Tilt and your tech cam - how often with short lenses?

[jlm]

here is the way I see it:

at all distances less than infinity, you are using something to judge focus: either the ground glass, the distance scale or best of all, your tethered laptop, and something to set the lens to sensor distance, usually a helical focus thread. All of these methods will have tolerance errors far in excess of .02mm. If you are satisfied with the tethered laptop as a judge of focus, none of those error tolerances will matter.

Infinity is another story. On the one hand, you can evaluate focus at infinity using the laptop, but if your helical threads and the infinity stop won't let the lens travel far enough, you simply won't get focus. This travel limit is what Alpa is adjusting by back shimming and no more, and they assume that there is sufficient uniformity between lenses that if one lens works, they all will

not sure what you are meaning here, in reference to my quote:

"Unlikely. The very nature of a lens being calibrated is that the infinity marking is calibrated since that is the reference mark for calibration. Thus, then consider that the turn of helical focus is large compared to the focus travel of the helical and that is why I believe what you stated is not correct."

 

[Anders_HK]

I see parts of what you mean now. Certainly tethering will be most precise to determine if acceptable focus since it is same as will be the capture. All others include tolerances, however if they are suffice small (summed up) and within what is critical to achieve focus, then the image should be reliably and consistently sharp.

What I meant in what you quoted of me was that I thus believe the distance markings may well be within acceptable tolerance, since focus travel are along a much finer scale than the distance markings. However I do not see why the shimming would be overridden or why when using gg that shimming is not a factor. The gg should adhere to the tight tolerances of the Alpa system, and specifically the shimming is bringing the digital back to within those tolerances of the Alpa system per my understanding.

 

[cly]

Thanks a lot for all your replies to my original question and for your comments, very much appreciated!

It seems to me that the topic is moving away from tilt and towards the merits or problems of two different focusing systems. So I thought, it's time to explain why I started this thread - as the issue of focus is what I am still thinking about now.

When I posted my tilt questions, I made a stupid mistake (which was pointed out to me in a personal conversation): I was wondering if would miss the possibility of tilt with short lenses and I was thinking about the floor/ground. I didn't think about the simple fact that there is also a ceiling which I want to have in focus, and that there is something between floor and ceiling ... So I have come to the conclusion that I won't miss anything, in my particular situation.

Camera-wise the background of my question was Alpa. I had tested a Cambo WRS which, in a way, was a pleasant surprise. It seems to me that it is a remarkable camera but: it is plagued by serious flare problems. To me this is a show-stopper. At the same time I had an Arca RM3d for a brief test. It's an interesting camera but I am not convinced. Yes, focusing using a disto and a small table with the helicoid settings simply works. But I'm pretty sure that I would never be able to memorize the tables of three different lenses, so I would have to look up the correct value. Moreover, I had the impression that, at least for my way of working, the camera isn't optimal: for example, why on earth is it impossible to check the spirit levels if the camera is mounted at eye-level? In a way, this is a trivial problem but to me it is extremely important. It's definitely a nice camera but I wouldn't call it well thought-out. Enter Alpa. I don't like their marketing speech (and they should hire someone for proof reading before talking about all these things at the top …), I don't like the way they write about Arca on their web site, I don't believe in what they say about precision when it comes to a sliding back. But: So far I am really impressed by the Max. And that's the reason why I asked about tilt with short lenses.

I have no doubt that, from a theoretical point of view, the Arca focusing approach is superior to everything else I know of. But I am not convinced that I won't be able to get the type of images I want using an Alpa with one of the new focusing rings. Obviously, this is something I have to make sure before making a decision. But at the time being I am gravitating towards Alpa. It seems to me that, for what I want to do with a tech cam, the Max is a better thought-out and more flexible tool.

Chris

 

[David Klepacki]

Chris, You started this thread by referencing Jack's image using tilt /swing. If this kind of focus control is needed, then there is really no other way to achieve this without such movements.

For landscape and architectural shots, I often rely on tilt. For product shots, I rely more on swing. It is all about how you wish to "lead the eye" into your image.

Maybe you should first consider which lenses are more attractive to you. If you prefer the Schneider Digitars, these lenses need to be stopped down a little more for their optimum aperture (around F11). However, the Rodenstock HR lenses are diffraction limited designs and will give you highest contrast at wider apertures, but possibly at the cost of more distortion than the Schneider equivalent lens. If you prefer the Rodenstock HR wide lenses, then I would recommend a camera that offers some tilt/swing (Sinar, Cambo, Arca, etc.) in order to manipulate the plane of focus with the idea of using their wider optimum apertures (F5.6 - F8). A benefit of being able to shoot at wider apertures is not only for lower light situations, but also to allow lower ISO with faster shutter speeds for less noise and less possible "movement" in the image that may cause blur (e.g., foliage in light wind).

If these things are not an issue for you, and you typically work at/near F11, then the Rodenstock lenses may not be as good a choice as their rendering will soften (although still very good) due to diffraction when stopped down that much. In this case, tilt/swing is less of a necessity in terms of being able to work at the optimum lens aperture, but it may be desirable for compositional purposes (as in the example from Jack).

 

[dchew]

I think jlm makes a great point here. I recently rented a Cambo, P40+ with 35, 55 and 90mm lenses basically to see if I liked the overall tech-camera approach, which I did very much.

However, I had a very difficult time focusing, so I did the usual flat-wall test with focus brackets. I was 15 feet away from the wall, and everything finally came in focus when the lens was at infinity! This was the case with all three lenses.

By my rough measurements the back was about 10 thousandths of an inch off, which is huge. Furthermore, since all three lenses exhibited the same phenomenon, the tolerances were off somewhere between the camera, back adapter, back and sensor, not the lens mount forward.

There was just no way to get anything in focus at much beyond 30 feet. I'm not slamming Cambo in any way. This was a rental; who knows what it had been through.

I think every tech camera including Cambo has some method to adjust from the camera to the back. In the case of Cambo you need to send it back for calibration, where they remove the small painted-in bolts and presumably add or subtract shims (but do I have to send them my back??). Alpa and Arca give the user the ability to do it. Alpa's solution can handle out-of-plane adjustments (but what if I rotate the back?). Arca's can handle tolerance issues from the lens mount forward.

Dave

...Infinity is another story. On the one hand, you can evaluate focus at infinity using the laptop, but if your helical threads and the infinity stop won't let the lens travel far enough, you simply won't get focus. This travel limit is what Alpa is adjusting by back shimming and no more, and they assume that there is sufficient uniformity between lenses that if one lens works, they all will

 

[Anders_HK]

Alpa and Arca give the user the ability to do it. Alpa's solution can handle out-of-plane adjustments (but what if I rotate the back?).

Assuming the Alpa system do have the required tight tolerances, then it should work fine. That seems as the very key point with that system compared to the others. The shimming merely shims the back to near perfect to their system and which already should have the tolerances required. Thus their system should be within acceptable tolerances no matter if back is rotated. Question is then how Leaf's rotating sensors will work in relation to the very tight tolerances, but I gather one could shim such back for sensor in one position and hope/try to see how well it works also in the other.

For Arca if a sensor is too much out of tolerance then there seems there is no way to adjust parallelism, only longitudinal focus per say. For Cambo there was a post in LuLa of custom shimming a Cambo RS http://www.luminous-landscape.com/forum/index.php?topic=49977.0. However, assuming no shims are in there from factory it appers that one can only add shims, and which will need to be balanced by adjusting helical focus (three screws I believe on Schneider).

Assuming Alpa is within the alleged tolerances, it appears to me as a more simple system for a user in order to achieve optimum sharpness in images. The downside of Alpa is that they do not inform clear what applies to their system as far as how tolerances and more work. Indeed in marketing they appear somewhat arrogant, and by email tend to many times not give fact replies to queries. Those comments do not apply to Optechsdigital who gives truly excellent, credible and brilliant clear tech replies.

Regards
Anders

 

[Steve Hendrix]

Chris -

I agree, for interiors, use of tilt may have some drawbacks (ceilings, for example, as you referenced). That said, recently we had a group testing the Arca Swiss RL3D with the 43mm Digitar (and P65+). Without remembering all the exact specifics, what we found was that just adding a few degrees of tilt produced a significant range of focus from just a few feet in front of the camera to infinity. We did notice that focus of an overhanging ledge roughly 20 - 30 feet away, and about 15 - 20 feet in the air was also achieved. So, for landscape, we can certainly see an application for tilt, even on wide lenses.

As far as where this thread has veered in terms of cameras and what they can do with focus and tolerances and precision, etc, etc. I would say that first off, when it comes to the sensor, it is certainly possible to have a digital back adjusted at the factory if the tolerances don't seem to be met. It's great that Alpa also offers a solution for adjusting the image plane of the digital back. On the other hand, this does also assume that all lenses are at the same tolerances. In this case, the Arca Swiss R series offers a precise and easy to use method for adjusting focus for any lens or any back with any lens, regardless of whether all are in harmonic convergence or not.

I don't know that there is a right or wrong way, a better or worse way. I believe this is a very subjective topic. I also believe that there is a lot of misinformation about both products. In any case, one's preferences may come down to very individualistic terms about the cameras completely isolated from how they handle the lack of tolerances from any of the components.

The last car purchase I made came down to several tilting (pun intended) factors - my choice had a straight on, line of sight key entry to the ignition, the open trunk button was in an easy to access location right at my left elbow, I liked the color of the inside lights at night, and I loved the steering wheel, the way my hands fit it. I think it drove accurately, but that was kind of a given. Even if it didn't, I felt I had ample tools to adjust. I see these products very similarly. Incidentally, any manufacturer claims of superior this or that mostly fell on my deaf ears. My advice is to never buy into that, instead prove them yourself and focus on the features that matter to you.


Steve Hendrix

 

[gmerrell]

I don't mean to get off the "tilt" subject but---
I own a Cambo with the Schneider 47xl.
I have not owned any other tech cams so I can't compare the quality with other cameras. I am a machinist by trade. I own an aviation machine shop and I think I know a little about precision.

The Cambo in my opinion is a very well made and precision piece of equipment. It has an adjustment method that I think works better than shims. I don't think they want non technical people
to fool with it. I adjusted my Cambo/P45 using the Alpa video instructions, only using the Cambo adjustment screw instead of shims. It worked great.

I think they would do well to teach people to use their adjustments.
They are taking an undeserved beating being quiet.

Greg

 

[GrahamWelland]

Graham - I think that is a little overly dramatic about how you use the Arca. Most people only have a couple of lenses and any adjustment factor is really quite small and simple to remember. In addition with the wider lenses for landscape you are working with a pretty short range of the focus helical.

Terry,

Yes, you're probably right. I'm just going by the procedure as I understand it and projecting on to the typical scenarios that I find myself in. I agree with John about the guaranteed precision of the Arca approach in so far as it factors in both back and lens corrections but I also know that at sunrise when I can hardly even see the settings on my camera, let alone look up tables for distances, that things get difficult. Heck, I have to walk around the front of my camera to see what the exposure is set to, look down from above to see what aperture and distance I've set and then finally fire off the shutter. For me, that's enough complexity and i confess that in many situations I'll take the easy path of using my favorite apertures, focus stacking against the marked distances and then letting the back tell me whether I need to change the exposure which I do by feel normally. These days I don't bother with the distometer but actually use my M9 to focus and then use the distance on the lens as my distance guide for the foreground. Doubtless not entirely accurate but since I find that I focus stack more often that not now it's good enough.

Assuming Alpa is within the alleged tolerances, it appears to me as a more simple system for a user in order to achieve optimum sharpness in images. The downside of Alpa is that they do not inform clear what applies to their system as far as how tolerances and more work. Indeed in marketing they appear somewhat arrogant, and by email tend to many times not give fact replies to queries. Those comments do not apply to Optechsdigital who gives truly excellent, credible and brilliant clear tech replies.

Regards
Anders

Anders,

I agree - I like my Alpa but they certainly do come over as contrite and arrogant on their site. I also agree with you about Paul & Ted at Optechs - very good folks to deal with and voices of reason even to Alpa's hype.

 

[GrahamWelland]

For landscape and architectural shots, I often rely on tilt. For product shots, I rely more on swing. It is all about how you wish to "lead the eye" into your image.

David,

I absolutely agree with you on this. Even with 35mm digital I was a regular user of T/S at 24/45/85mm and it's one thing that I wish Alpa had a better solution for. For still life & macro I'm sure that the T/S solution works well (I don't shoot them much generally, and not at all with the Alpa). What I do shoot regularly though is landscape and its here that the other tech systems such as the Arca & Cambo do seem to offer better solutions for wider glass. As I mentioned a while back, I miss this and would definitely use it if I had it. I'm sure that most landscape shooters of the Phase/Mamiya DSLR would do so also if they could too!

 

[cng]

i should also add that the alpa concept shimming of the back does not carry over for multiple lenses. Wouldn't it have made more sense to shim each lens?

Hi everyone, first post after a short while lurking (sounds so sordid!). Continuing the OT discussion on camera systems/tolerances etc, I came to the same conclusion as Jim while recently evaluating tech cameras for architectural work. i.e. Shimming works optimally only for the lens that you conducted the shimming process with.

I've decided on the Cambo RS for anything needing movements, but have on my future wishlist an Alpa TC/SW with one lens. Both are beautiful systems IMO.

The marketing hyperbole notwithstanding, I think all these camera systems are gorgeous and made well, but with individual quirks that give them their distinctive personalities.

Stating the obvious: Every component in a system (lens, body, back) has a tolerance and these add up (average out?) when the components are assembled together. The big problems arise when something is out of tolerance, e.g. a dud lens or misaligned sensor.

The quest for technical perfection is admirable but I would argue that it is non-attainable and self-defeating. An example: The marketing tells us and testing shows that shimming even 0.01mm provides noticeable improvement in performance. However, lenses are made to +/-0.02mm tolerance. Say you shim your body/back with a lens that is +0.02mm (within tolerance) but you have another lens in your kit that is -0.02mm (also within tolerance). That means when you mount your second lens you have effectively introduced a net 0.04mm intolerance into your system.

I would argue that maybe we should accept that our gear is being made as good as possible given current technology and accept a certain "averaging" of errors/tolerances, apart from the manifestly dud lenses etc.

I am not criticizing any particular camera system or philosophy, but what frustrates me is the black & white marketing that glosses over this reality of tolerances and compounding errors that will add up in any system.

Actually, on second thought I WILL provide some constructive criticism to Cambo etc relating to comments made by others above: Alpa needs a good copywriter to remove from their website all the sarcastic comments regarding their competitors' products, plus maybe tone down the arrogance a bit. It's not professional and not attractive. Seems almost insecure. Cambo and Arca need to market themselves more aggressively, maybe even turn up the bling a little bit. Then again, it seems these companies are showing the face they want to show, doing things the way they want to do it, and are doing quite fine without me thank you very much. :salute:

So ... I know that was a bit heavy for a first post, but I didn't want to start with the old favorite "what camera should I buy if I have x dollars". :toocool:

 

[Terry]

CNG - welcome out from the lurking shadows!

 

[cng]

CNG - welcome out from the lurking shadows!

Thanks. And yes, I saw the Dante quote.

 

[dick]

I have not read every word of every post in this topic - but am I allowed to post anyway?

For a tech camera, it is also absolute important that the lenses be mounted and adjusted for the lens plane to be perfect parallel to the sensor plane (or back mount), and within tolerance to the acceptable tolerance to achieve focus
Regards
Anders

Why?

I appreciate that you have to work to high tolerances when the light path is different for focusing and taking the picture - as in an SLR.

Also you need accuracy if you use a ground glass focusing system.

But, if you use a tech camera without an infinity stop, and you are going to tilt and focus with live view, why do you need accuracy?

The one situation that I can imagine that would need some degree of accuracy is when pre-setting movements at ground level before cranking my tripod up to ten meters. (Where I would check focus with live view).

 

[jlm]

"But, if you use a tech camera without an infinity stop, and you are going to tilt and focus with live view, why do you need accuracy?"

agreed, especially with tilts, but even with no tllts i think the conceptual problem is that most seem to want to set their lens on the infinity stop and have objects at 1/4 mile be in best possible focus. After that, they would assume the lens barrel distance markings (and helical fine adjustments) are accurate enough for intermediate distances, also assuming they can figure a way to measure those distances.

I know what i am going to do when my cambo shows up:

pop on the back I will be using now with no shimming. fit the 43mm lens, set to the infinity stop and take shots to check focus, figure out how to adjust the lens flange so infiinity sharpness is achieved.

when the 70mm lens shows up, repeat above.

when the IQ back shows up, then shim the new back, if required, by checking images ,using the 43 as it is more sensitive.

 

[David Duffin]

Graham - I think that is a little overly dramatic about how you use the Arca. Most people only have a couple of lenses and any adjustment factor is really quite small and simple to remember. In addition with the wider lenses for landscape you are working with a pretty short range of the focus helical.

Terry, have you calibrated your Arca Rm3d + whatever back you use with more than one lens? If so I'd be curious to know how consistent your correction factors turned out to be... (thanks, hope I'm not trespassing too much)...

 

[Terry]

Terry, have you calibrated your Arca Rm3d + whatever back you use with more than one lens? If so I'd be curious to know how consistent your correction factors turned out to be... (thanks, hope I'm not trespassing too much)...

I've mainly used the 35XL with the Rm3d and I was not having focus issues. This shot was with about 1 degree of tilt and you can see I'm good close in and if I were to post crops you would see I'm good at the tree tops (and confirmed with focus mask in C1).

My 90 came much later and between my travel schedule and just messy weather every weekend I have not done any major testing/calibration on the lens and also haven't been unhappy yet. With a better LCD and focus mask on the new IQ backs I'm also not getting too uptight about it.

 

[David Duffin]

Terry, your 90mm must have been perspiring a bit! Love the composition!

Tomorrow if it's clear here I will be checking a new 150 and a 72 that just arrived ... will see if both are close to the +2 arcafactor applied with 47mm..

(A little off-topic, sorry)