C2PS 03 Basics - Sensor size

[Martin Ocando]

Michael and Jeff concurs that the decision for choosing a certain sensor size, and accompanying megapixels, has to do with the intended output. But what if different sized sensors share the same megapixel count? Other than obvious better noise performance, due to lack of photosite overflow, is there any other advantage of say, a 5D Mark 1 12.8MP sensor over a Rebel T3 12.2 MP or a Powershot A1200 12.1 MP? I know there is depth of field as well, but that doesn't have to do with IQ.

Is there an intrinsic quality value for the larger sensor image?

[bclaff]

Martin,

I'm confused by your question.
You seem to be inquiring about Image Quality (IQ) but don't want to discuss the noise differences?
Noise is probably the biggest IQ lever that there is!

Puzzled,
Bill

[Martin Ocando]

Martin,

I'm confused by your question.
You seem to be inquiring about Image Quality (IQ) but don't want to discuss the noise differences?
Noise is probably the biggest IQ lever that there is!

Puzzled,
Bill

No, I'm not ignoring it. I'm saying that noise performance is evident on sensors with larger photosites. What I'm trying to look for is, if there is any other advantage to have a file that came out of a bigger sensor, but not particularly more megapixels, instead of a higher MP file from a cropped one.
My point is trying to understand if an image from an old camera with the same amount of MP but with "old time" noise and IQ performance would present a significant advantage over a new camera, with a cropped sensor, the same amount of MP but with new and improved noise response, surpassing the old camera, or at least getting closer to it.

I mean, the old 1D was an 11MP camera, with high ISO up to 1250. Now, the Rebel T3, for example, is a 12.2MP camera, roughly the same size, but with high ISO up to 6400. Is logical to assume that the T3 have much more improved electronics, photosite lenses, sensor design, S/N ratio, dynamic range and D/A converters, and thus will render cleaner images at 1250, or at least as good as the 1D.
Does the 1D images present any significant advantage over the T3 images?

[ErikKaffehr]

Hi,

A larger sensor is usually better all other factors being equal. A larger sensor collect more photons, that is good for noise. A larger sensor will also have better MTF for small details, assuming that moderate aperture are used with a very good lens on both. I think the MTF stuff is often called microcontrast.

I don't necessarily think that you can tell apart a 24" wide print taken with an APS-C and FF camera if both have around 12MP.

A 12 MP FF camera will need quite strong OLP filtering, or it may produce aliasing artifacts. A smaller sensor camera would use a weaker OLP filter.

Best regards
Erik

Michael and Jeff concurs that the decision for choosing a certain sensor size, and accompanying megapixels, has to do with the intended output. But what if different sized sensors share the same megapixel count? Other than obvious better noise performance, due to lack of photosite overflow, is there any other advantage of say, a 5D Mark 1 12.8MP sensor over a Rebel T3 12.2 MP or a Powershot A1200 12.1 MP? I know there is depth of field as well, but that doesn't have to do with IQ.

Is there an intrinsic quality value for the larger sensor image?

[bjanes]

Michael and Jeff concurs that the decision for choosing a certain sensor size, and accompanying megapixels, has to do with the intended output. But what if different sized sensors share the same megapixel count? Other than obvious better noise performance, due to lack of photosite overflow, is there any other advantage of say, a 5D Mark 1 12.8MP sensor over a Rebel T3 12.2 MP or a Powershot A1200 12.1 MP? I know there is depth of field as well, but that doesn't have to do with IQ.

Is there an intrinsic quality value for the larger sensor image?

If you want to get into sensor analysis, you will have to go beyond C2PS. Other things being equal, a larger sensor collects more photons. However, quantum efficiency, fill factor, micro-lenses, and full well capacity also enter into the equation. The number of photons collected determines the shot noise, which is predominant except in the deeper shadows, where read noise predominates. The large format cameras that Michael and Jeff prefer collect large number of photons, but have high read noise, handicapping their high ISO and shadow performance. For a fuller discussion, see Roger Clark.

Spectral sensitivity is also an important factor, as demonstrated in the comparison of the Nikon D5000 and Canon EOS 500D by DXOMARK.

Regards,

Bill

[Martin Ocando]

Wow, thanks for the Roger Clark link. Amazing read, and clears out my concerns. Definitively sensor technology helps smaller sensors deliver comparable overall image quality than older bigger sensors. I learned that sensor specs and data are sometimes meaningless when you see the end result.

I just got to keep shooting and stop worrying about specs. A vote for Canon for leading in read sensor noise performance.

Thanks for all your help, guys.

If you want to get into sensor analysis, you will have to go beyond C2PS. Other things being equal, a larger sensor collects more photons. However, quantum efficiency, fill factor, micro-lenses, and full well capacity also enter into the equation. The number of photons collected determines the shot noise, which is predominant except in the deeper shadows, where read noise predominates. The large format cameras that Michael and Jeff prefer collect large number of photons, but have high read noise, handicapping their high ISO and shadow performance. For a fuller discussion, see Roger Clark.

Spectral sensitivity is also an important factor, as demonstrated in the comparison of the Nikon D5000 and Canon EOS 500D by DXOMARK.

Regards,

Bill