Foucsing and DOF

[ahmad sabry]

Hello Dears

I am trying to do some practice. I am trying to photograph a toy car the issue I am trying to get the whole car in focus but I could not I attached the latest trial at which I got the car front in focus and sharp but I cannot for the car rear.
ISO 100
A 16
S 30
175 mm

Using CANON 5D MARK III, 70-200 mm F 2.8 II lens

Is this could be fixed via adjusting the technique or I need a macro lens (specially I want the background to be totally blurred)

[Dave Humphries]

I am trying to do some practice. I am trying to photograph a toy car the issue I am trying to get the whole car in focus but I could not I attached the latest trial at which I got the car front in focus and sharp but I cannot for the car rear.
ISO 100
A 16
S 30
175 mm

Using CANON 5D MARK III, 70-200 mm F 2.8 II lens

Is this could be fixed via adjusting the technique or I need a macro lens (specially I want the background to be totally blurred)

Hi ahmad,

Thanks for providing most of the useful information, the only bits missing are;
What was the distance from sensor to subject nearest and furthest points? and
What range of focal lengths can your lens achieve? and
What is its minimum focus distance? (although that may well change with focal length)

A macro lens on that camera won't really help - the amount of DoF you get will depend upon the aperture used, so f/16 on a Macro lens is no different to f/16 on the lens you used.

The concept of Hyperfocal Distance suggests you should focus about 40% 'in' from the nearest point you want sharp, to make the best use of DoF. If you focus on the nearest corner, you'll waste nearly half of it by having sharp 'fresh air' in front of the model. Looking at your shot, I'd say focus on the figure "2" or the front edge of the door crack.

However, punching the numbers into a DoF calculator returned a DoF of about 1mm, so the above may be academic.

If you have access to a smaller sensor camera, a P&S or bridge camera, even a mobile phone, for this type subject, their smaller sensor gives more DoF and will help you achieve a car that's sharp from one end to the other.

I'm sure others will pop up with better ideas too.

Welcome to the CiC forums from ....

[Manfred M]

You want the model to be sharp, yet the background blurred. This is a tough one as you are trying to balance out two different but interrelated issues. Dave has already given you some hints here. I disagree a bit regarding his comments on using a smaller sensor camera; you can achieve the same effect by magnifying your image less and cropping out the unwanted bits in PP. You have lots of material to do that with as you are shooting full frame. You might also want to look at setting up so that your background is further away to help throw it out of focus a bit more.

First of all, a macro lens is not going to help you per se; the laws of optics don't change and you are going to continue to fight the in-focus issues, as Dave has pointed out. One alternative might be to try focus stacking software, i.e. taking different sharply focused slices of the toy car and building them up into a sharp image, yet with an out of focus background. I haven't used this software (I'm not a macro shooter), but some of the members that are could certainly steer you in the right direction.

You could also try to achieve the background blur effect in PP. Shoot the car so that it is sharp and then create a mask to ensure that the car and foreground are not touched, but apply a gaussian blur to the background.

[Andrew1]

DAVE!!!! Focus on the figure "2" ????? Even being far a-field you should know that's the iconic Cobra from one of the most historic autos on the planet. For shame! Spank yourself.

Ahmad, just tried the same shot with your settings on a similar vehicle. Hard to get that little thing entirely in focus.

[Dave Humphries]

DAVE!!!! Focus on the figure "2" ????? Even being far a-field you should know that's the iconic Cobra from one of the most historic autos on the planet. For shame! Spank yourself.

LOL, OK, put my specs on now, I see what you mean

[Inkanyezi]

I think focus stacking is the best way to accomplish much bokeh for a small object like this if you want to throw the background so far out of focus that you cannot really see what it is. I tried to take a toy car with horizontal tilt, to see whether it would be possible to achieve sufficient DOF, but it does not cover the width of the front of the toy car at f/16 with the 50 mm lens. However the tail of the car is reasonably sharp at the same time as most of the side and the windshield.

My toy car is not as beautiful as the Cobra, and after a dog chewed it, it'll never be what it once was. And my mirror has its coating under the glass, but anyway, it gives an idea of how sharpness may be distributed by tilting and how much bokeh there is at a small aperture. If you instead do it by stacking, you can use a large aperture to throw the background really out of focus. Then you would need a special rail for stacking or if it's a Canon, it can be done by micro adjustment of focus with Magic Lantern.

[Inkanyezi]

For stacking, taking the shots requires that the camera is moved only along the optical axis, or that focusing is altered in micro-steps through the range one wishes sharp. There is a macro rail called StackShot that can accomplish the micro-movements, or with some Canon cameras, Magic Lantern can do micro-adjustment of the lens focusing. Then after you have taken all necessary images, they are merged with for example the Zerene Stacker software.

[ahmad sabry]

Hello dears

Sorry for being late to reply on your great post , I started the thread after about 1 hour and half of trials just before go to bed to be ready for working day and also due to the time difference.
Now I am back from work and read all your nice and helpful posts really I am amazed with this amount of great post
As a reply to my brother Dave
I set the camera at 1.2 m which the minimum distance allowed by my lens
Also I done some trials but I didn’t wrote about them at the time of writing the thread
I went back with the camera and tried to use the maximum allowed focal length which is 200 mm

Also it tried to use 110 mm and go for cropping after that as a trial to get more DOF


I got a bunch of information from Dave I will try to do research on them to understand them. Also I like the test done by Inkanyezi I will give it a try specially the StackShot idea
I am new to image stacking so I will try it . I saw some videos on youtube shows that it could be done via Photoshop
And I will come back with the results

[ahmad sabry]

"However, punching the numbers into a DoF calculator returned a DoF of about 1mm, so the above may be academic"

Dave kindly could you please explain much more and what should be the practical settings

[Inkanyezi]

The toy I shot is about 7 cm long, and the distance was about 30 cm with another 25 cm to the box in the background. The lens is a 50 mm f/1.8 Fujinon with a 10 mm extension ring, stopped down to f/16 and tilted 8° horizontally on a µ4/3 camera.

I think the image may be less compressed if you take it with stacking at a closer distance, for example with a 50 mm lens. Canon with Magic Lantern would be an evident choice, as it would not alter perspective in the way a rail does.

If you take the image with a small aperture, you can always blur the background smoothly with software. To cover the entire car you might need about ten images or less if you stop down a bit, and many more if you shoot wide open.

[DanK]

As one simple step, turn the car so that it is more nearly parallel to the sensor. That will lessen the amount of DOF you need.

I agree with Urban that stacking is probably your best bet, and also that Zerene Stacker is a good choice for this. However, you can do it simply. Urban wrote:

For stacking, taking the shots requires that the camera is moved only along the optical axis, or that focusing is altered in micro-steps through the range one wishes sharp. There is a macro rail called StackShot that can accomplish the micro-movements, or with some Canon cameras, Magic Lantern can do micro-adjustment of the lens focusing. Then after you have taken all necessary images, they are merged with for example the Zerene Stacker software.

For your purposes, all you need is the micro-steps in focusing. I have hundreds of stacked images and don't even own either a rail or Magic lantern. At this level of magnification, all you need is a tripod. Then start by focusing on the nearest part of the car, and take repeated images, moving the focus point back a tiny bit between each pair of shots. Zerene (and some other software) will take care of changes in perspective.

As an example, I'll post a recent image of a milkweed pod and seeds that I created that way. It was created by stacking 20 images in Zerene, using the DMap algorithm. The pod has a depth of about 2 cm.

[Inkanyezi]

Of course Dan is correct. You can do the micro-adjustments manually, if you keep your camera steady. My suggestion of the Magic Lantern is because it is a very versatile aid if you have a Canon camera, and a program can take care of the shooting, so that you may have a cup of tea while the camera takes its umpteen exposures. It can be done with most cameras, provided focusing in micro steps is reasonably reliable.

[Dave Humphries]

"However, punching the numbers into a DoF calculator returned a DoF of about 1mm, so the above may be academic"

Dave kindly could you please explain much more and what should be the practical settings

My point is, there aren't any

Dan and Urban have more experience with stacking than I, so rather than say something else stupid ("2" indeed ), I would recommend you try what they are suggesting.

I might try the P&S idea tomorrow to see whether it works, just gotta find a suitable subject.

[Inkanyezi]

I might try the P&S idea tomorrow to see whether it works, just gotta find a suitable subject.

It does work, what you don't get is bokeh. I often used a compact for product shots, mostly Canon PowerShot G7, where I can control the f-stop down to f/8. but I have also used Nikon Coolpix 3200 and 5600 for product shots, see example taken with Coolpix 5600 (the EXIF is still in the file). When stopping down these small lenses, the diffraction limit is passed, so they are usually sharper only stopped down minimally.

[ahmad sabry]

so as per your advice brothers i will try to do in depth research for stacking and do some setup to test and practice it.

the only issue i am afraid to try what is called "Magic Lantern" since my camera is new i just got it from three weeks and i am afraid to apply this firmware on it

any way i will go for stacking direction

[ahmad sabry]

[QUOTE=Of course Dan is correct. You can do the micro-adjustments manually, if you keep your camera steady. My suggestion of the Magic Lantern is because it is a very versatile aid if you have a Canon camera, and a program can take care of the shooting, so that you may have a cup of tea while the camera takes its umpteen exposures. It can be done with most cameras, provided focusing in micro steps is reasonably reliable. [/QUOTE]

Urban can you explain more since i am new to these terms

[Inkanyezi]

Urban can you explain more since i am new to these terms

When you have the camera on a steady tripod and take care not to budge it too much when you adjust focus, you can turn the focusing ring very little between different exposures, to have focus "slices" over the model car. This method is reliable with most cameras, and small differences in positioning on the image as when you happen to touch the camera so that it points marginally different, are not important, as the stacking software can cope with that. So the stacking software takes all those sharply focused slices and put them together so that they combined make a sharp image with depth. Anything outside the slices you take will have bokeh according to the f-stop you use. With a large f-stop, you get much bokeh, but you must make many slices. A smaller f-stop needs fewer slices, but will not provide as much bokeh.

The Magic Lantern software is not a firmware, but a firmware extension, a plugin that is run together with the original Canon firmware.

Magic Lantern accepts programs that can be run in the camera for various tasks. For example a program for stacking can be set to start at a specified focal point and do incremental adjustments for each image until it reaches another defined end point. All these images will be taken by the program while it runs, and it will stop at the point determined by the photographer. Taking let's say forty images with long exposure time may be a tedious task if you have to adjust focus by hand and press the shutter button for each one, and it is prone to mistakes, while running it as a program in the camera firmware will free the photographer to other things during the time it takes to shoot the object. So if Magic Lantern needs fifteen minutes to make the whole set of images, it will run by itself once set up.

[Bobobird]

Unfortunately Magic Lantern does not have a release for the 5D3 just yet, only a working beta with a limited set of functions.

[ahmad sabry]

many thanks Urban for your nice explanation and for you too Bobo for your Hint

Urban is this plug-in can harm the camera firmware by any mean? and shall it violate the warranty ofthe camera?

[Inkanyezi]

There is a full explanation about the firmware plugins on the wiki. So far I don't know of any case where a firmware extension has harmed a camera, and the actual firmware of the camera cannot be harmed, as this extension does not alter it at all. It is a program that is loaded, and that can utilise all functions of the camera firmware, also some that are hidden from the user in the usual menu system.

The caveats that I have had explained regard possible damage to the camera hardware, if software limits of movements are exceeded, for example focusing limits.

If you don't feel confident with the program, don't use it. You can do the stacking manually too, it is just more work.