Does higher MP = better crop quality?

Canon RF 24-70mm f/2.8 L IS

12/19/2006 01:00:18 PM · #1

I hope to soon upgrade from my xt to either an XTI or a 30D. I like the build qualtiy and faster fire rate of the 30D. My question would be would the higher megapixel XTI produce better quality crops. I shoot a lot of wildlife and sometimes crop a image quite a bit.

ignite

Canon EOS-500D Rebel T1i

Canon EF 50mm f/1.8 II

12/19/2006 01:09:16 PM · #2

It most certainly will.. More MPs give you more height and width, and therefore you can crop a smaller area and still have a good quality image. Cropping 50% from a 5MP image will leave you with a 2.5MP image. Cropping 50% from a 10MP image will leave you with a 5MP image.

Edit : If you are cropping a fixed portion (like 300x400 pixels) from a 5MP image or a 10MP image, the quality will be the same.

Message edited by author 2006-12-19 13:11:21.

kirbic

Canon EOS R5

12/19/2006 01:24:51 PM · #3

It may, but the difference between the two will be small. Compare the image resolution tests for the two cameras on DPReview.com. That should provide a good reference point, since both tests were done the same way, imaging the same scene under identical conditions.
It is possible for a camera with lower Mpx to yield as much detail as one with a higher pixel count. All depends on the "acuity" of the system. In a perfect world, a camera with an ideal lens would produce a perfectly sharp image. Not usually the case. there's normally some intentional softness from the anti-alias filter, and if microlenses are present their performance is never perfect either.
Bottom line, look at actual test data, don't rely on Mpx count as the whole story.

ignite

Canon EOS-500D Rebel T1i

Canon EF 50mm f/1.8 II

12/19/2006 01:56:19 PM · #4

Rightly said kirbic, but more MP does give you the ability to crop more and yet retain enough pixels to print better. Cropping a lot from a high MP camera will still allow you to print larger sizes than would cropping from a low MP camera.

12/20/2006 10:27:20 AM · #5

The lens matters also. If, say, an 8MP sensor out resolves a given lens, then a 10MP sensor will be no better. On a DSLR like the OP mentions, the answer will depend on whether one uses the kit lens or "L" glass.

The OP mentioned wildlife. Here, the AF system may matter more than the MP (both speed and accuracy). And a teleconverter or longer lens may be better than a body upgrade.

Funny that most P&S camera makers don't talk about how good their lens is, just how many MP there are :-)

Message edited by author 2006-12-20 10:30:38.

Canon EF 17-40mm f/4.0L USM

12/20/2006 10:43:23 AM · #6

I Use 100-400 L for most wildlife and Sigma 105mm for macro, both cameras have the same 9 point A.F. system, so it is basically it comes down to build quality, rate of fire and mpixels.

asimchoudhri

Canon EOS-5D Mark III

12/20/2006 10:43:48 AM · #7

I haven't seen a good analysis of the noise levels on the new xti sensor. With an identical sized sensor, more megapixels means more noise (quantum mottle), which requires a higher quality circuitry to maintain an adequate signal-to-noise ratio. The 5D has more megapixels than the 30D, but also has a larger area and the area covered by each pixel sensor is similar... thus you get more megapixels without increasing the noise. 30D and xti have the same sensor size, so therefore each pixel on the xti sensor has a smaller area... .smaller area means fewer photons detected which (as mentioned above) requires more advanced circuitry to avoid significant increases in noise. Slightly fewer megapixels on a 30D will not limit your photos as much as more noise.

Separate from the circuitry improvements needed to get better performance from smaller pixels, the resolving power of various lenses in line pairs per mm starts to become a factor. If the lens can't focus light with a certain degree of accuracy, it doesn't make much of a difference how many pixels you capture it with.

I don't know the real-world performance of the xti with respect to noise, but I would suggest doing tests to see if this will have a noticible impact on your type of work with the types of lenses you will be using (assuming you can temporarily borrow the different camera bodies). In summary, I don't think it's a good idea to focus just on the megapixel number. Many people think that the xti is a way to work on improving their sensors with smaller pixels without there being an adverse impact on the prosumer and pro lines.

Edit: no change to above text, but the very literal part of me had to add the following addendum:

There are two main types of noise involved with the sensor (forget about the lens and other stuff).
quantum mottle is a physical principle which is related to the number of photons detected. this will only be impacted by the number of photons hitting the sensor. less light means more quantum mottle. smaller pixels means more quantum mottle. there is no way around this. This will always be there. With a given sensor you can only control the amount of light hitting the sensor, so you can decrease quantum mottle by increasing lighting on your subject, extending the exposure, or widening the aperture. Technically, you can also decrease the impact of quantum mottle by using a lower resolution for acquiring the images... if your sensor is 2000 x 1000 but you take the image as 1000 x 500, four sensor pixels will be summed together to give you one image pixel, and the uncertainty/random variations that lead to quantum mottle will be decreased. Downsampling on your computer has a similar net effect.

Then there is noise within the circuitry. This involves how efficient the sensor is at detecting pixels, and how cleanly it can send that information to the image processor and data storage medium. Improved quality of the components of the sensor can decrease this type of noise. External factors, such as temperature, can also impact this type of noise (warm sensors can have more noise). It is this part of the "noise" equation which manufacturers work on improving to get higher quality sensors with more megapixels.

(I could go on... but i won't... yet...)

Message edited by author 2006-12-20 11:16:11.

scalvert

Canon EOS-5D Mark III

Canon EF 24-105mm f/4.0L IS

12/20/2006 10:45:48 AM · #8

Save up for a 5D and it will be a non-issue. ;-)

kirbic

Canon EOS R5

Canon RF 24-70mm f/2.8 L IS

12/20/2006 10:54:38 AM · #9

Originally posted by ignite:

More Mpx is only useful if the acuity is the same (or greater). An image with more Mpx that has lower acuity might actually have the same or less detail than the image with the lower pixel count. In that case, there's no crop advantage.
Packing more Mpx into the same sensor taxes the lens to a greater degree, making that part of the optical system more critical.

12/20/2006 11:06:16 AM · #10

Originally posted by scalvert:

Save up for a 5D and it will be a non-issue. ;-)

But then he'd have to crop even more because he loses the "crop factor" :-(

Canon RF 24-70mm f/2.8 L IS

12/20/2006 11:08:11 AM · #11

Originally posted by JOHNBOY1970:

I Use 100-400 L for most wildlife and Sigma 105mm for macro, both cameras have the same 9 point A.F. system, so it is basically it comes down to build quality, rate of fire and mpixels.

The 1.4x teleconverter will help with wildlife, and its cheaper than either body. If you buy it now, it may tide you over until the 30D successor is released.

kirbic

Canon EOS R5

12/20/2006 11:30:25 AM · #12

Originally posted by hankk:

The 1.4x teleconverter will help with wildlife, and its cheaper than either body. If you buy it now, it may tide you over until the 30D successor is released.

Putting a converter on the 100-400 will prevent AF, unless it's either a 3rd-party converter that does not let the camera know it is there (like the cheaper Tamron) or unless pins are taped to disguise the converter. In either case, AF will only function in very good light.

scalvert

Canon EOS-5D Mark III

Canon EF 24-105mm f/4.0L IS

12/20/2006 11:38:30 AM · #13

Originally posted by hankk:

Originally posted by scalvert:
Save up for a 5D and it will be a non-issue. ;-)

But then he'd have to crop even more because he loses the "crop factor" :-(

I was mostly teasing, but you WOULD have 12.8MP, and the cleaner images produced by a 5D can withstand more enlargement than those produced by noisier sensors. That said, I suspect the Rebel XTi woud currently be the best bet for lots of zoom and cropping. The main advantage of the 30D for wildlife would be the faster frame rate.

faidoi

Pentax K10D

Pentax DA 18-55mm f/3.5-5.6 AF

12/20/2006 01:28:05 PM · #14

Originally posted by scalvert:

Originally posted by hankk:

Originally posted by scalvert:
Save up for a 5D and it will be a non-issue. ;-)

But then he'd have to crop even more because he loses the "crop factor" :-(

I was mostly teasing, but you WOULD have 12.8MP, and the cleaner images produced by a 5D can withstand more enlargement than those produced by noisier sensors. That said, I suspect the Rebel XTi woud currently be the best bet for lots of zoom and cropping. The main advantage of the 30D for wildlife would be the faster frame rate.

Yeah, save up for a Canon EF 1200mm f/5.6 USM or something.

12/20/2006 03:10:49 PM · #15

Originally posted by faidoi:

Originally posted by scalvert:

Originally posted by hankk:

Originally posted by scalvert:
Save up for a 5D and it will be a non-issue. ;-)

But then he'd have to crop even more because he loses the "crop factor" :-(

I was mostly teasing, but you WOULD have 12.8MP, and the cleaner images produced by a 5D can withstand more enlargement than those produced by noisier sensors. That said, I suspect the Rebel XTi woud currently be the best bet for lots of zoom and cropping. The main advantage of the 30D for wildlife would be the faster frame rate.

Yeah, save up for a Canon EF 1200mm f/5.6 USM or something.

I think Zeiss came out with a bigger one :-)

One of the issues with cropping vs bigger lens is exposure. The tighter you get in on the subject, the less the surroundings will affect the metering points (think of an animal in a shadowed part of the woods, on a bright day or a dark bird against a bright sky). Even if you use spot metering, the covered area becomes smaller with a longer lens.

12/20/2006 06:22:01 PM · #16

I already have the 1200mm 5.6 and rarely use it. It it not as light as my 100-400. I do a lot of in flight shots.