Lens testing. As easy as 1,2,3...or maybe not?

[f8orbust]

Revisiting Joesph Holmes' article on lens variability (my head still hurts) got me thinking.

I have neither a) his knowledge, b) his patience or c) a number of 'identical' samples of a lens to compare against one another....so, for a typical user, is there any strightforward set of steps that can be undertaken in order to check that a lens (and I'm thinking tech camera lenses here) performs as expected.

Basically: Does anyone use a specific workflow to test the lens, or is it shoot a brick wall at f8 and then pixel-peep 'til dawn?

Jim

 

[Jack]

I shoot images with any new gear and inspect them corner to corner looking for anomalies. If I don't find any, I am happy. If I find some, then it depends on how bad or what they are. I used to find a lot of issues with older lenses and earlier backs and earlier camera bodies, but with newer lenses, backs and bodies I generally find very few, or at least few that cause any concern.

As re the current crop of popular tech camera bodies, I am certain you will find more issues are related to user error or user misunderstanding than any substantive issues with the hardware itself. But of course, there are still occasional hardware issues. Lenses may be a different matter, but generally speaking the latest digital copies I've used over the past few years have all been very good to excellent...

Joe found many issues originally regarding back and body mis-alignments, and then optical anomalies like decentered elements or bad spots in the compound grinds on aspheric elements -- but he was testing mostly older glass and original bodies with the then new P45+ back which was pretty extreme resolution even by today's standards. The reality is these anomalies were always present in early digital and film days, it was just harder to spot them unless they were extreme. With today's higher resolution digital cameras and instant 100% view, it is a lot easier for everyone to be looking close, and hence much easier to spot. Moreover, manufacturers seem to be paying more attention to detail as well, and I suspect that is the reason why I have not had many issues with the latest builds of "digital" specific lenses.

 

[jlm]

spent some time fiddling with lenses this morning.

Specifically, I am using the H39 back, tethered to a macbook pro with external monitor, cambo WRS tech camera and testing the results at inf.

shooting out an open window at about 1000', maybe much longer in a few cases.

lenses are SK43,120 and R70.

basic idea is to peep a shot at high mag, then move away from the inf stop a bit and repeat, comparing results. basic settings are iso50, 1/500, f5.6

with these lenses, you can disengage the helicoid from the distance scale (and stop) by loosening three small screws. this allows moving past inf to be sure of that part of the range.

I almost went blind trying to peep this!

found the 70 and 43 were very close to correct (I might have moved the stop 1/32" on the ring, but it was a tough call to see the diff) the 120 was off by 1/8" on the ring; changing this made a significant difference, as previously, the inf stop was setting focus beyond inf.

basic problems encountered:
I am using Phocus (H back) and it takes three steps before the image settles into the final version: first pixellated, then very soft, then the final. I also think Phocus is not so hot for an accurate display.

it takes a bit of work to find part of the image that shows what you need to see. At first, i was using cyclone fences on far rooftops, satellite dishes, water tower lattices, etc. had better results using signage, as the comparisons were clearer.

it is difficult to see very slight focus changes at a very highly magnified part of the image; I set two images side by side in the viewer for direct comparison

I am not sure if Phocus is doing any sharpening to create that last image, i have the sharpening box un-checked

this was a preliminary run, as I am expecting to do it again when the IQ160 shows up (due in mid october now, h mount)

advice appreciaterd

 

[Shashin]

I am not impressed with the testing scheme or the authors reasoning. The OP is concerned about his lenses. I am wondering if the OP suspects his equipment is not working as well as he believes so that we can help solve a real problem.

Does that clarify my post?

 

[jlm]

i am concerned about focus in the context of back shimming. this is not a new issue; Alpa does the same using metal shims between the back and the body, a sort of one shimming suits all lenses approach. adjusting the lenses individually seems more useful, and as you can see, one lens was significantly different from the other two. Had i done this by back shimming, the results would have compromised at least one lens. That is the reasoning

I found the method hard to apply, but basically it is the same concept that Optechs describes for the Alpa shimming; I just found it visually challenging

 

[GrahamWelland]

There can be variability in the set up of tech cam lenses such that they may front or back focus slightly. Yes, even with these $$$ digital technical camera lenses!

If you are creating offset adjustments per lens (Arca) then you are compensating for this across your entire kit.

If you only have the ability to shim your digital back then you can't cater for any differences between all of your lenses. I had this with my SK 47XL digitar back focusing very slightly compared to my SK150/90HR-W on my calibrated Alpa. My dealer has a collimator and so was able to adjust the SK47 focus ring so that it nailed infinity at the infinity mark and thus was identical to my other lenses. I'm not sure I'd want to be trying that across several lenses using the manual process that is suggested here without a guaranteed consisted reference mark.

 

[Stuart Richardson]

Jim -- a large percentage of MFDB users are bald. It is not because they are older and more established because of the price of entry. It is because they pulled out all their hair trying to test their lenses and eke out the last micron of performance. There is a point of pride in doing this, but I would suggest if you want to keep your hair and your sanity, just shoot with your lenses and see if they give you results you are happy with. If they don't, then try another sample and see if that helps, or send the whole setup back to the manufacturer to be matched.

This was one of the things that appealed to me about the S2 (or the 645D for that matter) -- an integrated body is a bit less troublesome in terms of sensor placement than removable backs. The S2 bodies and lenses are tested in terms of their exact specs, and the body and the lens talk to each other so that the AF's plane of focus is exactly on the sensor plane regardless of the lens/body pairing. I am not sure if it is foolproof (probably not), or if other makers don't do something similar (they may well do it too), but damned if that isn't a lot easier than shimming. Some people are tinkerers and like getting this stuff all done perfectly themselves. Personally, I would rather not have to worry about it. So it is an integrated body for me...luckily I don't need a view camera or technical cam for the type of work I do...on the rare occasion I do need it, I can still shoot 4x5!

 

[GrahamWelland]

Hee hee, if you want to be a voyeur of equipment angst (and premature balding/aging) go visit a Leica M equipment forum sometime.

 

[Guy Mancuso]

Yea I got out fast enough to only get a receding hairline. LOL

 

[jlm]

i am a tinkerer, have most of my hair and way too much ability to tweak things. I see a lot of room for improvement with MF tech gear, but that doesn't stop me from shooting.

It is quite common in this forum to hear discussion of focusing difficulties, or camera shake, or the demands of high res backs, etc., so nothing new here. I'm not after that last micron of performance, but if my lenses aren't up to basic snuff, what point is there in the IQ160, for example?

Alpa shimming has been discussed many, many times, but this is the first time I have seen emphasis on lens variation. lots of praise for the SK 120, but if it can't focus at infinity, you will have soft shots, and i have. 1/8" out on the focus ring is quite a bit; trying to make that up by shimming the back would be challenging.

what i am trying to do with my Cambo kit is make it reliable, consistent and reasonably accurate so i can get what i think i am shooting. Shifting the focus ring actually is quite easy, easier than shimming, for sure. I am also modifying the Alpa HPF rings so they will work on my cambo mount 43 and 70. All Schneider and Rodenstock had to do was a more thorough job of engraving distances so you don't have to guess where 18' is on the scale. I don't think that it is asking too much to be able to measure the distance (distometer) and then set the focus (HPF rings) and then have the damned $5k lens actually be in focus

 

[GrahamWelland]

All Schneider and Rodenstock had to do was a more thorough job of engraving distances so you don't have to guess where 18' is on the scale. I don't think that it is asking too much to be able to measure the distance (distometer) and then set the focus (HPF rings) and then have the damned $5k lens actually be in focus

Certainly can't argue with that! There's a lot to be said for removing the variable doubt factor that you get when everything isn't perfectly in sync.

 

[jlm]

the irony is that the motivation for me in using a tech camera is to be in control of the shot; so i have tilts and swings and shifts...as close to a view camera as is practical. but then, one of the major attractions and funtionalities of the view camera, WYSIWIG on the ground glass, ain't there. So we have all these workarounds, like LCD screens, HPF rings, Distometers, peeping the shot, accessory viewfinders, focus mask, tethering, back/lens shimming, etc

 

[Stuart Richardson]

I should go on record too and say I think it makes perfect sense that you have shimming and other ways to control the plane of focus in tech cameras -- it would be a shame if they didn't. And John, you are of course luckier than most in that you have a machine shop and the skill to use it! I dealt a little bit with shimming for my Sinar back when I tried it with the V adapter, and it worked well. All I am saying is that I would rather not HAVE to do it. And with a tech camera, you do, since the manufacturer of the camera is not the same as the manufacturer of the lens, which is not the same as the manufacturer of the back -- generally 3 different companies, all with different specs and tolerances. If you want them to work properly you really need to do some precision matching! So for me, an integrated body works better since you relying on the factory to get it sorted all out -- with players like Leica, Hasselblad and Phase One, their quality control is generally high enough that you don't have to worry as much. Certainly you still need to check, but needing to send everything in to be calibrated is the exception rather than the rule.

The possibilities of tech cameras are great, but for me the fussiness in their use, the lens casts...it's just not worth it for my work. For others they are indispensable, certainly for architecture shooters.

On the more general basis of lens performance, I agree that if the lens is not up to snuff that is a huge problem for MFDB, but rather than assuming the glass is half empty and searching endlessly to prove your lens is not perfect, it's better to shoot and let the lens show you if it is not working correctly! Do some basic tests, sure...make sure it focuses correctly and is sharp across the frame, but the lengths some people go to to find minuscule errors sometimes misses the point, which is to take good images.

 

[Bill Caulfeild-Browne]

Well said, Stuart!
Bill

 

[vjbelle]

found the 70 and 43 were very close to correct (I might have moved the stop 1/32" on the ring, but it was a tough call to see the diff) the 120 was off by 1/8" on the ring; changing this made a significant difference, as previously, the inf stop was setting focus beyond inf.

I have the 150mm which also focuses about 1/8 inch beyond infinity. But this lens does not adjust as readily as the 100mm and lower as it uses a different helical - I believe a Linhof helical. The 120 is the same (I think). How were you able to adjust that lens?

Victor

 

[jlm]

the moving part has a knurled section (1) and a section with the markings and the stop (2). loosen three very small phillips screws in 2. set 2 on infinity stop, rotate 1 which will advance the helicoid until you have the focus you want, tighten screws

before starting, i put a pencil mark across 1 and 2 for a reference

 

[vjbelle]

the moving part has a knurled section (1) and a section with the markings and the stop (2). loosen three very small phillips screws in 2. set 2 on infinity stop, rotate 1 which will advance the helicoid until you have the focus you want, tighten screws

before starting, i put a pencil mark across 1 and 2 for a reference

You are right for all of the lenses below 150mm. I also thought that the 120mm had the same helical as the 150mm. I don't see the three screws on the 150mm. They are very visible on my other three lenses. My dealer is also unaware of an adjustment procedure for the 150mm. It would be great, although not critical, to set the infinity stop for infinity - not beyond infinity.

Victor

 

[Guy Mancuso]

Thinking of going out and checking infinity on my 35xl before I put the HPF ring on tomorrow. Probaly use live view and a lupe and see how it goes. I know I have a view out there for a 5 mile away mountain range. Hope that is far enough away

 

[jlm]

wasn't it Buzz Lightyear's mantra: "To infinity, and beyond!"

Guy i was using about 1000 yds and also checking at about 1/4 mile and a mile or greater. by compariosn, found that 1000yds was adequate with the 43, 70 and 120

radio towers get pretty small with a 35mm lens


side note: the 120SK lens requires a spacer, (between the body and the back) so there is another joint and another element that add to the tolerance buildup

 

[Guy Mancuso]

Great John great info for me to check today. Like to do it before putting my ring on the lens.