| Author |
Message |
    
Nikos
Tinkerer
Username: Nikos
Post Number: 22
Registered: 07-2006
Rating: N/A
Votes: 0 (Vote!) | | Posted on Thursday, September 20, 2007 - 11:27 pm: | 
|
The other day I was using an Exakta mount semi-automatic Biotar, mounted on my old Topcon RE Super. As there is no linkage of any kind between lens aperture and camera, I had, after focusing, to take a stop down reading. I noticed an appreciable shift of focus and I wonder whether this is a true shift of focus or just an optical illusion. So, is there a case for stop down focusing?
Nikos |
Rick_oleson
Tinkerer
Username: Rick_oleson
Post Number: 383
Registered: 07-2006
Rating: N/A
Votes: 0 (Vote!) | | Posted on Friday, September 21, 2007 - 03:49 am: |
|
This is basically an indication of spherical aberration in the lens: the outer areas are coming to a different point of focus from the center. When the lens is wide open most of the light is coming through the outer area which of course has the most surface area, and you don't see the light from the center... when you stop down this is all cut out and the center light is what you see. It's not a very desirable characteristic in a lens, but it might have been a trade-off that was accepted to some degree in exchange for high speed and good general performance in an early lens. Still, it seems that if it's a large error there might be something not quite right with your lens. |
Nikos
Tinkerer
Username: Nikos
Post Number: 23
Registered: 07-2006
Rating: N/A
Votes: 0 (Vote!) | | Posted on Friday, September 21, 2007 - 10:08 am: |
|
Thank you Rick. I should have thought of that. I will develop and scan the film tonight and see if there is something wrong with my lens. |
Rj_
Tinkerer
Username: Rj_
Post Number: 22
Registered: 08-2007
Rating: N/A
Votes: 0 (Vote!) | | Posted on Saturday, September 22, 2007 - 05:07 am: |
|
Adding a little to Rick's post, I find that stopping down the aperture covers the focus point through the extended depth of field, in front of and beyond the focus point of the wide-open aperture. This finding would suggest that the opposite is true, and that your final image captured on the negative, would indeed be in focus with more stopping down of the aperture. On the condition that other variables, such as shutter speed and camera shake are controlled.
The spherical aberration affects most lenses; even on modern Carl Zeiss f1.4 lenses for 35mm format, there is a significant focus shift at the wide-end of focussing, enough to result in focus error if the SLR/rangefinder technique of focussing is followed without compensation at maximum aperture.
This focus shift improves with stopping down, and as Rick has pointed out, is a consequence of most lens design, particularly with the curvature of the front element of the lens at such a shallow depth of field.
Still, it doesn't seem to explain the findings that are visible for you, and the position of the focussing screen, mirror, or any other implement in the viewing light path, in relation to the light path of the exposed image may need to be considered.
How did the scan go? (Provided the image is scanned at an appropriate height from the scanner glass  ) |
Bob_dodds
Tinkerer
Username: Bob_dodds
Post Number: 3
Registered: 09-2007
Rating: N/A
Votes: 0 (Vote!) | | Posted on Saturday, September 22, 2007 - 06:32 am: |
|
For hand-held shots at large apertures, narrow DOF, it is likely that you will move off focus while you are stopping down.
If you are braced or on a tripod, it is likely that increased DOF at smaller apertures will overcome imprecise focus.
I just did a crazy test where I focused a Zuiko 50mm f1.2 at f8 and shot at f1.2 and f8, and focused at f1.2 and shot at f1.2 and f8. Those were on a tripod, and the results turned out the same(comparing f1.2 to f1.2 and f8 to f8). |
Rick_oleson
Tinkerer
Username: Rick_oleson
Post Number: 384
Registered: 07-2006
Rating: N/A
Votes: 0 (Vote!) | | Posted on Saturday, September 22, 2007 - 10:32 am: |
|
It's drifting a bit off topic here, but it just occurred to me that this might explain the occasional phenomenon of the focus in the field of the groundglass not seeming to quite agree with the split image rangefinder spot. The RF spot is only looking at the rays coming in at about the f/5.6 position, while with a fast lens the majority if the image in the field is formed by rays from the outer region of the lens, well outside of the f/5.6 zone. If there is a significant focus shift between these two zones in a given lens, you may see a disagreement in the finder. |
Hovaness
Tinkerer
Username: Hovaness
Post Number: 43
Registered: 07-2006
Rating: N/A
Votes: 0 (Vote!) | | Posted on Sunday, September 23, 2007 - 07:07 pm: |
|
The worst case of this focus shift I have encountered is in the lens of the well-regarded Olympus 35RC. At full aperture (f2.8), there is a significant infinity focus shift, resulting in soft results. I confirmed this problem on multiple samples of the Zuiko 42mm f2.8 lens using a collimator. |
Charlie
Tinkerer
Username: Charlie
Post Number: 98
Registered: 07-2006
Rating: N/A
Votes: 0 (Vote!) | | Posted on Monday, September 24, 2007 - 06:44 am: |
|
Doesn't depth of focus cancel out focus shift, especially at infinity? Also, scale focussing would seem to have the same problem, how does the maker mark the focus scale if he doesn't know what f stop you will be using. I'm not a big fan of SLR's but stop down focussing would seem to be the only cure. Or better lenses. |
M_currie
Tinkerer
Username: M_currie
Post Number: 76
Registered: 07-2006
Rating: N/A
Votes: 0 (Vote!) | | Posted on Tuesday, September 25, 2007 - 06:29 am: |
|
Hovaness, since I've been using an Olympus 35 RC a lot recently and hope to continue doing so with optimal results, could you tell me in which direction that focus shift occurs, if there's any consistency to it? Can I compensate by focusing a little closer, or does it go out of range in the other direction?
I haven't used mine under sufficiently controlled conditions to judge when and why the occasional soft results occur. |
Chenard
Tinkerer
Username: Chenard
Post Number: 19
Registered: 11-2006
Rating: N/A
Votes: 0 (Vote!) | | Posted on Tuesday, September 25, 2007 - 02:36 pm: |
|
This is interesting. I noticed a very similar effect with my Agfa Isolettes, which are scale focused cameras. I discovered it while attempting to set infinity and close focus using a ground glass .
You can read about it in this thread:
https://kyp.hauslendale.com/classics/forum/messages/2/9283.html?1177951146
So...if there is focus shift due to spherical aberrations wide open, what is the correct aperture to use while setting the focus on a scale focused camera? Where in the image should the sharpness be judged while setting the focus--middle, corner, halfway?
Paul |
Rick_oleson
Tinkerer
Username: Rick_oleson
Post Number: 385
Registered: 07-2006
Rating: N/A
Votes: 0 (Vote!) | | Posted on Tuesday, September 25, 2007 - 07:52 pm: |
|
If it's spherical aberration, which would seem to be the obvious cause, I don't think it makes a lot of difference where in the field you're looking.... except that generally, the best sharpness will be in the center with most lenses so it's probably easiest to focus there.
In most respects, I think lens designers try to make their lenses perform best in the middle stops. This gives a better balance of overall performance than if they optimize it at one end or the other, as you're always within a couple of stops of the best setting that way. However, in the case of focus, this might be overruled by the change in depth of field: if you adjust it for best wide open focus, you have a better chance of the increased DOF covering the shift for you as you stop down. Optimize focus in the middle stops, and wide open you'll have a combination of focus error and shallow depth of field... not a promising mix. |
Nikos
Tinkerer
Username: Nikos
Post Number: 25
Registered: 07-2006
Rating: N/A
Votes: 0 (Vote!) | | Posted on Sunday, September 30, 2007 - 05:42 pm: |
|
Thanks everybody for trying to enlighten me. This turned out to be a rather interesting thread.
I did manage to develop my films. The only trouble is that I thought I was shooting (and developing) FP4 while actually the camera was loaded with Efke 25. The only useable frames were the ones intentionally overexposed. No, there was nothing wrong with my lens.
What intrigued me in the first place was the fact that the shift of focus was only in reference to the central split image microprism—something that I neglected to mention in my original post. I have no idea whether there was any focus shift on the groundglass. I was focusing using the central spot. |
M_currie
Tinkerer
Username: M_currie
Post Number: 78
Registered: 07-2006
Rating: N/A
Votes: 0 (Vote!) | | Posted on Monday, October 01, 2007 - 07:26 am: |
|
My limited understanding of split images is that they are not accurate at all apertures, but are optimized for a particular aperture, according to the angle of the prisms. For example, I have for my Nikon F a special split image screen designed for slower lenses, and it comes with a warning that it will cause focusing error with faster lenses. This is on the assumption that focusing is done wide open, but I could see the possibility of a focus shift if one tries it at other apertures. |
