View Full Version : Gaoersi 617 infinity focus probs

Pages : 1 [2]

Ian Grant
06-16-2009, 11:58 AM
You can't 'design' a lens to have any specific diffraction characteristic. It a physical function of aperture size. Its simple, the smaller the aperture, the more the diffraction (or the larger the bullseyes). Its physics of light, not design of the lens.

As aperture is relative to focal length it goes without saying that an LF lens at f22 is quite different to a an equivalent 35mm or 120 lens of the same field of view at the same aperture. Diffraction is more apparent with lenses of smaller formats and manufacturers of LF lenses take this into account when designing their lenses, it's not a co-incidence that virtually all of them state the optimum aperture is f22 for their LF lenses.

As you say it's simple but the diameter of an aperture of f22 on a 35mm camera is more likely to be equivalent of f45 or even f64 with an LF lens, and f90 or f128 with ULF lenses, hence the reason f22 is used for best overall balance with 5x4 lenses.


06-17-2009, 08:17 AM
hence the reason f22 is used for best overall balance with 5x4 lenses.


I have heard and read this reference, many times. However...

having studied their websites, and all of the hard copy documentation I have from Rodenstock, Schneider and Nikon... I can't seem to find any statements (or data) indicating that f22 should be used for best overall balance with 5x4 lenses.

Schneider and Rodenstock do publish MTF data for their lenses, which clearly shows that f22 may in fact NOT be the best performing aperture for many of their lenses. Additionally, resolution testing of LF lenses (multiple samples) by Chris Perez and Kerry Thallman would also bare this out... see http://www.hevanet.com/cperez/testing.html.

I wonder where this bit of photolore originated?

Ian Grant
06-17-2009, 10:13 AM
Try looking at the Doctor Optic website for the LF Tessar's, you will have to use the Internet archives, (my saved copy is in the UK), f22 is Zeiss/Doctor Optics recommended aperture for optimal performance. I originally found the link to the Doctor Optics site via Perez/Thalman a few yeras ago. Other manufacturers (Schneider/Rodenstok) give much of their data on coverage etc at f22, this is f-stop they general chose for measuring movements.

One problem is that lens tests like Chris Perez/Kerry Thalman are very subjective and made from images of test charts and while very useful they are based on small samples of each lens - usually just one, although there are exceptions ie 90mm Angulons etc. Few LF lenses are flat field lenses so shouldn't necessarily give high resolution results of a test chart at all apertures, but you would expect process lenses like G-Claron's or Artar's to possibly perform better under these circumstances.

Another issue not covered by Perez/Thalman or MFT testing is the actual use of a lens in real life situations, particularly when Depth of Field and/or use of movements may be critical. The general rule of thumb with most lenses from top manufacturers has always been that the smallest marked aperture is weak, but that one or two stops up will give excellent performance. But like all rules some lenses will break them. One thing the Perez/Thalman chart highlights is that most lenses perform well at f16/f22 some are actually far better at f22, others may have better overall sharpness at f22 but better central sharpness at f16 or even f11.

In reality we are both nit-picking because it's how well a lens performs in actual use, as I said earlier I've never noticed or been able to spot any drop in performance of my LF lenses at f22 or even f32 and f45 with some of them. My Tessars & Xenar very definitely only become sharp across the full field at f22, and this as I said before is exactly what Zeiss have always said, and Chris Perez confirms.

I really don't know where the f22 figure comes from for other manufacturers/lenses but it's common across them all for 5x4 format lenses, and in general use by photographers around the world.

Interestingly someone once did a breakdown of the Apertures used by John Sexton, and no-one can doubt the quality of his images, the figures were:

For example, I looked at the technical data for the 49 photographs in John Sexton's Listen to the trees, done with a 4x5. The breakdown was as follows:

f16 5%
f22 30%
f32 50%
f45 20%
f64 5%

Those figures can't be quite right as they add up to 110% but they do give a fair indication that f32 is the most common aperture he's using followed by f22 and f45. I'd also guess that they aren't far off other peoples use as well.


08-24-2009, 10:17 AM
Sorry to open up this thread again, but I am seeking some insight.

I have been trying to mount a Super Angulon 90mm f/5.6 on gen. 1 Gaoersi 617. I can't get enough distance to get infinity and a reasonable close focus distance at the same time. I even taken to putting a small spacer (3mm) between the flange and helecoid to give me a little more distance.

Unfortunately, the rear element of the SA bumps into the inside of the helicoid as it is extended, further reducing the close focus distance. Even with the spacer, the close focus distance is about 2.5m.

To top it off, the helicoid focusing mount only has about a half-turn of thread connecting it to the adaptor. While Gaoersi claims compatability with the f/5.6 SA, do I need a consider getting a smaller lens with a shorter flange to film distance?

Any suggestions will be appreciated.

Cheers, Blair

Ian Grant
08-24-2009, 10:50 AM
Send Gaoersi an email, they are very helpful.

I used my f6.8 Grandagon on my Gaoersi with no problems, I have an f5.6 90mm SA but never tried it and when I bought the cones the company set them up for my specific lenses, which might be your problem.


08-24-2009, 11:23 AM
Ian, would you do me a favour and measure the depth of your adaptor/cone. Mine is 24mm deep and suspect that a slightly deeper one is required for the SA. It's likely the one I have will work with your Grandagon with its shorter flange to film distance. Thanks.

Cheers, Blair

Ian Grant
08-24-2009, 11:44 AM
24mm deep, and the mount flange appears to be 55mm maybe 55.5mm,

I have a feeling there was a good reason why I used my main 5x4 lens WA rather than the Super Angulon I bought second hand at the same time as I was buying the 6x17, there's a bell ringing in my head about the 90mm f5.6.


08-24-2009, 12:18 PM
Ian, thanks. It looks like there is a Grandagon in my future. I should look at this as a positive experience that fully justifies the purchase of another lens.

On an entriely different front, the Gaoersi finder is approximate at best, especially as I wear glasses. Does anyone have any suggestions on either improving this finder, using another brand, or making one's own, e.g., some sort of wire frame. Thanks.


08-25-2009, 05:45 PM
After a suggestion from a LFP member and some head scratching on my own, I made a gasket out of an old mouse pad and inserted it between the body and the cone. This gives me a 5mm of extension, which provides both infinity focusing and a sub 2m close focus. The rear element of the SA still bumps into the inside of the helecoid, but helecoid is now securely screwed into the cone. There goes my opportunity to buy a new lens. :(

I also found that a 21mm finder for a 35mm rangefinder is a very close match for the angle of view for a 90mm @ 6x17. I need to figure out how to mount it to the Gaoersi and my vertical masks need to be reconsidered (post-its at the moment).


08-26-2009, 10:46 AM
Large format lenses are normally designed and optimised for use at f22, they are not diffraction limited to f16,

Diffraction is physics. No one can design it to be better or worse.