ALPA and depth of field

[mtomalty]

Christian,

I copy the following from Joe Cornish's website. Nowhere does he mention resolving power,diffraction
or,even, circle of confusion 
Personally,if I were you, my first step would be to buy the 24 and 45 tilt shifts for your 5d Mkll.
When used in a 'slow and contemplative' style, using left/right or top/bottom stitching,and
paying attention to post processing the files are extremely capable.

"I am now actively working to that end in four digital formats. They include a digital compact (Panasonic Lumix LX-3); so-called 4/3rds (Olympus E3); 'FX' or full frame 35mm digital (Nikon D-700); and finally, medium format digital (Phase One P-45+). The resolution of the first three formats is really very similar, around 10 or 12 megapixels, while medium format is in a different league of 39mp. The Phase One is usually deployed on the amazing Linhof Techno field camera, helping me preserve my link with large format discipline. Why so many formats? Because the variety is stimulating, and each camera has its own particular strengths (and weaknesses). Who knows, I may eventually concentrate on medium format exclusively; it is the natural successor to LF film. Yet as I write this I am enjoying the challenges that variety offers, and still feel in the early days of my photographic 're-learning curve'.

[Graham Welland]

And one small note, depth of focus is a product of the f-number, not the focal length. If you have to shim your wides, you need to shim your normals and telephotos if you are using them at the same apertures. If shimming is required at all. And this always confused me, since you shim the back, what happens when the tolerances in the lens mount requires different shims?

Well, the fact of the matter is that you DO need to have your lenses in exact calibration on their mounts. If they are off then shimming the back alone won't be sufficient. I had to have one of my Alpa mount lenses calibrated by my dealer because it was out of spec from the other three that all had perfect registration with my shimmed back.

Some vendors claim that shimming isn't' necessary because all lenses are calibrated at the factory, as are the back mounts. Well maybe they have better QC than anyone else because many of us know from bitter experience that even Leica can't manage that! Arca do have a calibration scheme that can accommodate both variance in the sensor position/back flange and also lens tolerances since each combination has it's own offset when setting focus - you just need to keep track of them although that doesn't seem so hard if you only have a few lenses which let's face it most tech shooters typically do.

[theguywitha645d]

Well, the fact of the matter is that you DO need to have your lenses in exact calibration on their mounts.

What do you mean by "exact"? Even Alpa has manufacturing tolerances and so perfection is impossible. And how do you focus? Live view? So if the image is sharp on live view, what difference is a bit of shimming going to make. Are you using a ground glass? Then the only thing that matters is the difference between the GG and sensor plane. Are you using the focus scale and small apertures, then the discussion of accuracy vs. precision is valid. And absolute precision is impossible and not required for any imaging device that uses a CoC, especially with objects with depth--if you photograph a landscape at f/11, is it going to make a difference if the focus is at 60ft or 61ft?

Cameras are precision instruments, but to say you need perfection is a mistake, especially since it cannot be achieved nor is it required. Do you shim your Sinar? (I believe you use a Hy/6, if I got the name right) Have you had focusing issues?--I believe you cannot shim that camera. I manual focus my Pentax 645D with manual-focus lenses and that has a viewing system that has to be more precise than a simple design such as an Alpa and I get sharp images all the time. I have no problem focusing my Phase back on a Linhof 679c with a sliding back and GG.

I am curious what Alpa thinks the CoC of the IQ180 is? We can then calculated the depth of focus and compare it to the shims it offers.

Now, I think Alpa cameras are great. I am not trying to knock the product--they are well worth buying and using. I am simply questioning their claim about shimming and its practicality--it sort of falls into recent discussion over diffraction at 100%. Not that MFD shooters suffer from OCD or anything...

[Graham Welland]

Different Graham - I shoot with an IQ160 & P25+ with Alpa STC & Phase One DF.

Remember that focusing and shooting with a technical camera is completely different to using a DSLR like the 645D. There is no direct view in practice.

In answer to your question, you need to have a fixed reference point on your lens if you want to use the marked focus distances on the helicoid. That reference point is as close to infinity as is practical. I use the HPF rings which do provide very granular distances that you can use along with a distometer close up or a rangefinder at further distances. This means that if something measures 5ft away then I can dial in 5ft on the focus ring and I know that it will focus there and not at 4.9ft or 5.1ft - at these distances the accuracy is critical. Now when you get to 60ft vs 61ft,  sure its not so much of a big deal so long as DoF covers it although if that's your primary focus subject and you want it spot on then you want to know that when you measure 60ft and dial that in, it really is tack sharp at that distance.

Focusing with a ground glass is impractical IMHO with digital backs vs film but it is an option - I personally find it almost impossible with a loupe on a 6x6 GG. I just don't bother with that.

Live view - hmm, well that's a work in progress in my view with the IQ. Yes, when it is clear enough and you've got everything set up correctly then yes you can absolutely dial in the critical focus on screen and all of the shimming & mount calibration is irrelevant. However, we're not there yet ... at least not outdoors in varied lighting conditions.

What I can tell you is that when everything is dialed in correctly and shimmed, when I set focus to infinity I get sharp at infinity. From there every other focus mark is correct on the ring. Now personally I do a fair amount of focus stacks that include infinity and so I want that sharp as one of my reference images. I'll then zone in from there for an image stack.

You could certainly argue that if you calibrate your own hyper-focal settings manually using the procedure I linked to in an earlier post, whether the marked infinity really is accurate or not for each lens/aperture probably doesn't matter so much either. However, that's no different than knowing your lens front or back focus and applying your own fudge factor every time you shoot (or more likely forgetting or not getting it right and getting softer shots than you expect).

[TH_Alpa]

And as we are talking about the overemphasis on certain technical matters, Apla does have a long history of overemphasizing precision in their marketing--hey, it helps sell cameras and more power to them. Personally, their marketing turns me off. Alpas are fine cameras and I would certainly recommend them. But the physics of photography is really not that hard to figure out. Shimming has really become overemphasized.

And one small note, depth of focus is a product of the f-number, not the focal length. If you have to shim your wides, you need to shim your normals and telephotos if you are using them at the same apertures. If shimming is required at all. And this always confused me, since you shim the back, what happens when the tolerances in the lens mount requires different shims? And how does all of this work with temperature changes? Personally, Alpas are great cameras and I would not worry about the shim thing.

With all due respect, you are wrong when saying that shimming is overemphasized and not that important, let alone that precision of a system isn't either. I invite you to come and visit us in Zürich, and we will show you what it means in practical terms and that it is not a question of marketing.

Concerning your question with the lenses: it is expected that each lens is absolutely well adjusted to infinity, which can be done as well.

Best regards
Thierry

[TH_Alpa]

One would expect that this calculation has been made, and yes it has been. But it needs only to shoot a couple of images at "infinity", taking away a shim or adding one to see the difference.

On a side note: nobody claims that Alpa is the perfection, at least not me nor Graham or anybody at Alpa, but what we try to say is that with the current high-res backs and their pixel size, a camera system has to be as precise as possible, with the minimum possible tolerances. A system is only as good as the weakest part in the whole chain is: better to have all elements of this chain from lens to DB  as precise as possible. That is what Alpa is trying to achieve, by different means and techniques in the manufacturing of the cameras and their accessories.

Best regards
Thierry

I am curious what Alpa thinks the CoC of the IQ180 is? We can then calculated the depth of focus and compare it to the shims it offers.

[ErikKaffehr]

Hi,

Theguywiththe645 has a good point on thermal expansion. On a 5 cm long body of Aluminium it would be 23 micron for a 20 degree change. The focal length of the lens may change to. That is part of the reason many lenses can focus past infinity.

Best regards
Erik

With all due respect, you are wrong when saying that shimming is overemphasized and not that important, let alone that precision of a system isn't either. I invite you to come and visit us in Zürich, and we will show you what it means in practical terms and that it is not a question of marketing.

Concerning your question with the lenses: it is expected that each lens is absolutely well adjusted to infinity, which can be done as well.

Best regards
Thierry

[EricWHiss]

All I can say is you guys with Alpa's are lucky - in two ways - 1st because its a great camera, and 2nd because mostly you'll never work with macro stuff.  That's the area where I experience the biggest challenge with DOF and diffraction effects.  It's not my experience that the losses don't show in prints - I see them in even 20x24 prints with subject matter that has fine detail.  But there are software workarounds that help recover this detail ie richardson Lucy, so it's not untenable. 

[theguywitha645d]

I invite you to come and visit us in Zürich, and we will show you what it means in practical terms and that it is not a question of marketing.

Best regards
Thierry

Thierry, when I am lucky enough to visit, I would be delighted to take you up on your offer.

Graham, sorry for getting you and your camera confused. I have used technical camera, but since it was a film one, I did not want to mention it and open up a can of worms.

I guess what I may be reacting to is this monster that MFD is somehow extremely difficult, unless you set your camera on a steel pier with a shutter speed of 1/250 at f/11 and at the base ISO. You just have to wait for the sun to give enough light to take the image. I have not found for photographers that have mastered the photographic process, both professional and amateur, that it is really that difficult a leap. Perhaps LCCs and the effects of lens cast may be "hardest" thing about the process. But shooting MFD is no superhuman task, which I find it seems to be portrayed.

And there have always been different flavors of photographers. Some are very detail oriented and others are looser in their process and work. Both ends of the spectrum produce fine images--we can disagree on aesthetics, but we are in the realm for personal taste at that point. MFD is fine for all photographers and I would encourage an skilled photographer to pursue MFD if the interest is there and trust they will have the skills to get what they need. I have not found shooting MFD any different from any format I have shot--from 16mm to 8x10. And I have been printing much of my work all the way up to 4x12 feet.

There was a thread recently about using an Alpa for handheld portraiture in the field. Many folks really did not like the idea of that. Having done similar work, I thought it was quite a valid way of shooting. (I am sure I could do that with an Alpa off-the-shelf with no shimming ) Because of cost, MFD has gone where the money is, mostly commercial work which has a very specific approach to photography with a definite technical bent--I know I am painting with a broad brush. I guess I am an advocate of seeing MFD hit areas of photography that have been underrepresented. I think want may put some folks off is this perception that it is somehow different and really hard.

I will get off my soapbox.

Peace...

[Graham Welland]

No worries - hey, I actually agree with you about how easy it is in reality to shoot MF with technical cameras actually. Personally I almost always get better results when I use my Alpa vs my DF if only because it's a more considered shooting experience and not technically difficult, other than of course remembering to do everything in the right order. Ditto LCCs - simple stuff if you have a process that you follow. If I can do it, it must be easy.  

As regards handheld shooting & zone focus - yup, simple if you've got enough light. Shimming will still matter if you want to shoot at infinity though - but yes you have a very valid point.

[dchew]

I think shooting with a tech camera can be as simple or as difficult as you want it to be.

It's one of the things I like about it!

Really. No meter, no ai/oneshot/servo/manual focus mumbo, and barely a viewfinder (I lived for almost a year without one - still have to pick it up from my dealer). Just me and the outdoors!

Oh, and I never grab the zoom ring by mistake instead of the focus ring.

Dave

ps. If someone comes out with a zoom lens for a tech camera I will hang myself. And no the Hartblie doesn't count. 

[D Semick]

Hello All,

I have to admit, I was a little skeptical as to the importance of shimming a back. Having recently returned from visiting the folks over at ALPA, I can attest that the difference of .1 mm is clearly apparent. The ALPA system is really quite impressive. (yes, I do have some bias here being an ALPA dealer).

Having said that, I cut my teeth in photography with a 4x5 using tilts and swings for focus control. I'm very excited to see what ALPA has in the works regarding future T/S possibilities being incorporated into their wider lenses.


Denny

With all due respect, you are wrong when saying that shimming is overemphasized and not that important, let alone that precision of a system isn't either. I invite you to come and visit us in Zürich, and we will show you what it means in practical terms and that it is not a question of marketing.

Concerning your question with the lenses: it is expected that each lens is absolutely well adjusted to infinity, which can be done as well.

Best regards
Thierry

[TH_Alpa]

Hi Denny,

Welcome in the Alpa team. And thanks to confirm my statement.

Just to be clear myself as well:

I was myself as well much skeptical before seeing what shimming means, but I was well aware from my Sinar time and the experience with tousands of DBs in the market that sensor alignment was a serious issue.
However, it is most of the time difficult to see that an image is actually OOF or that the focus is slightly (or even much) at the wrong place, simply because one has no reference to compare. An image can seem to be absolutely perfectly focussed/sharp and though it isn't the best possible which can be reached.
By doing a test and by changing the position of the sensor, means the distance to the lens plane by shimming + or - by increments of mms, it clearly shows visually when comparing 2 images what e.g. 1/100 mm can change in terms of sharpness.

Best regards
Thierry

Hello All,

I have to admit, I was a little skeptical as to the importance of shimming a back. Having recently returned from visiting the folks over at ALPA, I can attest that the difference of .1 mm is clearly apparent. The ALPA system is really quite impressive. (yes, I do have some bias here being an ALPA dealer).

Having said that, I cut my teeth in photography with a 4x5 using tilts and swings for focus control. I'm very excited to see what ALPA has in the works regarding future T/S possibilities being incorporated into their wider lenses.


Denny

[ChristianRandwijk]

Hello again,
   thank you all for providing loads of information for me to chew on. And I'm very excited about a possible future t/s system for Alpa. I am guessing a good disposition for me now would be to find a 24mm TS-E v2 to try out the movements on my 5D2, and then see what the next year or so holds for me, budgetwise. So again, thank you all for the input.

[D Semick]

Ah..yes...decimal points matter.

That should have been .01 mm.

...the difference of .1 mm is clearly apparent.

[ErikKaffehr]

Hi Thierry,

Would yo be able to post a comparison image demonstrating the effect of a subject being 0.01 mm defocused (in the image plane)?

Not that I'm overly skeptical, but my understanding is that the "CoC" caused by defocus in the image plane is defocus divided by f-stop, so 0.01 mm of defocus at f/4 would give a circle of confusion of 2.5 micron, much smaller than the sensor pitch.

Best regards
Erik

Hi Denny,

Welcome in the Alpa team. And thanks to confirm my statement.

Just to be clear myself as well:

I was myself as well much skeptical before seeing what shimming means, but I was well aware from my Sinar time and the experience with tousands of DBs in the market that sensor alignment was a serious issue.
However, it is most of the time difficult to see that an image is actually OOF or that the focus is slightly (or even much) at the wrong place, simply because one has no reference to compare. An image can seem to be absolutely perfectly focussed/sharp and though it isn't the best possible which can be reached.
By doing a test and by changing the position of the sensor, means the distance to the lens plane by shimming + or - by increments of mms, it clearly shows visually when comparing 2 images what e.g. 1/100 mm can change in terms of sharpness.

Best regards
Thierry

[adammork]

This is from an old post of mine, from a topic that dealt with camera tolerances, it's a couple of years old but still valid IMO:

"To be honest I always thought that all the talk about 0.01 mm was to much marketing talk and I thought with a smile on the face, that the fussiness about so small tolerances had something to do with the water they drink in switzerland :-) - that was until I got a new back that had a bit different focus plane than the old, I took the adjustable Alpa V adapter and started to shim it - I don't smile anymore at the swiss, only at my own ignorance....... 0.1 mm (millimetre) change the focus a lot and yes you can just see a difference in 0.01 mm - it do not ruin the image at all, but there is a difference..... yes it's too scary"

/adam