Dept of Field Data

I have raised this issue on this forum on several occasions but I'm still not ready to give up on it.

More often than I care to admit I have an out of focus subject in a group because I missed my depth of field guess.

Yes I have tables and iPhone apps that I can use to calculate depth of field to great accuracy,

but in the heat of battle changing f stops and ISO and focal lengths (not to mention white balance) I get an out of focus subject in a group

only because I let the depth of field slip. Or I move three feet closer and the whole depth shrinks. Then in post processing I realize ooops.

Ok what I want is a depth of field display in my view finder to remind me, it could even blink if it wants to so I pay attention.

All the necessary data is in the camera. Look on photoshop bridge metadata. Its right there focal length 70, lens 24-70 aperture f2.8 camera 5d MARK III,

and subject distance 2.7 meters.

What is not there is the depth of field which in this case is 24cm, why could it not display or flash this data.

Or a nice SIRI voice that says "are you really sure you can fit all your subjects in to this 24 cm space".

Does anyone else have DOF issues?

Thank you.

Naive question - have you tried using the DoF Preview button?

Yes I have used the DOF preview button but it is not the same. If I had the data I would have a better handle on how to fix the problem.
The data would be much more valuable.

I always found the DOF preview not that good.

I am sure the data could be in camera, it probably hasn't been in the past because of the pixel and processor power race. Dedicating processing to IQ and FPS is more important. Maybe we will see a DOF calculator in the future, but why stop there? Display subject distance, FOV and other info as well.

That sounds nice to have. I often forget to check half the crap it throws at me though.
I'd rather have zebra or focus peaking in the viewfinder. Could be what they are trying in the next M model.

Quote:

---

Originally Posted by iND

I have raised this issue on this forum on several occasions but I'm still not ready to give up on it.

More often than I care to admit I have an out of focus subject in a group because I missed my depth of field guess.

Yes I have tables and iPhone apps that I can use to calculate depth of field to great accuracy,

but in the heat of battle changing f stops and ISO and focal lengths (not to mention white balance) I get an out of focus subject in a group

only because I let the depth of field slip. Or I move three feet closer and the whole depth shrinks. Then in post processing I realize ooops.

Ok what I want is a depth of field display in my view finder to remind me, it could even blink if it wants to so I pay attention.

All the necessary data is in the camera. Look on photoshop bridge metadata. Its right there focal length 70, lens 24-70 aperture f2.8 camera 5d MARK III,

and subject distance 2.7 meters.

What is not there is the depth of field which in this case is 24cm, why could it not display or flash this data.

Or a nice SIRI voice that says "are you really sure you can fit all your subjects in to this 24 cm space".

Does anyone else have DOF issues?

Thank you.

---

I’m afraid an in-camera DOF calculator isn’t that easily implemented. There are a few important parameters unknown to the camera: The size of the final image/print, the viewing distance and the eyesight of the viewer. And if you don’t know these parameters you can’t calculate the DOF.

Of course you could guess for example that the long side of the final print will be 25 cm, and that the viewer is able to distinguish 5 line pairs per mm at a viewing distance of 25 cm. The resolution 5 lp/mm corresponds to a CoC (circle of confusion) of 1/5=0.2 mm on the final image. Since the sensor image must be enlarged by a factor 25/3.6 = 6.94 (for FF sensor) to get the final size, it corresponds to a CoC of 0.2/6.94 = 0.029 mm on the sensor. This value is commonly used by “DOF calculators”, but is based on the above guesses.

I guess the camera manufacturers are not willing to add a function that would attempt to calculate an exact number of something that’s actually quite ambiguous and subjective. I suppose they let us either calculate for ourselves or simply check if the result is sharp enough.

Quote:

---

Originally Posted by cls

calculate an exact number of something that’s actually quite ambiguous and subjective.

---

That's something that they are quite good at;)

Quote:

---

Originally Posted by cls

I’m afraid an in-camera DOF calculator isn’t that easily implemented. There are a few important parameters unknown to the camera: The size of the final image/print, the viewing distance and the eyesight of the viewer. And if you don’t know these parameters you can’t calculate the DOF.

---

Print size wouldn’t matter as far as the calculator goes. The camera does know focus distance, it displays this in the exif files with some programs. My near sightedness doesn't affect the DOF of my pictures. I think it is easily implemented, and should be as accurate as the ones that are on-line. On-line calculators only need to know a few things; lens, aperture, distance and sensor size. All of the pieces of information are available in camera.

Quote:

---

Originally Posted by andnowimbroke

That's something that they are quite good at;)

---

Very true, white balance and exposure are two of the things that the cameras sometimes calculate differently from what I think. I should have been more carful with that conclusion ;).

Quote:

---

Originally Posted by HDNitehawk

Print size wouldn’t matter as far as the calculator goes. The camera does know focus distance, it displays this in the exif files with some programs. My near sightedness doesn't affect the DOF of my pictures. I think it is easily implemented, and should be as accurate as the ones that are on-line. On-line calculators only need to know a few things; lens, aperture, distance and sensor size. All of the pieces of information are available in camera.

---

Yes, but when they need to know sensor size it's because it is put in to the "magnification factor" calculation (from sensor size to the assumed print size of 25 cm / 10 inch long side). The assumption of the viewer's ability to distinguish 5 lp/mm at a viewing distance of 25 cm is also built in (typically these parameters are not inputs to DOF calculators). But sure, the same simple algorithm as can be found in various DOF calculators could be implemented also in cameras. But the result would still be based on assumptions.

Quote:

---

Originally Posted by cls

Yes, but when they need to know sensor size it's because it is put in to the "magnification factor" calculation (from sensor size to the assumed print size of 25 cm / 10 inch long side). The assumption of the viewer's ability to distinguish 5 lp/mm at a viewing distance of 25 cm is also built in (typically these parameters are not inputs to DOF calculators). But sure, the same simple algorithm as can be found in various DOF calculators could be implemented also in cameras. But the result would still be based on assumptions.

---

That would be true, since the DOF is subjective. Any calculator you have will be the same.

Quote:

---

Originally Posted by HDNitehawk

The camera does know focus distance, it displays this in the exif files with some programs.

---

I wouldn't put it quite so strongly. The camera records a value for focus distance in the EXIF, that I won't dispute. But sometimes that value is egregiously inaccurate, making any reported DoF estimage what the engineering folks call a WAG (wild-ass guess).

On all of the pictures you take do you measure the distance to the subject, or do you wag the distance?
Are our wag's better than the cameras? Making a good living off wag's from my experience I can say the cameras wag is not all that bad.

After all, DOF is subjective. Depending on your pov even the on- line calculator might be nothing more than a wag.

I suppose things like these have been discussed within the manufacturers' development units. My guess is that they hesitate to introduce functions that aren't really necessary and that will inevitably cause discussions and complaints. "The in-camera DOF meter said a DOF of 59 cm. Then why isn't the entire group sharp on my photo when the distance between front and back row was only 41 cm? I've studied the photo in 100% magnification and can clearly see that the back row is less sharp - this camera sucks!" But for those who can handle such a function and take estimations for what they are, I agree it would be helpful.

You could easily think of a customizable function where you could set the camera to show different parameters in the viewfinder, e.g. the set focusing distance, near/far DOF limits, DOF, hyperfocal distance etc. In this customization they could also allow the user to choose between the default 25 cm measures or set their own parameters for that too.

I'm not sure about the focusing distance measurement, maybe it's too rough. If so, you could imagine the complaints: "My camera focuses at 1.2 m when I carefully measured the actual distance to be 1.284 m. Please make a warranty repair of my eqipment." But even if the distance estimation isn't usable, it could still be helpful if the camera could assist in estimation of hyperfocal distance. (Still this would require that the users can accept the fact that the presented numbers are only estimations for typical use of the image.)

In some ways, Canon answered this 'problem' already. Rebel/xxxD bodies prior to the T4i/650D have an A-DEP setting on the mode dial, which would have the camera pick an aperture to get the nearest and furthest subjects in focus (limited by what was under the AF points). Older high-end bodies had a more sophisticated version, DEP mode - found on the EOS 1 and 3 film bodies (maybe the EOS 7?), and the original 1D and 1Ds. Similar to multi point metering, you'd focus on a near subject then a far subject with an AF point (two separate steps) then the camera would AF to the hyperfocal distance and pick the widest aperture that would get far and near subjects in focus.

But they seem to think we've outgrown the need for these 'crutches' and the DEP-type modes are gone now...

That mode doesn't sound to accurate. Knowing how inaccurate the AF system is on some of the older bodies I bet the results were very poor or random. You would only have to miss one point with the AF and the whole series would be bad.

Still I think any of the extras we can think of, until processors reach a break away point from the sensors and IQ we are not going to see to many extra frill's. It is similar now to computers in the 90's and the expensive cycle of more memory and faster computers, then programs use the memory and speed up as fast as it advances. Now any new processor speed is going to things like noise reduction and increasing pixel density. I would think developers would look at DOF calculators and their first though is that everyone has a smart phone now, use it.

Thanks for all the great discussion. This is why I keep coming back to this forum.

I would settle for some inaccuracies in the DOF calculation it would not have to be exact.
Like I said maybe just a idiot light with an adjustable setting:
Something like a blinky if the DOF of field falls below a preset value.
Example: If my DOF panic light (even if not accurate to the cm) would blink if say the DOF fell below 1 meter, then I could back up and reframe, and not suffer the to post processing disappointment.

I would even settle for the values in the app based softwares that use the known data

Here is a good example where I started a series of shots then moved in to fill the frame, forgetting that my DOF fell to 27cm at one meter.
Result is (since I used a tight group of focus points (Mark III) the second girls is unacceptable. Not anyones fault except my own as I failed to think
SHRINKING DEPTH OF FIELD as I moved in. Wish I would have had a quick reminder. I could have simply stepped back, filled the frame with some unwanted detail but that could have been an easy crop in post processing (starting with 22mpx and is no big deal)




Attachment 1680

iND, so what you’re asking for is an idiot light in the camera to tell you the DOF is incorrect.:p

Taking wildlife pictures if it has happened once it happened hundreds of times, in the heat of the moment I see something and grab the camera and start taking pictures. I am usually in manual mode and after about four or five frames I realize I am underexposing by about five stops. I need an idiot light that says "hey stop and think".

I made charts a few years back, using one of the online calculators and adapting the chart.

Honestly I used them very little, and I seldom use a calculator.
But in making the chart it gave me a better grasp of what to expect, and helped make the thought process second nature when taking pictures.
It is the thought process that screws up many of my pictures. I still have the charts in my bag in case I want to check.

Canon 5D (Mark II) Focal Length: 100 mm
	f/2.8	f/2.8	f/4	f/4	f/5.6	f/5.6	f/8	f/8	f/11	f/11	f/16	f/16	f/22	f/22	f/32	f/32
Distance
(feet)	Total	DOF In Front	Total	DOF In Front	Total	DOF In Front	Total	DOF In Front	Total	DOF In Front	Total	DOF In Front	Total	DOF In Front	Total	DOF In Front
1	0		0		0		0		0.02	50%	0.02	50%	0.02	50%	0.04	50%
2	0.02	50%	0.02	50%	0.04	50%	0.04	50%	0.07	43%	0.1	50%	0.14	50%	0.19	47%
3	0.04	50%	0.06	50%	0.08	50%	0.12	50%	0.17	47%	0.23	48%	0.34	47%	0.47	47%
4	0.08	50%	0.1	50%	0.15	47%	0.21	48%	0.31	48%	0.43	47%	0.61	46%	0.87	45%
5	0.12	50%	0.17	47%	0.24	50%	0.35	49%	0.48	48%	0.69	46%	0.98	45%	1.39	43%
6	0.18	50%	0.25	48%	0.35	49%	0.5	48%	0.7	47%	1	46%	1.43	44%	2.04	42%
7	0.24	50%	0.35	49%	0.48	48%	0.69	48%	0.97	46%	1.38	45%	1.97	43%	2.84	40%
8	0.32	50%	0.45	49%	0.64	48%	0.91	47%	1.28	46%	1.82	45%	2.61	42%	3.77	39%
9	0.41	49%	0.57	49%	0.81	48%	1.15	47%	1.63	45%	2.31	44%	3.37	41%	4.92	37%
10	0.54	44%	0.74	46%	0.98	49%	1.46	45%	2.01	45%	2.84	44%	4.17	40%	6.11	36%