tennclay
Offline
• • • •
Upload & Sell: On
|
 p.5 #18 · Is film really going away??? | |
brainiac wrote:
In tennclay's 65mm (effective) example above, the angle of view should be about 15 degrees, so the swing is probably no more than 5 degrees. The calculated error at one meter is about 4mm. Funnily enough, the swing does appear to have sharpened up the 19 marker a little, and certainly the 18 marker and the text under it are sharper after the swing. Also, look at the upright line behind the ruler which is sharper at the edge of the frame. Compare the 23's. Focus has moved back. Would it be fair to say that tennclay's result corroborates rather than refutes a calculated back-focus of a few millimetres? And would 4mm be enough to take the crispness out of eye glints?
As I said before, the effect becomes more frustrating if you use bright normal and wideangle lenses, where the swing can be 20 or 30 degrees. Telephoto examples like this one aren't relevant....Show more →
The error demonstrated in your calculation would indicate a problem, ceteris paribus. But, again, the question is whether it is a problem in practice. There are more factors in play than you account for.
For example, you ignore DOF in the "equation". Your calculation gives 12 cm of focus error, looking quickly at DOF master you will see a calculated DOF of 27cm (35mm/1.4 Film). Of course DOF is very subjective and arbitrary, but why would you ignore it? Is DOF enough to overcome the focussing error? I believe it depends - where does the lens best point of focus lie within the DOF (front, rear, middle), what is the size at which you will view it, and what is acceptable focus for you personally?
Using your formula and substituting 1 meter for 2 meters, you'll see that the focus error will decrease by about 1/2. Doesn't this mean that there would be an increasing focusing problem at larger distances? Of course not, in practice one knows that it more of an "frustration" at close distances. This is the practical impact of DOF and its inverse relation to focusing error as distance changes.
Another contributor mentioned previously is optics - take a look at this link and you'll see that it would impact the situation as well. http://www.vanwalree.com/optics/astigmatism.html
Now regarding the irrelevance of my test, I won't argue the point of narrower FOV making a difference. But I am interested in your estimation of 5% swing with my test. Could you reference a formula for that ?
Also - I did test with a wider lens (28mm) , though it is only a F2.0 and I do see backfocus in an extreme close focus test. However, I would rarely take a shot under these circumstances and never a portrait due to the obvious perspective problems. You are saying that is a critical criteria for you, so again, it seems the frustrations of a RF would not satisfy your needs. BTW, when I repeated the test at about 2.5 meters, there is no noticeable problem.
Now with regard to your seeing backfocus in my demonstration, don't place much stock in the minute things you are seeing (personally I don't see it the same way). This was an impromptu test - no tripod - just me doing what I would do in a practical situation when recomposing. How much my head shifted when swings the camera is unknown.
Was the shot you posting with a 35/1.4 WO on a fullframe camera? What would be the degree of swing if recomposing? If that is a good example of the type of shot that has frustrated you, I don't think it would be a RF that would hold you back. It looks like the man is not particularly in focus, but I can see whay it would be hard to focus accurately with a dim SLR VF, even in that relatively high level of light.
Again, I am sure this would not be a significantly frustrating problem the majority of people. And those who do find it a problem - there is the SLR, thank goodness.
Edited on Jul 21, 2008 at 07:06 AM
|
