FFrame crop v. 1x6 plus I want it all.....

[mark_sam]

With all the 5DMkII talk it's got me thinkin again. I'm not a brainiac as it relates to the 1.6 crop factor. I have a 40D.

My kids are in High School Sports. I also live in pretty country and like to shoot landscapes. Plus, I like to

photograph animals. So I have the need for a bigger reach as well as the full frame for landscapes. I really don't

understand the 1.6, I know much has been written about it, but my question is more specific. EXAMPLE: If I have a

100mm lens on my 40D, I shoot a photo of a deer for example and then shoot the same pic on the 5D, I know it

appears closer on the 40D due to the 1.6 crop. BUT, if I crop the full frame pic so I see the same dimensions as on

the 40D, is the deer going to appear as close on the full frame as on the 40D (once it's cropped?) A 100mm lens

should ......well, never mind. I'm getting too old. Help.

[chrissyone]

If you crop the image from a FF it will look exactly the same as the one from the 40D. The question here becomes a

matter of pixel density. The old 5D had bigger pixels (or less density) than the 40D, but the new sensor changes that. I

don't know the exact math on this, but it's pretty close to being the same now, so you should be able to crop an image from

the 5D II and get about the same thing as you'd get on a crop sensor.

[tdigi]

We did have a similar conversation on here a while back

 

http://www.photo.net/canon-eos-digital-camera-forum/00PSbq

[ralph_jensen]

Not sure EXACTLY what the question is, but basically, the "full-frame" sensor of the 5D is 1.6x larger than the "crop-frame" sensor of the 40D.

 

For any focal-length lens you use (100mm, for example), the image (of "the deer") projected by the lens onto the sensor is the same size. But the "full-frame" sensor, being 1.6x larger, will show more area around the deer. (It's the same as if you set up a slide projector to project an image of a deer onto a 4' screen and then change the screen to a 6' screen at the same distance: there'll be a lot more room around the deer but the deer will stay the same size.)

 

Photographically speaking, you can crop the "full-frame" image to be identical to the "crop-frame" image --BUT, as Christina alluded to above, if the "crop frame" camera has higher pixel density (more pixels per square centimeter) than the "full-frame" camera - for example, if they are both 12 megapixel sensors but are of different sizes the smaller sensor will have more pixel density - then the image that was cropped from full frame will not be as detailed as the image that was shot with the crop-frame (because the former used only 8 megapixels of the original 12 megapixels, while the latter retains all 12 megapixels).

 

I hope this is clearer than it sounds as I write it! I'm sure there are better explanations elsewhere, including Tommy's link.

[chrissyone]

Update - my math skills may be wonky, but I believe a 10MP crop sensor would be roughly equal to a 23.8 MP full frame

sensor as far as pixel density goes.

[chrissyone]

Sorry, it should be 26.3 MP, not 23.8.

 

Here is a graph to explain this:

 

http://en.wikipedia.org/wiki/Image:SensorSizes.png

 

So a Canon APS-C sized sensor is 329 square mm.

And a FF sensor is 864 square mm.

 

864/329 = 2.63

 

2.63 x 10mp = 26.3mp

[chrissyone]

So, at 21mp, the 5D II isn't quite on par with the 40D's density, but it's pretty close. Other variables like lower noise enter

into the equation as well, so I think it's probably a wash.

[elliot1]

I did the equivalent test with a 12mp Nikon D300 and a 5mp D3 'crop image' at ISO 3200. After resizing the D3 image, both looked pretty much the same. I would image the same would be true for the 5D and 40D.

 

Here is a link to the test comparison shots I made. The top of the first image is from the D300. The bottom shot is from the D3. The 2nd image is the full frame so you can see how much I cropped the images.

 

http://www.photo.net/photodb/folder?folder_id=859843

 

You mention the 5D and the 5DMKII so I am not sure which camera you are referring to . In any case, the 5D ' 'cropped' image would be about equal in quality to to a full 40D image and the 5DMKII would possibly fare slightly better because of the slight increase in resolution.

[mendel_leisk]

*If* the 1.6 crop sensor and the full frame sensor have the same pixel density (pixels per inch or whatever), when you

crop the full frame picture taken with any lens down to by the 1.6 crop factor, you're result is identical to shooting

with the 1.6 crop body. Well, apart from the fact that they're 2 different pictures, time passed etc, *and* those

sensors may have different characteristics, different generation of sensor, etc, etc.

[jason_hall5]

Some one check me out in case I am wrong. 12.7mp / (1.6x1.6) = 4.96mp

 

So the 5D cropped to equal a 1.6 crop field of view, would give you 4.96mp. A far cry from the 40Ds 10mp.

 

The 5DMkII however..... 21.1 / (1.6x1.6) = 8.2mp

 

The even the 5DMkII will not equal a 40D after it is croped to the same FOV it is prety much right on the 30Ds and 20Ds specs.

 

So yes, if you crop the image it will basicly the same FOV wise, however you loose a fair amount of resolution. But to end up with 8.2mp ain't all that bad. Unless you compare to the 50D. :o)

 

Jason

[jason_hall5]

BTW, here is were I found the info....

 

http://www.nicolasgenette.com/Labo/Articles/CropFactor/index_us.php

 

Jason

[chrissyone]

@ Jason

 

I think your math might be better than mine, but we're both pretty close to each other, so we might be on to something...

>.<

[JDMvW]

I really LOVE it when this topic comes up. It's good for at least 5 or 6 minutes of uninterrupted chuckling. I'm not making fun, I mean it literally.

 

Quelle Spaß! C'est frommage. etc.

[chrissyone]

Watching me do math is always good for a laugh.

[scott_ferris]

Hang on Mark you can have it all!

 

I think Christina and Jason's maths is wrong.

 

The 5D MkII sensor is 36mm X 24mm or 864 square mm, the 40D is 329 square mm. The 40D sensor is 38% the size of the 5D MkII, 38% of 21,000,000 is 7,980,000 or 8MP, very close to the 10MP of the 40D BUT the 5d MkII's pixels are bigger, bigger pixels make for far higher IQ.

 

I don't think anybody could tell the difference between a cropped 5D MkII image and a full 40D image, go buy the 5D MkII in peace knowing that it will do everything you want it to :-)

[scott_ferris]

I'm sorry Christina and Jason, I didn't read your full posts well did I? I was too wrapped up in the maths! :-)

 

Looks like we all agree.................

[JDMvW]

Christina, it's not just you. I've done this too on other topics. It's just the intricacy of the arguments, the sudden insights, in short, "the maths"

 

☺

[chrissyone]

<i>I don't think anybody could tell the difference between a cropped 5D MkII image and a full 40D image, go buy the 5D

MkII in peace knowing that it will do everything you want it to :-)</i>

<p>

^^^ I completely agree with this. AAMOF, it's exactly what I'll be doing in a couple of months. =)<p>

As far as the Math™ goes, we're all in the ballpark, I think we just came at it differently. It would settle this to know the actual size of the

40D pixels. The 1Ds III and the 5D II have 6.4 �um square pixels. (whatever the hell that means, and no, I can't figure out how to make that

special "u" with the little tail on it.). The old 5D's were 8.2. Anybody know the size of the 40D pixels?

<p>

BTW, I got these sizes from Ken Rockwell's site.

<p>

<a href="http://www.kenrockwell.com/canon/5d-mk-ii.htm">http://www.kenrockwell.com/canon/5d-mk-ii.htm

</a>

<p>

-c

[scott_ferris]

Hi Christina,

 

The 40D pixel size is 5.7 microns square, so it is 80% the size of the 5D MkII. A 25% increase in pixel size for the 5D MkII over the 40D, I would have thought that equated to a visible improvement that would more than make up for the slight lack in pixels.

[matthijs]

To summarize: the biggest magnification in a 100% crop (with a canon and using the same lens) will be from the upcoming 50D which has the smallest pixels.

 

Assuming that the IQ of the 50D will be adequate for your purposes that's the camera to get to get wildlife the closest.

[brad_farwell]

I think you're all missing an important aspect of this comparison:

 

5 x 10 = 50, Q.E.D, the 50D has *ten times* the pixels of the 5D.

 

Even with the 5D mk2, it would only have 2x it's original pixels, a paltry _1/5_ the number of pixels of the 50D!

 

Why anyone would buy the 5D mk2 (it has only as many pixels as the 10D from years ago!) is beyond me.

 

And besides, a 1.6 crop factor clearly makes the image *bigger*! I mean, you multiply something by 1.6 and you get a *larger* number, not a smaller one. So with that crop factor figured in, the 50D is really more like a 80D! It's the biggest camera out there!

 

(and don't get me started on the 1Ds MK3. SO overhyped.)

 

 

(heh.)

[chrissyone]

lolz!

[don_juan2]

AFAIK, there should be an issue of DOF for the same lens on different sensor. Mathematically u need to move far

away from the subject, if you use 1.6x crop factor sensor to capture same FOV as the FF sensor does. That is why,

physically is it not true that u should get narrow DOF in the 1.6x body (here 40D) for the same setting as 5D?

(I admit i know very little of photography, so corrent me)

[mark_sam]

AAHHH, Consensus at last. The deer looks closer....or does it?

[jason_hall5]

The deer, for the most part, will look the same in both the 40D image and the croped 5D image. That is assumeing the same lens and all other variable are the same. However the draw back is that the final image in the cropped 5D photo is much smaller in size. This means lower resolution and less detail when pixel peeping.

 

Jason