Camera crop factor (1 Viewer)

[pe'rigin]

Mark’s Right,


Somewhere with the onset of the digital world we have acquired these misleading words ‘crop factor’.

PROFESSIONALLY IT MEANS NOTHING….

. more than the part reduction of an original.


All you need concern yourselves with is resolution, which will give you an enlargement factor (magnification) at the correct dpi for your output device.


At the higher end of reproduction on digital images the ability to enlarge is restricted, especially when compared to film.

It doesn’t matter a figs dot whether you have small or large pixels, because the difference is minuscule at the higher printing.

 

[Photovisions]

Mark’s Right,


Somewhere with the onset of the digital world we have acquired these misleading words ‘crop factor’.

PROFESSIONALLY IT MEANS NOTHING….

. more than the part reduction of an original.

Correct, thats my point and my concern is that some think they are getting extra magnification, which they are not!


All you need concern yourselves with is resolution, which will give you an enlargement factor (magnification) at the correct dpi for your output device.

Agreed again


At the higher end of reproduction on digital images the ability to enlarge is restricted, especially when compared to film.

There are many 35mm Digital camera's that can now match 35mm film and what about the Hasalblad camera's with 39MP?


It doesn’t matter a figs dot whether you have small or large pixels, because the difference is minuscule at the higher printing.[/

The difference is small by a factor of 1.44 comparing the largest and smallest pixels in modern day digital camera's, but to achieve an A2 print without pixelation relates to the total resolution of the camera. For example I can get fairly good A3 prints with my 5D, but a bird in the frame is often quite small.
Adrian

 

[gordon g]

Sorry, couldn't resist. Don't get me wrong, I'm enjoying this thread ( and learning something from it as well ) but really, I mean, we are are talking about taking photographs of birds here and to say that hairs are being split is putting it mildly, particularly when you put this debate up against fieldcraft , that is getting a bit closer to the bird! So OK, I guess some or all of you may already be pretty good in the fieldcraft stakes but what about composition? What are we looking for here, a technically 100% perfect ( or as close as theoretically posible) picture of the bird or birds in question or a pleasing image, that is an image that is aethetically pleasing and/or containing information about the birds behavior or biology? Crop factor, pixel density, chroma/luminance noise ratio is always good fun to debate and long may it continue but at the end of the day, what bearing does it have on taking a good, well composed, photograph?
Apologies once again.

Chris

Thanks Chris - it was good to be reminded of why I actually use a camera in the first place!

 

[mjmw]

I dont think I can explain this any easier, unless I showed two photos side by side.
Adrian

Adrian, I do understand what you are saying and mainly agree with you, but I just want to seperate resolution of the recorded image from camera crop factor.

Going back to my example of the 20D and 1D2N - both have roughly the same number of pixels, but the 20D records less of the image from the lens, because the sensor is smaller...so the resulting images have the same resolution and the 20D image is magnified compared to the 1D2N's.

So, the camera crop factor has not reduced resolution...thats the point I was trying to make.

People have seemed to compare camera crop factor with cropping pixels in post-processing - they are different things and have different effects on the image's resolution.

 

[kevindurose]

Doesn't this all depend on which is the limiting factor; which could be either the lenses optical resolution or the cameras sensor. In theory if the classware was optically perfect there would be no limit on the recordable detail if the sensor was up to it?

 

[tjsimonsen]

Doesn't this all depend on which is the limiting factor; which could be either the lenses optical resolution or the cameras sensor. In theory if the classware was optically perfect there would be no limit on the recordable detail if the sensor was up to it?

Well, the ultimate limit is the wave-lenght of visible light 3
But that's probably a bit hypothetical: a macro lens with a mag power of 400:1 is in theory possible, so we still have quite some way to go....

Sorry, couldn't resist it :-O

Thomas

 

[mjmw]

There are many 35mm Digital camera's that can now match 35mm film and what about the Hasalblad camera's with 39MP?

The sensor on the latest Hasselblad H3D is 48x36 mm, so has some of the 'opposite crop factor' I talked about earlier...and going this route, you will only get pe'rigin started about 8x10 sheet film 3

 

[RAH]

Want to get people started - isn't this discussion related to the Sigma Foveon chip - i.e. quality of pixels versus quantity? Talking about the merits of this chip often gets folks frothing at the mouth!

 

[Photovisions]

Want to get people started - isn't this discussion related to the Sigma Foveon chip - i.e. quality of pixels versus quantity? Talking about the merits of this chip often gets folks frothing at the mouth!

Yeah, foveon technology has been around for years, but hasnt really taken off. I dont know how many Sigma's with foveon chps there are out there, but look at this forum and I havent seen one yet.

Two points I would like to make about the Sigma SD14:
1. The resolution is quoted at 14MP, but because of the foveon technology with overlaying RGB layers, the physical size is really quite small 21x14mm equating to a 1.67 crop factor!

2. Because colour data is recorded by an effective single pixel compared with a Bayer matrix of 4 pixels, the foveon sensor should offer about 2x increased resolution and 4x increased pixel density.

I would love to borrow an SD14 to try it out. Reviews say it has excellent colour rendition and improved detail, but a relatively small chip.
Adrian

 

[Photovisions]

Adrian, I do understand what you are saying and mainly agree with you, but I just want to seperate resolution of the recorded image from camera crop factor.

Going back to my example of the 20D and 1D2N - both have roughly the same number of pixels, but the 20D records less of the image from the lens, because the sensor is smaller...so the resulting images have the same resolution and the 20D image is magnified compared to the 1D2N's.

So, the camera crop factor has not reduced resolution...thats the point I was trying to make.

People have seemed to compare camera crop factor with cropping pixels in post-processing - they are different things and have different effects on the image's resolution.

Comparing the 1DMkII and 20D is difficult because the 1DMkII has the bigger sensor, but also the larger pixel size, so in fact the 20D will offer better pixel density for an object and hence resolution, but the difference is not large. I would have the 1DMkII any day for its better focusing, speed etc.
Adrian

 

[pe'rigin]

The sensor on the latest Hasselblad H3D is 48x36 mm, so has some of the 'opposite crop factor' I talked about earlier...and going this route, you will only get pe'rigin started about 8x10 sheet film 3

Morning Mark,

I love film, bring it back!

Going back to resolution and what this thread is really about, the ability to enlarge an image or part of without the loss of definition.

35mm film is almost dead, digital will eventually replicate what film has given us.

If I could take two issues where film had an advantage, colour and the ability for very big enlargements. Digital is slowly addressing these, maybe on colour they can do a lot better, especially Canon.

On the Haselblads, we are talking a different ball game, I use photographers with this gear and their investment has been huge, (so are their fees). So we have to park that to one side and concentrate on what the average chap (chappess) can afford on this forum.

To get on an equal footing with film enlargement, I think that we will see larger mega-pixelled chips come in, hopefully through the entire range. Let's hope that they price things more sensibly as well.

Once these chips are available to everyone, maybe we can confine the wretched words 'crop factor' to the waste bin.

 

[gkrpepper]

Hi, all

Am I missing something?

I always thought the only reason "Crop Factor" exists is because a Digital Camera is compared to a 35mm Full Frame camera.

Which to me means:

SMALLER SENSOR + SAME LENS = SMALLER FIELD OF VIEW (as compared to a full frame camera)

Which to me means if I enlarge an image to the same size as a "full frame" print, the smaller sensor will have to be enlarged more, giving the impression of making it bigger.

Which also means Adrian Canon's D5 Camera doesn't have a crop factor

Am I wrong?

 

[RAH]

SMALLER SENSOR + SAME LENS = SMALLER FIELD OF VIEW (as compared to a full frame camera)

Which to me means if I enlarge an image to the same size as a "full frame" print, the smaller sensor will have to be enlarged more, giving the impression of making it bigger.

The first part, in capitals, is correct. However, if the smaller sensor camera has the same number of megapixels as the full-frame camera, you don't have to "enlarge" the image out of the small camera to bring it up to the size as the full-frame. If both cameras are the same MPs, both cameras produce the same size image. But the image out of the small will look closer because of the smaller field of view. This is why some folks (e.g. me) say this magnifies any lens. This is what the argument is all about.

 

[mjmw]

This is why some folks (e.g. me) say this magnifies any lens. This is what the argument is all about.

Me too...but we may be the minority...

 

[tonky]

I've totally lost track now of who's supporting which argument here and exactly what the argument is. |:S|

It took me a while to come to terms with the fact that my 20D is producing notably better small birds shots with more fine feather detail than the 1DMK2n, but it is. Both cameras produce exactly the same size image on the sensor, but it's the size of that small bird when you view the two images at 1:1 reproduction where the 20D really scores heavily, with a larger image of the bird and a significantly larger physical file size when both images are cropped to produce identical framing.

The point has been well made earlier that there's no substitute for getting closer to the bird regardless of which lens you use, and it's quite true, although that can bring it's own DOF problems, and it isn't always possible to get as close as you'd like. Some of my sharpest most detailed small bird shots have been taken with the Canon 180L Macro lens at a range of around 5 or 6 feet, but the fact is that even at that sort of range, regardless of which body I use, the images still require quite substantial cropping, so the 20D still wins and ends up with the larger file size after cropping.

The problem is largely peculiar to very small bird shots of course , (or very long range shots), and the differences become less marked with larger birds where you can fill the frame better.

So do I regret forking out the wonga for the 1DMK2n as soon as it came on the market? Of course not, it's built like a camera should be and makes the 20D feel like a toy in comparison. There are also occasions for flight and diving shots where the faster AF of the 1DMK2n is going to be the obvious choice and it will still AF with the 2X converter whereas the 20D wont.

It has changed my thinking though on my next body purchase. The results from a test of the 40D and the 1DMK3 would yield larger 1:1 images in each case, but the 40D would still win the crop comparison by a similar margin. The 1DSMK3 with 21 million pixels would be a great all round camera, superb for weddings portraits and everything else and still not be very far behind the 40D in cropping superiority, but it's £6K!

The 40D is looking better to me all the time! ||

 

[Mil]

Also small contribution from me to this topic. One ask above how much we crop pictures? I never crop more then up to 20% (cut out).
Best!
Milan

 

[tonky]

Also small contribution from me to this topic. One ask above how much we crop pictures? I never crop more then up to 20% (cut out).
Best!
Milan

LOL. I often crop a lot more than that Mil!

This shot is an uncropped shot taken at around 6 feet - probably more like 5 feet, with the Canon 180L Macro on the 20D body. This is about as good as it gets for for me for filling the frame with birds of this size, regardless of which lens I use.

http://www.ctomkinson.com/imgs/gallery/4336_202370523347335a068f307.jpg

And the cropped version...

http://www.ctomkinson.com/imgs/gallery/4336_182381838346238172d9766.jpg

Clearly if I'd used the 1DMK2n body I'd have recorded a smaller image of the bird in the first instance (at 1:1 reproduction) and would have had to crop even more to get the same framing.

I never worry about how much I crop, just the quality of the image I'm left with. Where cropping does become a huge problem is trying to hit the bar with 50mb interpolated image library images when it's often well nigh impossible to do so and show the bird at a decent size within the frame afer cropping, but there again the smaller sensor starts out with the better image.

What I really need is a 1DMK3 with a 1.6X crop sensor, But Canon aint making it. ||

 

[postcardcv]

What I really need is a 1DMK3 with a 1.3X crop sensor, But Canon aint making it. ||

perhaps I'm missing something here, but I thought the mk III did have a 1.3x 'crop factor'...?

 

[tonky]

perhaps I'm missing something here, but I thought the mk III did have a 1.3x 'crop factor'...?

Doh! I've amended my typo. :t:

 

[Photovisions]

Tonky

Your living proof that small pixels do help!
The Canon camera's with the smallest pixels are the 400D and 40D (see my table earlier in the thread), even smaller than the 20D and I reckon the 40D is nearly the perfect camera for bird photography and fairly priced.
Go get it

Adrian

I've totally lost track now of who's supporting which argument here and exactly what the argument is. |:S|

It took me a while to come to terms with the fact that my 20D is producing notably better small birds shots with more fine feather detail than the 1DMK2n, but it is. Both cameras produce exactly the same size image on the sensor, but it's the size of that small bird when you view the two images at 1:1 reproduction where the 20D really scores heavily, with a larger image of the bird and a significantly larger physical file size when both images are cropped to produce identical framing.

The point has been well made earlier that there's no substitute for getting closer to the bird regardless of which lens you use, and it's quite true, although that can bring it's own DOF problems, and it isn't always possible to get as close as you'd like. Some of my sharpest most detailed small bird shots have been taken with the Canon 180L Macro lens at a range of around 5 or 6 feet, but the fact is that even at that sort of range, regardless of which body I use, the images still require quite substantial cropping, so the 20D still wins and ends up with the larger file size after cropping.

The problem is largely peculiar to very small bird shots of course , (or very long range shots), and the differences become less marked with larger birds where you can fill the frame better.

So do I regret forking out the wonga for the 1DMK2n as soon as it came on the market? Of course not, it's built like a camera should be and makes the 20D feel like a toy in comparison. There are also occasions for flight and diving shots where the faster AF of the 1DMK2n is going to be the obvious choice and it will still AF with the 2X converter whereas the 20D wont.

It has changed my thinking though on my next body purchase. The results from a test of the 40D and the 1DMK3 would yield larger 1:1 images in each case, but the 40D would still win the crop comparison by a similar margin. The 1DSMK3 with 21 million pixels would be a great all round camera, superb for weddings portraits and everything else and still not be very far behind the 40D in cropping superiority, but it's £6K!

The 40D is looking better to me all the time! ||