Help...I've tried everything for sharp images!

[jerryph]

Oh heck, nothing to apologize for!

[Kit]

[QUOTE=patterfr;1497779]Wanna see where the plane of focus is?
Look at the grass. It will always tell and the kid is in the plane of focus and the tree and adults are not.
It's that simple.

QUOTE]
Excellent tip! Thank you, now I'll always look for that.

also.. this thread has some of the most helpful info on focusing that I've seen on any forum. You guys are very generous with your time.

[blumesan]

Originally Posted by jerryph

Not so much "wide" as simply "wider", and the reason is for increasing the DOF. A lens of 50mm at F/5.6 has a greater DOF than a 200mm at F/5.6.

Hi Jerry,
You might want to plug this scenario into the DOF calculator. At a distance appropriate for say a family group of 3 (magnification 0.02), the 50mm lens will give you just 24mm (about 1 inch) more DOF than the 200mm.

[Gene51]

Your lens is extremely sharp in the center, and very sharp in the edges, achieving maximum sharpness across the field at F2.8, so I would say that F3.5 is not going to be an issue as far as sharpness is concerned, regardless of how much the lens cost.

The issue seems to be more seated in post processing. Certainly, the depth of field could be greater, to include the child and the parents in the zone of focus. Even if you were to use ISO 200, you should be able to still use F4 - F4.5, and focus on the tree then recomposing. But your post processing will make that image really pop. if you are using Photoshop CS5, you can easy do three levels of sharpening. First you will need to do capture sharpening - its a given when recording raw images. I usually do a modest amount of sharpening in camera raw. just enough to see some detail, but not enough to add grain. You could add a little "light" by using an adjustment brush to liven up the faces and another adjustment brush to soften and darken the background, as a means of increasing the separation between subject and background. Finally, when you open the image in Photshop, you can do a bit of local contrast enhancement - using unsharp mask, set the percentage to around 20, then the radius up to 40 or more, and watch what it does to the image. After you get that lookin ok, you can do the final sharpening with a small radius and a reasonable percentage. Last but not least, in CS5 you can execute the Edit-Fade Unsharp Mask as the command immediately following sharpening - so that you can minimize the sharpening on surfaces and apply it to just the edges. If you were to send me a copy of the raw file I could give it a shot to illustrate what I am talking about.

[jerryph]

Originally Posted by blumesan

Hi Jerry,
You might want to plug this scenario into the DOF calculator. At a distance appropriate for say a family group of 3 (magnification 0.02), the 50mm lens will give you just 24mm (about 1 inch) more DOF than the 200mm.

Then I would have to question it's accuracy.

Calculaters are useful... within the confines of the mathematics, but if the difference between a 50mm has only an inch more DOF than a 200mm, why did my real life results make such an obvious bigger difference with less changes in focal lengths than what you mention?

How did I make an F/2.8 ;ems have a shallower DOF than an F/1.4 lens? In fact the visual equivalent of a DOF of around F/1.0? (in reference to the blog article I wrote)

I have nothing against the DOF calculator, but there are times to leave the calculator at home... and look at the results and let them speak for themselves.





Both photos were taken at F/4 and the same shutter and ISO numbers and same camera to subject distances.One was at around 70mm, the other at 200mm.

Why is the background so much less blurred in the first photo vs the second? In the top, the fronds are nearly individually visible, in the 2nd, they are a mass of single colour variations, no individual stocks are visible. Literally all that was changed was focal length from 70mm to 200mm (and perhaps some minor lateral movement), nothing else?

Rather than make reference to some calculator on the internet, why not do the same tests I did... in real life? This is not all that hard to do, as I just showed.

At 200mm if I placed you 4 feet behind my subjects, you would begin to be blurred to the point of it being very easy and obvious to see..

At around 70m, you would be clear as day... reference the clear LACK of any blur in the steps in the snow a good 10 or more feet behind the subjects in the first photo. Look at the clarity of the individual branches in both photos too. At which focal lengths are the individual branches easier to see?

Why can I get EVERYTHING in a room in focus at F/2.8 with a 15mm fisheye lens and that same F/2.8 at 200mm blurs the crap out of anything that is more than 6 inches behind whatever I focus on? I mean... F/2.8 is F/2.8, no?

I guess not... maybe it makes a bigger difference than most are aware of, calculator or not.

[blumesan]

Originally Posted by jerryph

Does that take into consideration all 5 factors or just focal length? Does it take into consideration sensor size? How about relative positions of the 3 people?

First of all I am speaking from practical experience, which I took the trouble to confirm with a reliable DOF calculator. My example considers focal length, aperture, and image magnification (such that the object width at the plane of focus is the same, i.e. the same framing of the subject). It assumes the same size sensor (FX) in both cases.

Calculaters are useful... within the confines of the mathematics, but if the difference between a 50mm has only an inch more DOF than a 200mm, why did my real life results make such an obvious bigger difference with less changes in focal lengths than what you mention?

Why is the background so much less blurred in the first photo vs the second? In the top, the fronds are nearly individually visible, in the 2nd, they are a mass of single colour variations, no individual stocks are visible. Literally all that was changed was focal length from 100mm to 200mm (and perhaps some minor lateral movement), nothing else?

First of all I do not accept that the background in the first photo is less blurred. I think at the same enlargement you would find the relative blur to be the same in both photos. Beyond that, you have answered your own question. By remaining at the same distance from your subject but just changing focal length, the DOF of the shorter lens is much greater than that of the longer lens. No great mystery. But your original suggestion was to use a shorter focal length lens to increase DOF (at the subject), not to obtain a wider fov. So it's a fair assumption that, when switching to a shorter lens, one would move closer to the subject in order to achieve the same framing. And when one does that the increase in DOF is vanishingly small. So why not go outside in that lovely snow and try this with that handsome couple and show us the results.

Or you could just plug the numbers into a DOF calculator of your choosing and see what comes up.

[jerryph]

Wow... ok. If you cannot accept what you see with your own eyes, and then say:

... I think at the same enlargement you would find the relative blur to be the same in both photos.

That is saying to me "if the first photo and the second photo were at the same focal length, they would have the same amount of blur"... uhmmm... yeah they would, but that just again proves the point. It is obvious that the aperture is not the only thing that controls DOF, focal length is a big factor, not the only factor, but it *is* a factor.

Again, if this alone is not enough to prove it, forget the calculator, go out there and do the test yourself and see whether or not 50mm and 200mm share a 1 inch DOF difference or not. You will see exactly what I saw, that it is a lot more than 1 inch. As mentioned before, if you need irrefutable proof, look at the 2 photos above. Even if I crop in that 50mm at the distant trees to the same POV that the 200mm has, the 50mm will have *still* have branches that are way more distinct over the 200mm ones.

One would not have to think that no excuses need to be made at this point... it's not just obvious... but damned obvious... lol

[Preeb]

Originally Posted by blumesan

First of all I do not accept that the background in the first photo is less blurred. I think at the same enlargement you would find the relative blur to be the same in both photos.

Try it yourself! Even though both of Jerry's photos are quite small, I copied both into Elements, then enlarged the top one until the subjects were the same size in both. The difference in background blurring is still obvious, even with the rather extreme pixelation of the enlargement. If you pan the view in the enlargement to the left you can see the same background trees in the same size as they are in the 200mm photo, but they are not nearly as blurred.

I don't dispute your statement that if the subject framing is the same, then DoF is similar, but just changing focal length without changing distance to subject does change DoF and background blurring significantly. In a case where a different bokeh is desired, using a shorter lens and cropping can create a different effect.

[jerryph]

Originally Posted by Preeb

...just changing focal length without changing distance to subject does change DoF and background blurring significantly. In a case where a different bokeh is desired, using a shorter lens and cropping can create a different effect.

... and if we are seeing differences in amount of blur, that proves a change in DOF, and that change is not limited to one aspect of the frame, but the entire frame (only exception to that would be a tilt-shift lens, which I own, but it was NOT the lens used in the making of these 2 photos, I used the Nikkor 70-200 F/2..

Depth of field increases with nothing more than a change of focal length is irrefutable.

Again, is it the only way? No, it is one of the 5 factors that defines what the DOF will be, but no one should dispute that it is not a factor... and that DOF calculator may have a glitch in it somewhere or not take into consideration all factors for most accurate results.

[jerryph]

Originally Posted by blumesan

My example considers focal length, aperture, and image magnification (such that the object width at the plane of focus is the same, i.e. the same framing of the subject). It assumes the same size sensor (FX) in both cases.

I've found what I think is the discrepancy.

You took into consideration:
- focal length
- aperture
- sensor

You did not take into consideration the apparent distance between:
- camera to subject
- camera to background

This ratio is critical to the proper calculation of DOF.

You are confusing magnification with the whole thing... expecting that a wide angle lens and a long lens to have the same framing. This is not possible, not unless you change distance between you and the subject (by moving back or forward), at which point you also alter the entire ratio of camera to subject vs camera to background.

You cannot change one without changing the other... or if you change only one... DOF *will* change.