If I have learned anything from my large format journey so far, it has been that sample variation is frustratingly more important than any other factor, at least with used modern lenses. If you find a good lens, never sell it. I am still waiting to try the lenses I bought (I can update in September), but it is really hard to get much objective information, as the results themselves seem to vary too much between lenses and people. This has not really been my experience in smaller formats, but in LF, it seems that the best way is to acquire several lenses that are generally regarded as "good" and then just test.
It is true that there is a sample to sample variation for the ultimate peformance, but this hapens at contrast extintion with high lp/mm, and in a region in what regular film anyway is not resolving all what a lens can deliver so while there is certain impact in ultra fine detail... that it's hard to be seen in a real photograph.
At the end we are not in a lab with flat targets, we usually photograph 3d objecs and nothing is in absolutely perfect focus, some times we need to stop the lens for DOF sake and limiting the lens performance. Modern 4x5 lenses are diffraction limited by f/22.
Rodenstock, Schneider, Fuji and Nikon were very serious manufactures and it's hard to find a lens from them that is not perfectly able, delivering tons of image quality.
The weakest link in the LF chain, IMHO, it's the photographer himself. Today we all can have gear that's way better than the one owned by AA, or Karsh. Another thing is making a comparable work...
Hi Pere,
This has not been my experience. Perhaps it is not "sample variation" so much as buying lenses that are broken. Or it could be that my workflow is more demanding. I am not trying to be arrogant or say that I am a better photographer than the greats or that these lenses could not make great images, only that I see very clear visible differences in my workflow between the lenses I have bought, and the quality of the lens IS the limiting factor with several of my lenses. I am often using flash in nature at apertures from f/11 to f/22, and I get phenomenal results from a few lenses, but from others the results have been less spectacular. This is consistent over time and throughout testing. I cannot say if it is the lens itself that is broken or suboptimal, or if it is that the "great" lenses are just better suited to my camera. But the fact remains that I can get great results with some lenses, not others. I don't think this is my failure as a photographer...I think this is the failure of the lens to perform as well as another lens. I run a lab printing exhibition prints and exhibit my own work often at large scale, so I am quite accustomed to working within fairly demanding constraints. I know when something is possibly thrown off because of wind or missed focus. I am not talking about that, I am talking about consistent results over a large sample of images indicating that one lens is performing better, and how that seems to be less tied to the maker of the lens and its model than to which lens copy it is. For example, I have had opportunity to test two APO Sironar S 150mm lenses, (well three, but one turned out to be a frankenstein lens that someone sold me with a Rodenstock front cell and Schneider rear cell), and one was clearly better than the other. I sent my copy to Rodenstock to repair, so my hope is that it comes back better, but we will see. And when I say test, I mean I produced an exhibition consisting of 12 1mX1.25m prints and 30 40x50cm prints for the National Museum of Iceland, of which I scanned and printed all the originals. So I saw a lot of samples...I then shot it alongside my own copy of the lens making test images both in the studio and in the field, and I could see variations.
If you believe you can, or you believe you can't... you're right.
Optimal shimming it's easy to check. I placed a microscope eyepiece (periplan, etc) in a metal plating sized like a 4x5 film holder, a 30x is perfect:
Then, with GG removed, you unscreew the front lens until you have we have best image in the center. By the number of tours and pitch we know the shimming we need. Note that we have to iterate unscreew+focus+check.
It may happen that optimal shimming for the center is not the optimal for the corner.
If we want to check that we can make some test shots of a resolution chart in we focus optimally for the center, but in each shot we screewed/unsecreewed the front cell from the optimal shimming for the center, then we judge if we may move the shimming a bit (from the optimal position for the center) in a way we get the best balance.
My guess is that we can make a shimming that's slightly better for the center or for the corners.
Instead the eyepiece, we can also place a dslr in the back, without the dslr lens, of course...
OK.
But as soon as you loosen the front element you introduce tilt due to clearance in the threads.
That throws the front cell off-axis relative to the rear cell.
???
A proper optical test jig holds both cells rigid, and allows the spacing to be changed accurately.
Then the spacing is measured, compared with the shutter thickness, and shims calculated.
- Leigh
If you believe you can, or you believe you can't... you're right.
Bookmarks