Bokeh on crop vs ff

jamiebonline

jamiebonline

Messages
194
Name
Jamie
Edit My Images
No
Hi guys,

I guess most of you know Tony Northrup. I have watched the video where he explains you must multiply the aperture by the crop factor to match ff and crop images i.e. if you put a 50 1.8 on Nikon crop body you get 75mm but approx. 2.8 aperture. To test this you just need to look at what 75mm 2.8 looks like on ff. They should be identical. Now here is the part that makes little sense...

If you use a 50mm 1.8 lens you actually should get the same bokeh on ff and crop. isn't that right? i mean the frame is smaller on crop but the bokeh should be identical... or am I confused? If this is the case then it is kind of irrelevant when buying primes. I have my first ff camera and thinking of getting a 50 1.8 believing the bokeh would be better than the same lens on crop. This is wrong.

Thoughts? :)
 
You're pretty much correct there. A 50mm at f1.8 gives the same depth of field regardless of what camera it's mounted on, you're just cropping the edges. Cropping a full-frame image in post wouldn't affect the depth of field, and neither does it if you crop via a smaller sensor.

The reason for the equivalents that Northrup discusses is that often on smaller-sensored systems you want to compare lenses with equivalent angles of view. So for a fast normal on m43 you need the 25mm f1.4, which has the same depth of field as a 50mm at f2.8, because 25mm is a shorter focal length. Exposure values are consistent with f1.4 though.
 
jamiebonline said:
Hi guys,

I guess most of you know Tony Northrup. I have watched the video where he explains you must multiply the aperture by the crop factor to match ff and crop images i.e. if you put a 50 1.8 on Nikon crop body you get 75mm but approx. 2.8 aperture. To test this you just need to look at what 75mm 2.8 looks like on ff. They should be identical. Now here is the part that makes little sense...

If you use a 50mm 1.8 lens you actually should get the same bokeh on ff and crop. isn't that right? i mean the frame is smaller on crop but the bokeh should be identical... or am I confused? If this is the case then it is kind of irrelevant when buying primes. I have my first ff camera and thinking of getting a 50 1.8 believing the bokeh would be better than the same lens on crop. This is wrong.

Thoughts? :)
Click to expand...

Tony N knows his stuff, and if that's the video I'm thinking of, it's a good and clear explanation.

You have to compare like with like. Where I think you're going wrong is you're not adjusting either focal length or shooting distance to maintain the same framing.

martsharm said:
You're pretty much correct there. A 50mm at f1.8 gives the same depth of field regardless of what camera it's mounted on, you're just cropping the edges. Cropping a full-frame image in post wouldn't affect the depth of field, and neither does it if you crop via a smaller sensor.
Click to expand...

Not true. Check it out here http://www.dofmaster.com/dofjs.html

The reason for the equivalents that Northrup discusses is that often on smaller-sensored systems you want to compare lenses with equivalent angles of view. So for a fast normal on m43 you need the 25mm f1.4, which has the same depth of field as a 50mm at f2.8, because 25mm is a shorter focal length. Exposure values are consistent with f1.4 though.
Click to expand...

Yes :)
 
jamiebonline said:
Hi guys,

I guess most of you know Tony Northrup. I have watched the video where he explains you must multiply the aperture by the crop factor to match ff and crop images i.e. if you put a 50 1.8 on Nikon crop body you get 75mm but approx. 2.8 aperture. To test this you just need to look at what 75mm 2.8 looks like on ff. They should be identical. Now here is the part that makes little sense...

If you use a 50mm 1.8 lens you actually should get the same bokeh on ff and crop. isn't that right? i mean the frame is smaller on crop but the bokeh should be identical... or am I confused? If this is the case then it is kind of irrelevant when buying primes. I have my first ff camera and thinking of getting a 50 1.8 believing the bokeh would be better than the same lens on crop. This is wrong.

Thoughts? :)
Click to expand...

I've done a load of this stuff before giving up and just getting on with it :D

I'd agree that when looking at DoF the crop factor works so for example a 25mm f1.4 on Micro Four Thirds (MFT keeps the mental arithmetic easy as it's a x2 crop :D) will give the same DoF as a 50mm f2.8 on a full frame camera but as you're using different lenses the bokeh may well look quite different.

If you use the same lens on FF and on MFT the camera to subject distance comes into it.

If you shoot with each camera (with the same lens and at the same aperture) from the same camera to subject distance the final images will look quite different as the FF shot will have a wider field of view but if you then crop the FF picture to match MFT picture they should look pretty much the same (see my closing below) if you don't run out of resolution when cropping.

Of course you probably wouldn't shoot FF and MFT with the same lens from the same distance. You will probably move position as the field of view will be quite different. When using MFT you'll probably move back because of the change in field of view from 50 to 100mm and that changes things and you'll get deeper DoF as you've increased your camera to subject distance. Conversely, if you frame your shot with a 50mm lens on MFT when switching to FF you will probably move forward and that'll change things and you'll get shallower DoF because you've decreased your camera to subject distance.

Keep in mind that when you alter the camera to subject distance you also alter the perspective.

If you don't reduce the camera to subject distance when shooting with the FF camera the MFT picture will appear to have less depth of field but if you crop the FF image to the same field of view they'll look very similar and any differences will be down to... noise, resolution and other stuff affected by the chip and camera rather than just the lens, aperture and camera to subject distance and these things may well affect how the bokeh looks if you go looking.
 
anibap said:
This article explains it better. The dof of a 50 1.8 lens at 1.8 on a FF camera is shallower than the dof of the same lens at 1.8 on a crop body. Result - even if you crop the FF shot to the same frame as the crop image, the FF image will have shallower depth of field resulting in better OOF areas and bokeh

http://neilvn.com/tangents/full-frame-vs-crop-sensor-cameras-comparison-depth-of-field/
Click to expand...

I don't think that's a great example as he seems to be using different focal length lenses at the same distance and aperture, 85mm on FF at f2.8 v 58mm on APS-C (cropped in camera) at f2.8. It should be mildly obvious that the 85mm set up with give the shallower DoF here as when shooting from the same distance with these setups you'd have to use a wider aperture when using the 58mm lens in crop mode.

An easy way to look at this is how big is the aperture? At 85mm and f2.8 the aperture is 85/2.8 = 30.56mm but 58/2.8= 20.71mm.

If anyone wants some test shots I could shoot MFT and FF with a 50mm f1.8 and post the results or even 50 v 85mm or 25 v 50mm but it's a faff so keep me as a last resort :D
 
Last edited:
This doesn't sound right at all.

anibap said:
This article explains it better. The dof of a 50 1.8 lens at 1.8 on a FF camera is shallower than the dof of the same lens at 1.8 on a crop body.
Click to expand...

No. As stated there, with no reference to adjusting shooting distance to maintain framing, DoF on full-frame will actually be greater.

Result - even if you crop the FF shot to the same frame as the crop image, the FF image will have shallower depth of field resulting in better OOF areas and bokeh
Click to expand...

No again. If you crop the full-frame image, you've simply turned it into a crop format camera. DoF will be identical to the cropper.

To make valid comparisons, you really have to change the focal length. Then you can shoot from the same distance, so framing will be the same, and perspective will also be the same.

Example: 50mm lens on full-frame at f/4, will produce an identical image to a 31mm lens at f/2.5 on a Canon 1.6x crop format camera. Same framing at the same distance, with same perspective and same DoF. For equivalence, both focal length and f/number are adjusted by the crop factor. In DoF terms, the difference between full-frame and crop format is a smidge over one stop.
 
anibap said:
This article explains it better. The dof of a 50 1.8 lens at 1.8 on a FF camera is shallower than the dof of the same lens at 1.8 on a crop body. Result - even if you crop the FF shot to the same frame as the crop image, the FF image will have shallower depth of field resulting in better OOF areas and bokeh

http://neilvn.com/tangents/full-frame-vs-crop-sensor-cameras-comparison-depth-of-field/
Click to expand...

In that article he's using different focal lengths to achieve the same angle of view, not the same lens on different cameras.
 
There are lots of "use cases" which will give rise to different logical outcomes. My favourite way of thinking about it is this: if you took the classic picture of a ruler at 45 degrees to the lens, using the same lens on a variety of cameras, and printed them so the gradations on the ruler were the same size in each print, the DOF would be exactly the same in each.

The circle of confusion calculation that gives different results for DoF when calculating using DoFMaster does so because it assumes an uncropped print is being made - which means our ruler gradations would all be incomparable sizes anyway.

If you crop to the same size, or print at a magnification such that the object remains the same, it matters not what format camera you used, the same lens with the same aperture will give the same depth of field.
 
you have to include the distance to the subject
50mm on full frame 3m from subject will have the same DoF and Bokeh as a crop but the crop will not show as much of the image.
if you use the crop equivalent (say 35mm) so
35mm on crop 3m from subject the DoF and Bokeh will be different to the 50mm lens but the framing* will be the same as 50mm on a full frame.

*see https://www.lensrentals.com/blog/2016/05/how-lens-focal-lengths-will-effect-background-compression/ for what happens to the backround
 
I used to watch Tony and his wife, but only because of his wife :)

Tony spends far to much time analysing numbers and looking at graphs for my liking. Just get out there and enjoy photography.
 
If you're about to 'invest' (I use that word loosely lol) in either FX or DX then this may have a bearing on that decision, assuming cost differences weren't important, if you've already got either then do as I do and stop thinking about what the other formats may do differently and just enjoy shooting :)

I can't imagine anyone thinking... "My bokeh could be better so I'll sell all my DX gear and spend a few thousand £s on FX gear instead" :D

Be wary of being too techie and too little a photographer, I know a couple of people like this close to me; they can probably quote you the LPM of every lens since the mid 60s but can't shoot for sh-it

Dave
 
mountainpenguin said:
you have to include the distance to the subject
Click to expand...

Yes, to maintain same perspective.

50mm on full frame 3m from subject will have the same DoF and Bokeh as a crop but the crop will not show as much of the image.
Click to expand...

No and yes. The two images will not be comparable because the framing will be so different, but also the DoF will not be the same. FF image will actually have more DoF in that example.

if you use the crop equivalent (say 35mm) so
35mm on crop 3m from subject the DoF and Bokeh will be different to the 50mm lens but the framing* will be the same as 50mm on a full frame.
Click to expand...

Yes, a good and valid comparison - framing the same, perspective the same, FF image will have less DoF at same f/number.

*see https://www.lensrentals.com/blog/2016/05/how-lens-focal-lengths-will-effect-background-compression/ for what happens to the backround
Click to expand...

Shows perspective changes well :)
 
Last edited:
Depth-of-field in 30 seconds...

- DoF is an optical illusion, based on one basic fact. That is, the human eye cannot detect details smaller than 0.2mm in an image/print 10in wide, when viewed at a distance equal to the diagonal, ie 12in.
- All DoF calculations are worked back from that basic assumption.
- If the viewing distance is maintained as equal to the diagonal, DoF calcs hold good for different sizes of print, ie a huge street poster still looks sharp from over road. And equally, viewing at 100% on screen throws DoF calculations out of the window (unless you view from the far side of the room).
- DoF is all about magnification. Everything that changes the size of any part of the imaging chain affects the calculation, including: shooting distance, focal length, f/number (the diameter of the aperture), sensor size, print size and viewing distance.
- Good DoF calculator here http://www.dofmaster.com/dofjs.html Note how when you input the camera model, it changes the circle-of-confusion value on the bottom right according to sensor size.
 
DG Phototraining said:
If you're about to 'invest' (I use that word loosely lol) in either FX or DX then this may have a bearing on that decision, assuming cost differences weren't important, if you've already got either then do as I do and stop thinking about what the other formats may do differently and just enjoy shooting :)

I can't imagine anyone thinking... "My bokeh could be better so I'll sell all my DX gear and spend a few thousand £s on FX gear instead" :D

Be wary of being too techie and too little a photographer, I know a couple of people like this close to me; they can probably quote you the LPM of every lens since the mid 60s but can't shoot for sh-it

Dave
Click to expand...

Until quite recently, I would agree with you Dave, but I think things are changing. If we, as enthusiasts and professionals, are to differentiate our work from the billions of iPhone images and raise our game, then advanced equipment and techniques will play an increasing role. We have to look at the things smartphones cannot do.

Shallow DoF and bokeh effects give that 'professional look' sought by wedding couples, and it drives the appeal of full-frame and the resurgence of interest in fast prime lenses. Other things we can do that smartphones can't include a mixture of equipment and techniques. Such as: flash photography, studio work, very long lenses, very wide lenses, macro, very long exposure times, low light work, very short exposure times. These are the kinds of thing that will make high level equipment and advanced techniques worthwhile.

"Be wary of being too techie and too little a photographer..." Why? If you enjoy it? Photography is both an art and a science, involving specialist equipment, advanced techniques, and an infinite number of subjects. It's not 'all about the image' just as it's not all about the gear either, but photography can be enjoyed on any number of levels and aspects, and all are perfectly valid IMHO.
 
Ive not read all the posts but here's my understanding.

50mm f1.8 on each with the same framing DOF will be smaller on FF due to the closer subject distance.

50mm f1.8 on each from the same distance then the crop with have smaller DOF due to the greater effective focal length.

I would imagine if you stood in the same spot with a 50mm f1.8 on a 1.5x crop and a 75mm on FF (hypothetically speaking) then DOF would be the same?
 
No, when shooting from the same spot the crop factor works for focal length and dof so you'd have to use a wider aperture with the crop body and lens. Set everything to account for the crop factor and you'd end up with pretty near the same shot, framing and dof wise.
 
Last edited:
HoppyUK said:
Until quite recently, I would agree with you Dave, but I think things are changing. If we, as enthusiasts and professionals, are to differentiate our work from the billions of iPhone images and raise our game, then advanced equipment and techniques will play an increasing role. We have to look at the things smartphones cannot do.

Shallow DoF and bokeh effects give that 'professional look' sought by wedding couples, and it drives the appeal of full-frame and the resurgence of interest in fast prime lenses. Other things we can do that smartphones can't include a mixture of equipment and techniques. Such as: flash photography, studio work, very long lenses, very wide lenses, macro, very long exposure times, low light work, very short exposure times. These are the kinds of thing that will make high level equipment and advanced techniques worthwhile.

"Be wary of being too techie and too little a photographer..." Why? If you enjoy it? Photography is both an art and a science, involving specialist equipment, advanced techniques, and an infinite number of subjects. It's not 'all about the image' just as it's not all about the gear either, but photography can be enjoyed on any number of levels and aspects, and all are perfectly valid IMHO.
Click to expand...

Yes, but apart from those points :D

Dave
 
*If everything is kept the same except for FX/DX, then FX will have ~ 1stop greater DOF.
*If the subject framing is kept the same (shorter FL or greater distance w/ DX) the DX will have ~ 1 stop greater DOF. Because FL/distance have 2x the effect on DOF that aperture does.
*If everything is kept the same and the FX is cropped for the same subject composition, then the DOF will be the same. Because you have the same area remaining and it will require the same amount of enlargement.

DOF DOES NOT EXIST until the image is being viewed... and it is dependent on "how" the image is being viewed.

But I largely agree w/ Dave... use what you have. My only caution is that when considering a system keep in mind that those expensive fast primes for the smaller systems are *required* if you want to do the same things you can with slower/cheaper lenses on larger formats. And they do *not* allow you to do the same things as you can with an equally fast lens on larger format.
 
Last edited:
snerkler said:
Ive not read all the posts but here's my understanding.

50mm f1.8 on each with the same framing DOF will be smaller on FF due to the closer subject distance.
Click to expand...

Yes.

50mm f1.8 on each from the same distance then the crop with have smaller DOF due to the greater effective focal length.
Click to expand...

Yes.

I would imagine if you stood in the same spot with a 50mm f1.8 on a 1.5x crop and a 75mm on FF (hypothetically speaking) then DOF would be the same?
Click to expand...

No. Aperture must be adjusted. F/number x crop factor.
 
DG Phototraining said:
Yes, but apart from those points :D

Dave
Click to expand...

:)

Where's Pookeyhead when you want him? :D

I think the points I made above are actually very important - the stuff that smartphones just cannot do. It's what will ultimately keep Canon and Nikon etc in business (I hope) and us provided with advanced and affordable kit.

The comparison pictures here were a bit of a wake-up for me, iPhone 6 vs Nikon D800 :eek: http://www.petricek.cz/iphone6svsd800/ That bluddy iPhone is shockingly good, but only in the right situation.
 
Last edited:
  • Like
Reactions: mpe
There is a problem I have with these explanations "Supposing that all is same except for FX/DX...".

Photographers (FX, DX or any other format sensor size) tend to shoot similar things at similar shooting distances and try to achieve similar composition depending on what they want to achieve with the picture. Often they have to use a different lens (or zoom setting).

Considering that, gear with larger sensor makes it easier to produce pictures with greater subject isolation, shallower DOF or more pronounced bokeh features.
 
Last edited:
sk66 said:
DOF DOES NOT EXIST until the image is being viewed... and it is dependent on "how" the image is being viewed.
Click to expand...

I actually disagree with this view and if anyone is at all interested they can read on and I'll explain why...

Once you've picked up your camera and lens, dialled in the settings, framed the shot, pressed the shutter button and taken the shot that's it and you're stuck with it.

Whatever DoF you then perceive at whatever output size and viewing distance you choose is set by the camera, lens, settings and camera to subject distance and you're stuck with it.

If you wanted deeper dof, tough. If you wanted shallower dof, tough. It's too late. What you see including the dof at whatever output size and viewing distance you choose is set when you take the picture.

The shooter should think about a few things before even picking up the gear and rushing out to take the shot. You should think about what you are going to shoot and how you are going to get the shot, you should think about the final image, how you are going to display it and how big and how you are going to view it and of course you should think about how you want it to look including the subject, perspective, depth of field and anything else you want to throw into the mix.

If you don't think about these things and use them as a starting point to decide what gear and settings you choose you may well be stuck with a final image that isn't what you wanted. Seems obvious to me :D

So I'd say that the dof does exist before you display and view the image and that it's just waiting to be revealed and perceived and that it's set when you press the shutter button and then you're stuck with it.

To me saying "DOF DOES NOT EXIST until the image is being viewed... and it is dependent on "how" the image is being viewed" seems like saying that the dof is something you can decide on later by deciding how big you are going to print or display and at what distance you are going to view. Viewed in these terms this is IMO madness.

There's little point deciding that the 36" high portrait print you'd planned will now be printed the size of a postage stamp because at 36" one eye is sharp and the other is out of the dof and the only way to hide that and get both eyes sharp is to print it the size of a stamp... or tell people to look at it from a bus ride away.

What you should have done is thought about the image you wanted to capture and how you were going to display and view it and used this information to help you to choose your gear and settings and how you took the shot.

To me no one should start with the gear. You should start with the end result and work out how you are going to get it and viewed in these terms dof exists right through the process and is set when you press the shutter button.

Maybe I'm mad or maybe this goes back a long way for me. Years ago I made artwork to be displayed on stage and it needed to be visible by the audience sat at the back and they had to be able to see (the whatever.) That was the starting point and I still think like that.

If I want a small picture with shallow dof I should know that I'm going to have to have some very shallow dof so it'll look right in the final small image and that's going to influence what gear I choose and the settings. Conversely if I want a big picture with apparently front to back sharpness I'm going to have to use appropriate gear and settings so that the final big picture looks right. So for me dof exists when I press the button and will be seen in the final image, and once I've pressed the button I'm stuck with it.
 
Last edited:
DG Phototraining said:
Yes, but apart from those points :D

Dave
Click to expand...


Well, without drawing too far away from the subject, there is one other major factor; Field of View (or equivalent Focal Length).

I use FF, APS-H and APS-C cameras. In a fixed position I need to know what the effect of mounting each different sensor type onto a lens will have.
Stick the wrong sensor onto the wrong lens in any particular situation and I could well end up either seriously over or under lensed.
 
woof woof said:
To me saying "DOF DOES NOT EXIST until the image is being viewed... and it is dependent on "how" the image is being viewed" seems like saying that the dof is something you can decide on later by deciding how big you are going to print or display and at what distance you are going to view.
Click to expand...

Think of all of the things that are outside of your control... you cannot typically really control how far an image is going to be viewed from. I.e. your stage example... it will be different at the back and front of the stadium.
And on the web you cannot control how large an image is going to be viewed (monitor size/settings). When I view someone's image on the web it will very likely be a different size (but similar distance) than someone using a larger monitor/lower resolution setting. And what if I select to view the image full size instead of downsized? All of those things will change the apparent DOF in the image.

But, you are correct in that the other factors affecting DOF are fixed at the time of capture. If you do have a specific size/distance requirement then DOF needs to be considered more carefully. And maybe the default DOF/sharpness calculations are wrong for your use.
 
Last edited:
Most of my images are viewed by people in the same room as the picture or the output device, so yes, I have some control.

When taking a picture I often have the final size and viewing in mind and I find it ever so slightly amazing that the output size and viewing distance should come as a surprise but then I don't take pictures and throw them out of a window for someone else to find and view as they see fit. Typically I think "I'd like to take a portrait and mount it on that wall..." or "I want to take some pictures whist on holiday in Thailand and email them to people."

These decisions decide what kit and setting I use, not the other way round."
 
Last edited:
You must log in or register to reply here.
Back
Top