Pages: [1] 2   Go Down

Author Topic: Focusing Almost at Infinity  (Read 7587 times)

John R Smith

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1357
  • Still crazy, after all these years
Focusing Almost at Infinity
« on: November 18, 2010, 08:54:16 am »

A subject which was briefly touched on in the Pentax 645 review was the problem of focusing at distances which are almost, but not quite, at infinity. In the article this referred to AF lenses, but I am having trouble with this very problem using my old CZ MF lenses.

The issue arises principally with the 150mm Sonnar. If my subject is say 200 to 400 feet away from me, and I set the lens focus to infinity, although DOF at f11 or f16 should theoretically cover things, when I view the result on the PC screen the subject is noticeably soft, made worse by the fact that the background is pin-sharp. However, when I view the scene through the 4X magnifier on my camera, I simply can’t see the difference in focus between the subject and the background no matter how careful I am. And the scale on the lens stops at 100 feet, beyond which there is only a tiny rotation to the end stop at infinity, so guessing the distance and setting it on the scale is kind of hit-or-miss.

This is really annoying me, because I have now (with care and a lot of practice) just about cracked all my other focus issues and very rarely miss. Any advice you chaps might have would be most welcome.

John
Logged
Hasselblad 500 C/M, SWC and CFV-39 DB
an

ondebanks

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 858
Re: Focusing Almost at Infinity
« Reply #1 on: November 18, 2010, 09:08:13 am »

 John,

IIRC you shoot with a digital back on your Hassy? Then you could either (a) blindly, manually "focus bracket" and assume that one shot at least will nail it or (2) if you have the time, "chimp": zoom in on the LCD after each shot, check focus and focus bracket in an informed manner. This of course assumes that it suits your subject matter and style of shooting.

Ray
Logged

cunim

  • Full Member
  • ***
  • Offline Offline
  • Posts: 130
Re: Focusing Almost at Infinity
« Reply #2 on: November 18, 2010, 11:22:06 am »

I asked HUSA about this very thing.  I was setting to infinity at f9, seeing soft with objects that should have been in focus, while the distance was sharp.  There is no on-the-lens setting that discriminates this type of thing.  Answer "That's just the way it is".  I was advised to use AF which does appear to nail it with some targets - others not so much.  Trouble is my aging eyes can't fine tune focus of small objects in a view finder.  Easier on ground glass.  I wonder if this problem is endemic.
Logged

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: Focusing Almost at Infinity
« Reply #3 on: November 18, 2010, 11:54:15 am »

Hi!

I'd say that infinity is much farther away than 300-400 feet with a 150 mm lens at a critical viewing distance.

http://www.cambridgeincolour.com/tutorials/DOF-calculator.htm

Best regards
Erik


A subject which was briefly touched on in the Pentax 645 review was the problem of focusing at distances which are almost, but not quite, at infinity. In the article this referred to AF lenses, but I am having trouble with this very problem using my old CZ MF lenses.

The issue arises principally with the 150mm Sonnar. If my subject is say 200 to 400 feet away from me, and I set the lens focus to infinity, although DOF at f11 or f16 should theoretically cover things, when I view the result on the PC screen the subject is noticeably soft, made worse by the fact that the background is pin-sharp. However, when I view the scene through the 4X magnifier on my camera, I simply can’t see the difference in focus between the subject and the background no matter how careful I am. And the scale on the lens stops at 100 feet, beyond which there is only a tiny rotation to the end stop at infinity, so guessing the distance and setting it on the scale is kind of hit-or-miss.

This is really annoying me, because I have now (with care and a lot of practice) just about cracked all my other focus issues and very rarely miss. Any advice you chaps might have would be most welcome.

John

Logged
Erik Kaffehr
 

ndevlin

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 679
    • Follow me on Twitter
Re: Focusing Almost at Infinity
« Reply #4 on: November 18, 2010, 09:33:19 pm »

...If my subject is say 200 to 400 feet away from me, and I set the lens focus to infinity, although DOF at f11 or f16 should theoretically cover things, when I view the result on the PC screen the subject is noticeably soft, made worse by the fact that the background is pin-sharp. However, when I view the scene through the 4X magnifier on my camera, I simply can’t see the difference in focus between the subject and the background no matter how careful I am. And the scale on the lens stops at 100 feet, beyond which there is only a tiny rotation to the end stop at infinity, so guessing the distance and setting it

Yes! That's EXACTLY the problem I have had with a number of MF systems, including Hassy C with Phase backs, Mamiya DF and, to a limited extent, the 645D.

It's annoying beyond words, especially since so much landscape work is done at these distances, and MF's ability to render ultra-fine detail (even at these distances) is one of its principal advantages.

- N.
Logged
Nick Devlin   @onelittlecamera        ww

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: Focusing Almost at Infinity
« Reply #5 on: November 19, 2010, 02:23:29 am »

Hi,

What I observed is on full frame digital that DOF is very limited. If we look at pixel level the circle of confusion becomes something like pixel pitch. DOF scales on MF are probably based on CoC (Circle of Confusion) of 1/15 mm, that is 67 microns. Some folks recommend the use of f/8 marks for depth of field when stopping down to f/16. That practice would still give a CoC of 33 microns. Sensor pitch on the Pentax 645D is 5.93my, so about 24 pixels would fit into what is considered sharp on the focusing scale, using the conservative approach. If normal (1/15 mm criteria is used) a blob of 96 pixels would be considered sharp.

Additionally, there is no guaranty that the infinity mark is exactly at infinity, because of manufacturing tolerances. It is reasonable to presume that the mechanical stop is beyond infinity, because of the said tolerances. Otherwise correct focus at infinity could not be achieved on half of the lenses!

DOF scales were invented for smallish prints, not pixel peeping. With digital and small pixels the DOF at the pixel level is virtually nil.

Best regards
Erik


Yes! That's EXACTLY the problem I have had with a number of MF systems, including Hassy C with Phase backs, Mamiya DF and, to a limited extent, the 645D.

It's annoying beyond words, especially since so much landscape work is done at these distances, and MF's ability to render ultra-fine detail (even at these distances) is one of its principal advantages.

- N.
Logged
Erik Kaffehr
 

Rob C

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 24074
Re: Focusing Almost at Infinity
« Reply #6 on: November 19, 2010, 03:22:04 am »


DOF scales were invented for smallish prints, not pixel peeping. With digital and small pixels the DOF at the pixel level is virtually nil.
Best regards
Erik





I can tell you that one of the best tips I was ever given was that one should treat DOF as a myth. And that was in film days, and when using 35mm cameras. The advice, decades ago, was to make your prime point of interest the one on which you base focus.

Guess the adviser was prescient.

Rob C

Bart_van_der_Wolf

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 8913
Re: Focusing Almost at Infinity
« Reply #7 on: November 19, 2010, 07:57:30 am »

What I observed is on full frame digital that DOF is very limited. If we look at pixel level the circle of confusion becomes something like pixel pitch.

Hi Erik,

That's correct. The "Circle of Confusion" concept is not being applied properly by most.

On film, there was less of a tendency to pixel-peep. Not only did we not work with discrete pixels, an optical loupe was the tool du jour, we worked much more with final output sizes as a measure of sharpness / fitness for intended purpose. That is how the COC should be used, is it 'sharp enough' for the intended viewing distance, the only quibble being what 'sharp enough' means. For the most stringent use, one could define it as 'not possible to increase the perceived sharpness' (something close to 8 lp/mm for someone with 20/20 vision at normal reading distance).

When pixel peeping, we can perceive small differences down to the pixel level (magnified on our relatively low resolution 2 lp/mm displays), so we would need to use an appropriate COC for that, but one might question the relevance of the close-up observation. It might be more relevant when we view the display from 4x the normal viewing distance. Do we then still see a difference between infinity and far focus? That would come closer to the 'not possible to increase the perceived sharpness' qualification.

Everything becomes much easier to predict when the proper COC is used in a DOF calculator. For large output, that means that one cannot improve on the resolution of the sensor array divided by the magnification factor for output, unless one resorts to stitching with relatively longer focal lengths (which only reduces the DOF for pixel peeping, not for higher resolution printing quality).

Cheers,
Bart
« Last Edit: November 19, 2010, 10:02:44 am by BartvanderWolf »
Logged
== If you do what you did, you'll get what you got. ==

yaya

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1254
    • http://yayapro.com
Re: Focusing Almost at Infinity
« Reply #8 on: November 19, 2010, 08:11:33 am »

A subject which was briefly touched on in the Pentax 645 review was the problem of focusing at distances which are almost, but not quite, at infinity. In the article this referred to AF lenses, but I am having trouble with this very problem using my old CZ MF lenses.

The issue arises principally with the 150mm Sonnar. If my subject is say 200 to 400 feet away from me, and I set the lens focus to infinity, although DOF at f11 or f16 should theoretically cover things, when I view the result on the PC screen the subject is noticeably soft, made worse by the fact that the background is pin-sharp. However, when I view the scene through the 4X magnifier on my camera, I simply can’t see the difference in focus between the subject and the background no matter how careful I am. And the scale on the lens stops at 100 feet, beyond which there is only a tiny rotation to the end stop at infinity, so guessing the distance and setting it on the scale is kind of hit-or-miss.

This is really annoying me, because I have now (with care and a lot of practice) just about cracked all my other focus issues and very rarely miss. Any advice you chaps might have would be most welcome.

John


A simple ruler test with a standard lens, wide open, should be able to tell you if the sensor is where it should be or not. If you do it tethered with Live View you'll also be able to tell if the focusing screen/ prism are out of alignment
Logged
Yair Shahar | Product Manager | Phase One - Cultural Heritage
e: ysh@phaseone.com |

John R Smith

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1357
  • Still crazy, after all these years
Re: Focusing Almost at Infinity
« Reply #9 on: November 19, 2010, 09:57:50 am »

Thank you all for the useful replies.

I think actually I am probably talking about more like 500 feet away from me, rather than 200 to 400 as I said. I can see focus at 200, just about. Perhaps focus bracketing, as suggested, is the only practical option. This is an example - you won't be able to see it in the jpeg, but the subject (the boat) is just slightly soft, whereas the background is pin-sharp. This is OK in a 10x8 print, but it would not stand a really big enlargement.

John

« Last Edit: November 19, 2010, 10:37:58 am by John R Smith »
Logged
Hasselblad 500 C/M, SWC and CFV-39 DB
an

Rob C

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 24074
Re: Focusing Almost at Infinity
« Reply #10 on: November 19, 2010, 11:27:50 am »

John, my first question was where the hell's the boat?

Now, that's easy to see as a slight, but it isn't meant as such. The point is that focus determination (to the eye) is very dependent of sharp edges and was where the split image screens came in useful. Do you have one for your 500 Series? You could get them at one time. Straight ground-glass focussing is always a dodgy thing; with the relatively slow lenses on 'blads (my 150mm was an f4, if I recall) it is even more difficult, but in less than terrible light the split-image screen would still work for you without going half-black, but you'd still need the straight lines.

Rob C

claudefiddler

  • Newbie
  • *
  • Offline Offline
  • Posts: 21
Re: Focusing Almost at Infinity
« Reply #11 on: November 21, 2010, 06:51:34 pm »

John

Use the moon, (it's close to or at full tonight), to test where your lens focuses on infinity.

I'm not kidding.

Claude Fiddler
www.wildernesslight.com

Logged

Policar

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 56
Re: Focusing Almost at Infinity
« Reply #12 on: November 21, 2010, 07:38:47 pm »

John

Use the moon, (it's close to or at full tonight), to test where your lens focuses on infinity.

I'm not kidding.

Claude Fiddler
www.wildernesslight.com



Clever.  I will remember this.

Any advice for testing film/ground glass registration on LF, btw?  I am scared to shoot wider than f22 at the moment with my old fidelity film holders.
Logged

John R Smith

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1357
  • Still crazy, after all these years
Re: Focusing Almost at Infinity
« Reply #13 on: November 22, 2010, 03:39:23 am »

The point is that focus determination (to the eye) is very dependent of sharp edges and was where the split image screens came in useful. Do you have one for your 500 Series? You could get them at one time. Straight ground-glass focussing is always a dodgy thing; with the relatively slow lenses on 'blads (my 150mm was an f4, if I recall) it is even more difficult, but in less than terrible light the split-image screen would still work for you without going half-black, but you'd still need the straight lines.

Rob C

Rob, I have every type of screen pretty much, including the one with the centre split-prism. But at that distance I can't see any difference in the split prism across a straight line. Between 100 feet and the infinity mark on the 150mm lens is only about 1/4 inch of movement on the helicoid. In this case I know that the boat was 200 feet away from me (measured on my GIS system), so I was operating in the last 1/8 inch of movement in the helicoid, probably. And given that this is a forty-year old lens with a bit of slack in it, as well . . .

When I am shooting B/W film at these apertures (f11 or f16) I just set the old girl on infinity and my prints are fine. But this digital sensor picks up differences in focus that I never even knew were there in the past.

John
« Last Edit: November 22, 2010, 04:16:26 am by John R Smith »
Logged
Hasselblad 500 C/M, SWC and CFV-39 DB
an

Rob C

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 24074
Re: Focusing Almost at Infinity
« Reply #14 on: November 22, 2010, 01:44:09 pm »

Rob, I have every type of screen pretty much, including the one with the centre split-prism. But at that distance I can't see any difference in the split prism across a straight line. Between 100 feet and the infinity mark on the 150mm lens is only about 1/4 inch of movement on the helicoid. In this case I know that the boat was 200 feet away from me (measured on my GIS system), so I was operating in the last 1/8 inch of movement in the helicoid, probably. And given that this is a forty-year old lens with a bit of slack in it, as well . . .

When I am shooting B/W film at these apertures (f11 or f16) I just set the old girl on infinity and my prints are fine. But this digital sensor picks up differences in focus that I never even knew were there in the past.

John


John

Maybe the problem isn't the camera or you at all, perhaps it's just that one decade's technology isn't always best mated with another's.

All the years I used 500 equipment I had few problems I can remember. The only repeat one seemed to be a faulty delayed action device on the shutter of my 50mm Distagon.

The 150mm gave me great results every time - any failures were due to either user error, such as shooting into the light at the wrong angle, if that's a user error, and I would have wanted to change it for only one lens, the 180mm which didn't exist when I bought the 150mm.

I still have problems accepting that quality at print sizes can be less with digital than with film; at 100% yes, I get that, but at 16 x 20 prints? Dammit, the same errors must have existed in the day, add less resolution from film and film should still look inferior to digital...

Rob C

John R Smith

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1357
  • Still crazy, after all these years
Re: Focusing Almost at Infinity
« Reply #15 on: November 23, 2010, 03:26:01 am »


John

The 150mm gave me great results every time - any failures were due to either user error, such as shooting into the light at the wrong angle, if that's a user error, and I would have wanted to change it for only one lens, the 180mm which didn't exist when I bought the 150mm.

I still have problems accepting that quality at print sizes can be less with digital than with film; at 100% yes, I get that, but at 16 x 20 prints? Dammit, the same errors must have existed in the day, add less resolution from film and film should still look inferior to digital...

Rob C

Rob

I agree entirely that the 150mm Sonnar is a great lens. I get terrific results from it with insane amounts of detail when everything goes well. But I am not saying that my prints from the CFV-39 are in any way inferior to film, that is not the issue. Now I have got the hang of things a bit, my prints from the DB are superior to those from scanned B/W film in pretty much every respect, as far as IQ goes. The snag is, because they are so grain-free and smooth and have better resolution for the same print size, you can see problems with focus, DOF, and camera-shake that I never knew were there before. So in many respects this has been really good for me, because it has been a wake-up call and I have had to improve my technique in all sorts of ways. Since January, when I combined this piece of very expensive hi-tech with my venerable camera collection which most folks would give to a museum, I have had to re-examine how I meter exposure, get focus, assess DOF, and avoid mirror-slap and camera shake. As a result, I can go out for the afternoon now and come back with 90% or more usable technically spot-on frames. When I first started with the DB I was lucky to come back with 50%.

John
Logged
Hasselblad 500 C/M, SWC and CFV-39 DB
an

Rob C

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 24074
Re: Focusing Almost at Infinity
« Reply #16 on: November 23, 2010, 03:39:22 am »

Well, you're winning the battle of percentages to say the least!

I wouldn't consign my 500 stuff to a museum - if I still had it (idiot, Rob, blind, credulous, stupid moron!) I would have probably gone for one of those eye-wateringly expensive 120 scanners and be relatively pleased with myself. As it is, scanning forty-something 35mm trannies over the past week and trying to bring something worthwhile out of it all is making severe dents in my enthusiasm for photography as a soothing, gentle, assisting-me-to-slowly-the-tomb enterprise!

;-)

Rob C 

David Grover / Capture One

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1324
    • Capture One
Re: Focusing Almost at Infinity
« Reply #17 on: November 23, 2010, 03:40:14 am »

John,

Are you winding the lens all the way to the infinity stop?

If so you are very likely to be focussing 'beyond' infinity... Which of course doesnt exist in real terms but will on the lens.

The infinity stop has to be this way as due to ambient temperatures changes this can change a little. Therefore if the infinity stop was at actual infinity it might then be impossible to achieve infinity focus in diffent environments.

Yaya gives good advice suggesting to check the focus accuracy of the system on a 45 angled ruler.

David.
Logged
David Grover
Business Support and Development Manager

Will Ophuis

  • Newbie
  • *
  • Offline Offline
  • Posts: 40
Re: Focusing Almost at Infinity
« Reply #18 on: November 23, 2010, 04:40:51 am »

^ ah that makes a lot of sense, Ive found I will turn the focus right to infinity and then will have to turn back slightly.
Logged
Hasselblad H4D-40.

www.williamophuis.com/blog

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: Focusing Almost at Infinity
« Reply #19 on: November 28, 2010, 04:40:09 pm »

Hi!

I made a small experiment trying to emulate depth of field in lab conditions and have a small writeup on my preliminary findings here:

http://echophoto.dnsalias.net/ekr/index.php/photoarticles/49-dof-in-digital-pictures

The findings are that there is an obvious and visible loss of image quality on a 150 mm lens at f/8 focused at infinity at a distance of about 240 m.

The article is short on explanations but will be extended later on.

Best regards
Erik


A subject which was briefly touched on in the Pentax 645 review was the problem of focusing at distances which are almost, but not quite, at infinity. In the article this referred to AF lenses, but I am having trouble with this very problem using my old CZ MF lenses.

The issue arises principally with the 150mm Sonnar. If my subject is say 200 to 400 feet away from me, and I set the lens focus to infinity, although DOF at f11 or f16 should theoretically cover things, when I view the result on the PC screen the subject is noticeably soft, made worse by the fact that the background is pin-sharp. However, when I view the scene through the 4X magnifier on my camera, I simply can’t see the difference in focus between the subject and the background no matter how careful I am. And the scale on the lens stops at 100 feet, beyond which there is only a tiny rotation to the end stop at infinity, so guessing the distance and setting it on the scale is kind of hit-or-miss.

This is really annoying me, because I have now (with care and a lot of practice) just about cracked all my other focus issues and very rarely miss. Any advice you chaps might have would be most welcome.

John

Logged
Erik Kaffehr
 
Pages: [1] 2   Go Up