Crop Factor; Is it as big an issue as it seems?

[Lornek]

I just bought myself my first camera, a Rebel T2i. I knew a decent amount about the Canon lineup before making my decision, but never really paid much mind to the 1.6x crop factor that most of the cameras have.

After doing some calculations:

A 50mm 1.4f lens on a 1.6x body gives you equivalent results to an 80mm 3.6f lens on a 1.0x body. Not in terms of amount of light captured, but in terms of your DoF results.

You basically diminish the beautiful DoF effects that you pay so highly for in expensive lenses by not having a 1.0x body it seems.

So is it a big mistake buying up these 1.6x cameras? Seems you lose a lot of shallow DoF potential by doing so. With the amount of megapixels on cameras now, you don't really lose out on much if you decide to 1.6x crop pics from a 1.0x body anyway.

[Clactonian]

It is surely only the angle of view that is affected. By careful selection of the aperture you can still control the depth of focus. It will probably take a while to get your head around the fact that your 'standard' lens is now effectively a short telephoto lens, but it soon becomes second nature.

[Lornek]

Yeah my standard lens is a short tele lens, but now only has the potential DoF shallowness of a tele lens with the aperture size multiplied by [Crop Factor ^ 2].

So my 50mm 1.4f is now an 80mm mini-tele, but it only generates as shallow DoF as a 3.6f 80mm on a full frame body.

I can't help but feel a bit bad about my camera choice now. I love playing with DoF and using really big apertures for shallow focus effects.

[Colin Southern]

Hi Lornek,

It's really a case of "horses for courses".

If you and I are out shooting wild bears, each with a 600mm lens - me with a FF camera, you with a CF camera, then I'll have to get a lot closer to them to get the same FoV ... and I don't like getting close to wild animals. So in my opinion, in that situation the CF camera wins. If you're shooting macro eg insects or even the likes of dental photography then the CF camera wins again. However, if you want shallow DoF the the CF camera loses out to it's FF counterpart - however - perhaps the question isn't so much "which is better" as it is "is the shallow DoF I'm getting from my CF camera sufficient"? Is a Porche 911 twin turbo fast enough to go shopping, or do I need a (400km/hr +) Bugatti Veyron?

With regards to cropping a FF down to a CF FoV ... yes / no / maybe ... the (obscure) point I'm trying to make is that you have to take the relative pixel counts into account, you can't just use the crop-factor. Case in point ... a collegue shoots with a 10MP Canon 1D3 (1.3 CF), and I shoot with a 21MP Canon 1Ds3 (FF). He claimed a 1.3x reach advantage when shooting speed boats one weekend ... but in reality, If I crop my FF shot down to the same FoV (assuming same physical lens focal length here) I actually out-resolve him by around 30%. So in terms of resolving power (assuming a "perfect" lens), you'll get a LOT more detail from, say, an 18MP 7D than you will from a cropped 21MP 1Ds3 or 5D2.

If you're losing sleep, then just go and buy a FF camera - you know you want to

[revi]

Hi Lornek,

It's really a case of "horses for courses".

If you and I are out shooting wild bears, each with a 600mm lens - me with a FF camera, you with a CF camera, then I'll have to get a lot closer to them to get the same FoV ... and I don't like getting close to wild animals.

(...)

If you're losing sleep, then just go and buy a FF camera - you know you want to

And drag all that extra glass around? Sure?

Remco

(To go back to the wildlife: FF with 600 mm can do the same as a CF with 400mm, which is (a lot) less than half the weight..., ignoring pixel counts etc. for the moment)

[Peter Anson]

How come the DoF is altered? I don't understand this at all.
I suspect it can be explained mathematically.

[Eric M]

FF does produce less DOF. This is either a blessing or a curse depending on what you are trying to achieve. Despite the pure IQ benefits of a FF sensor, crop cameras have several advantages for many uses.

Keep in mind that bokeh effects also have to do with the distance to the subject, distance of the subject from the background, focal length of the lens, and aperture.

[Lornek]

How come the DoF is altered? I don't understand this at all.
I suspect it can be explained mathematically.

The reason is that a FF camera needs to be placed 1.6x closer to the subject than the CF camera for it to occupy the same amount of the frame. The closer you are with the same lens, the shallower the DoF you can capture.

The get the data on what the lens equivalent on a 1.6x body would be:

(Focal Length) x (Crop Factor) to get the effective mm length of the lens.

(Aperture Size) x (Crop Factor^2) to get the effective DoF size.

So a 50mm lens with 1.4f on a 1.6x CF body would look the same as a 80mm 3.6f lens on a FF.

Honestly with the sheer amount of MP on DSLRs now, I don't mind losing 40% of 21m pixels for the vast majority of work I do. Plus then I have the raw data of a FF camera too.

Oh well it's my first camera. I'll deal with the 1.6x business and next time get FF if I still want it.

[Wirefox]

If you're losing sleep, then just go and buy a FF camera - you know you want to

Never a truer word. All I will add is that if your style of shooting means that minor differences in DoF and focus blur are critical it is odd that you did not go for a full frame in the first place. If your style has developed over time and you now predominantly use extreme shallow DoF as a trademark feature of your work then its time to upgrade. If on the other hand this concern is an academic insecurity brought on by forum overdose I would say get out and shoot with the kit you have and then judge the resultant images rather than the technical specifications of the equipment.

It sounds harsh I know but its a place we have all been to greater or lesser degrees. I still read the odd article that makes me question my choice of gear. But I head straight for the freezer get a large bag of frozen pea's and get WireVixen to slap me around the mush with them. Cures the insecurities instantly and does wonders for creativity.

Steve

[RichB]

Sounds like a good excuse to go and buy a wide aperture expensive lense to me.

[rpcrowe]

Eric wrote:

"FF does produce less DOF. This is either a blessing or a curse depending on what you are trying to achieve. Despite the pure IQ benefits of a FF sensor, crop cameras have several advantages for many uses.

Keep in mind that bokeh effects also have to do with the distance to the subject, distance of the subject from the background, focal length of the lens, and aperture."


Full frame cameras DO NOT produce less depth of field than 1.6x crop cameras. In fact, if used with the same focal length lens and shot from the same distance, they produce a greater depth of field. Considering that the images in all the calculations below were shot at f/2.8:

1.6x crop camera with 50mm lens focused at 10 feet = 1.29 feet DOF
Full frame with 50mm focused at 10 feet = 2.06 feet DOF

What gives you the more shallow DOF when shooting with full frame equipment is that in a lot of photography, portraiture for example, the photographer using full frame equipment will either choose a longer focal length lens or will shoot closer to the subject in order to achieve the same framing as with a 1.6x camera.

As an example, in order to achieve the same framing as a 50mm lens provides on a 1.6x camera, the photographer must use an 80mm lens.

Full frame with 80mm focused at 10 feet = .79 feet (a shorter DOF than the 1.29 feet above)

Using a 50mm lens on a full frame camera and shooting with a decreased lens to subject distance would also decrease the DOF.

However, if I and another photographer were standing on the sidelines of a football game and we were each shooting with a 400mm lens at the same f/stop and the same player... The photographer using the 1.6x equipment would have a more shallow DOF than the photographer shooting with full frame equipment.

[Colin Southern]

Honestly with the sheer amount of MP on DSLRs now, I don't mind losing 40% of 21m pixels for the vast majority of work I do.

I'm afraid that you lose a lot more than 40%. At the same pixel density (which by definition it must be), you're essentially reduced to a 8MP camera.

5616 / 36 * 22.3 = 3479 (convert horizontal 5D2 pixel dimension to 550D pixel dimension)

3744 / 24 * 14.9 = 2324 (convert vertical 5D2 pixel dimension to 550D pixel dimension)

...

3479 * 2324 = 8,085,196 (calculate new MP count)

...

((21-8)/21) * 100 = (approx) 62% (convert to % loss)

[Colin Southern]

Sounds like a good excuse to go and buy a wide aperture expensive lense to me.

It's OK for you, you're Rich (sorry, couldn't resist!)

[Colin Southern]

If on the other hand this concern is an academic insecurity brought on by forum overdose I would say get out and shoot with the kit you have and then judge the resultant images rather than the technical specifications of the equipment.

Never a truer word spoken there too!

I think a whole photographic generation has spawned that is totally hung up on specifications, and yet has absolutely no concept as to how these specifications relate (or as is more often the case don't relate) to a visible difference in a real-world photo. In many cases folks seem to worry about how pixels look at 100 (or 200%) magnification, whereas in reality to see the same thing in a real world print they'd have to print it 6 feet wide x 4 feet high ... and then inspect it with a magnifying glass!

[Peter Anson]

The reason is that a FF camera needs to be placed 1.6x closer to the subject than the CF camera for it to occupy the same amount of the frame. The closer you are with the same lens, the shallower the DoF you can capture.

The get the data on what the lens equivalent on a 1.6x body would be:

(Focal Length) x (Crop Factor) to get the effective mm length of the lens.

(Aperture Size) x (Crop Factor^2) to get the effective DoF size.

So a 50mm lens with 1.4f on a 1.6x CF body would look the same as a 80mm 3.6f lens on a FF.

Honestly with the sheer amount of MP on DSLRs now, I don't mind losing 40% of 21m pixels for the vast majority of work I do. Plus then I have the raw data of a FF camera too.

Oh well it's my first camera. I'll deal with the 1.6x business and next time get FF if I still want it.

Thanks for for this.

[Peter Anson]

Thanks to rpcrowe for the detailed explanation.

[RichB]

It's OK for you, you're Rich (sorry, couldn't resist!)

Wish that it were true.

[PhotoWeb]

Never a truer word spoken there too!

I think a whole photographic generation has spawned that is totally hung up on specifications, and yet has absolutely no concept as to how these specifications relate (or as is more often the case don't relate) to a visible difference in a real-world photo. In many cases folks seem to worry about how pixels look at 100 (or 200%) magnification, whereas in reality to see the same thing in a real world print they'd have to print it 6 feet wide x 4 feet high ... and then inspect it with a magnifying glass!

Outstanding point! And good for the ulcer.

[carregwen]

I have three cameras with three different crop factors. A 5D (FF), a 50D (x1.6 CF) and a Panasonic G1 (x2 CF). TBH they are all really good in certain situations. The 50D outstrips the 5D for macro and plant photography. The 5D slightly outstrips the 50D on landscape.... but.. a pano with the 50D outstrips (easily) the 5D! The G1 is just great at most things, and weighs about a third of the other two.

[PopsPhotos]

It does cut both ways. I can go out with my D40 or D70 with the 200mm or 210mm lens and leave the Soligar 400mm at home, dropping about 5 pounds in lens and 6 pounds in tripod in the process. I can put the 26mm E series on either of them and not have a wide enough field of view for the church.

If I'm in a quarterhorse race I don't want a Percheron or Clydesdale. If I'm hauling a corn combine, I don't want burros. Horses for courses, as was so aptly said above.

Pops