry Pupator!

I linked the article because I get this question all the time from my family members. It's hard to explain to them why they don't need a 8mp point and shoot.

corbosman

Is this supposed to be about P&S or also about dSLR? When I switched from my D100 to a D2X I noticed a huge improvement in print quality, especially in the blue tonal range, which is kind of important in underwater photography. So even though perhaps the 6->12 megapixel didnt make the difference (although I think the 25% increase does contribute), more megapixel usually means a new camera model, with maybe different CCD technology or different software or all of the above, which produces a better end product.

Cor

Pupator

corbosman wrote:

So even though perhaps the 6->12 megapixel didnt make the difference (although I think the 25% increase does contribute), more megapixel usually means a new camera model, with maybe different CCD technology or different software or all of the above, which produces a better end product.

Cor

And that's exactly the point. If they spent more time working on tonal quality, good noise reduction, in-camera sharpening -rather than just seeing how many pixels they can cram on the same tiny ccd - everyone (point and shoot folks especially) would get better image quality.

Malte

Pupator wrote:

And that's exactly the point. If they spent more time working on tonal quality, good noise reduction, in-camera sharpening -rather than just seeing how many pixels they can cram on the same tiny ccd - everyone (point and shoot folks especially) would get better image quality.

I say more pixels is better if it means a bigger sensor, leaving the pixelsize the same or bigger.

Malte

Baldy

I made those comments because Andy and I fight the high-ISO noise wars every day, especially with Nikons. Bad prints almost always come from noise, poor color fidelity, lack of shadow detail or exposure lattitude, almost never from lack of pixels. If it's lack of pixels it's because someone has gone to an extreme crop and then they have other problems anyway.

People were stunned at the quality of Mars Rover pics, which were just 1 megapixel. But those were good pixels...

I was a geophysicist for 16 years and we had cameras circling the earth taking satellite images in the infrared, UV and visible bands. The watchword was good pixels, not a lot of bad ones, and boy did we get great shots.

wxwax

Baldy wrote:

I made those comments because Andy and I fight the high-ISO noise wars every day, especially with Nikons. Bad prints almost always come from noise, poor color fidelity, lack of shadow detail or exposure lattitude, almost never from lack of pixels. If it's lack of pixels it's because someone has gone to an extreme crop and then they have other problems anyway.

People were stunned at the quality of Mars Rover pics, which were just 1 megapixel. But those were good pixels...

I was a geophysicist for 16 years and we had cameras circling the earth taking satellite images in the infrared, UV and visible bands. The watchword was good pixels, not a lot of bad ones, and boy did we get great shots.

You should have a discussion with Snoid, a film devotee, who says digital has a long way to go before it catches-up to the resolution of film. I'd like to read that one.

claudermilk

Baldy wrote:

I made those comments because Andy and I fight the high-ISO noise wars every day, especially with Nikons. Bad prints almost always come from noise, poor color fidelity, lack of shadow detail or exposure lattitude, almost never from lack of pixels. If it's lack of pixels it's because someone has gone to an extreme crop and then they have other problems anyway.

People were stunned at the quality of Mars Rover pics, which were just 1 megapixel. But those were good pixels...

I was a geophysicist for 16 years and we had cameras circling the earth taking satellite images in the infrared, UV and visible bands. The watchword was good pixels, not a lot of bad ones, and boy did we get great shots.

And that's the crux of the argument.

Heck, I've printed a 3.3MP at 16x20 and at the normal viewing distance for that size of a print it looks great--those were all good pixels. The same camera in more challenging settings made 3.3MP of mainly bad pixels. I'd rather have a camera that makes each pixel count every time. My intended upgrade path? 20D->1D MkIIn (note, the same 8.2MP) and I could care less about the 10MP XTi or any 10+MP P&S.