Arca Helical Calculator

[jagsiva]

I have been finding both the Arca lens charts and the eModule far too course in their distance scales for use especially with longer lenses like the SK60XL or SK120ASPH.

With the 120 for example, a foot at near distance can make a difference of several major scale markings on the Arca helical. So the look up charts and the eMC are not fully utilizing the refinement of the Arca helical as far as I can tell.

So I built a calculator in Excel. It is currently only in feet and has coverage for 32HR, 40HR, 35XL, 43XL, 60XL and 120ASPH. You put in your CoC and offset for the back. These numbers do not change from lens to lens. The you input your distance in feet. The app returns helical values for all the supported lenses and shows the color of the helical revolution as well (White, Red, Blue, Green, Yellow).

The UI is not the greatest, and I as soon as I get some time, I will create a version in "Numbers" so it can be run on an iPhone.

I had to post here as a zip file, so it will need to be uncompressed. You will also have to enable macros when you open it.

Appreciate any thoughts and how it can be improved.

This free, but please don't make it better and sell it back to me

Cheers...

 

[dougpeterson]

We have a similar in-house tool we use to create charts at smaller increments for clients that wish them, or wish to have cards listing multiple specific lenses on one card, or wish to have a PDF perfectly fitting their phone screen-of-choice.

Did you account for transition between thin vs thick lens equations (i.e. does it match the Arca provided cards at all distances) or did you use a single regression? We found that a tough nut to crack. If you don't get that right then either the far or near distances will be significantly off.

 

[Paul2660]

Hate to ask a stupid question.

How do you determine you CoC and back offset?

As far as I know I have no other offset than the mounting (Mamiya) and the thickness of the rm3di.

Thanks for doing this, iPhone version would be a great asset.

Paul

 

[jagsiva]

We have a similar in-house tool we use to create charts at smaller increments for clients that wish them, or wish to have cards listing multiple specific lenses on one card, or wish to have a PDF perfectly fitting their phone screen-of-choice.

Did you account for transition between thin vs thick lens equations (i.e. does it match the Arca provided cards at all distances) or did you use a single regression? We found that a tough nut to crack. If you don't get that right then either the far or near distances will be significantly off.

Doug,

It is based on all the Arca distance points given in their charts. Following this, I used multiple regression to generate the formulas, the formulaic results are very close to the arcs chart values, BUT, the arca chart values are rounded off, so there may be some very small differences. For the most part, the curve fit is in the 99.x% range. So far, with my testing, the numbers have been bang on.

My frustration with the cards is that they are too course. Same for the eMC. For example, with the 120SK, value at 10ft is 12.4-green, and 9ft is 21.8-green. This is a massive amount of focus range in within a very narrow DoF. Now, this is at a foot interval, and neither the charts nor the eMC go to this level. I really think we need to be at inches at close distances.

So far, my results have been good.

 

[jagsiva]

Hate to ask a stupid question.

How do you determine you CoC and back offset?

As far as I know I have no other offset than the mounting (Mamiya) and the thickness of the rm3di.

Thanks for doing this, iPhone version would be a great asset.

Paul

Paul,

CoC is specific to the back. In my case, the IQ180 is 5 microns or 0.005mm. I believe the IQ260/160 are 6 microns. All the pixel pitch sizes are available on the Phase site for all their backs. The main info you are getting here is the acceptable DoF and HF distance. As the pixel size drops, the acceptable area of focus is reduced.

The Offset is an Arca thing. This is the baseline calibration value you dial in or add to any standardized helical setting. This is very similar to the Alpa shims. Given that there are small variations in backs, plates, and bodies, this allows you to calibrate infinity focus to your specific kit. Again, you just do this once, and the same value holds true for all lenses.

I just realized the Office mobile is now available for IOS. I will get cracking on an iPhone version. I think I will also have the "actual" Arca charts in there as well so they can compared as a sanity check.

 

[dougpeterson]

My frustration with the cards is that they are too course. Same for the eMC. For example, with the 120SK, value at 10ft is 12.4-green, and 9ft is 21.8-green. This is a massive amount of focus range in within a very narrow DoF. Now, this is at a foot interval, and neither the charts nor the eMC go to this level. I really think we need to be at inches at close distances.

Just imagine using one of the "other" tech camera options with standard helicals where a typical interval might be [5 ft] [10ft] [25 ft] [infinity]!

Still, your frustration is why we have our in house tool to generate more finely stepped charts. Since the helical itself is so fine grain it screams out for a very fine grain chart.

 

[Paul2660]

What is needed, ,, CMOS full frame that is tech camera friendly with Live View. At a reasonable price.

Paul

 

[jagsiva]

Why not just have an "electronic" ground glass? CMOS based EVF or rear LCD. Sits on the same plane as the DB. Cant be too expensive to make, just take a look at any sub $1000 1.6x crop camera out there. Certainly should be cheaper than the optical VF from Arca or Alpa!

This way, I can still have my CCD

 

[alajuela]

Hi

Wish Cambo with come out with "Alpa" style rings for their lens with tilt shift.

My project for the weekend is setting up two tripods, distro and focus aid (lens align) and a tape measure on my floor, tethered to get at least the missing numbers on my focus ring for my 90 and 120.

I did the 40 and 60 which came out good. I just hope my little stickers do not come off. It is a bit tedious -- good job for a rainy day.

And............

I also have a question - Is there a chart which would have variables of focal length and distance (and I guess sensor size although I sorta think that does not matter if we correspond in 35mm terms), that shows when the image goes off axis.

For example if you were shooting a honeycomb, the dead center (on axis) would be two dimensional. As you move away from dead center the sides of the honeycomb start to appear, the further you move away towards the outside of the image circle the more the sides appear.

If interested I can explain why I am asking - :chug:

Thank you for the indulgence.

Phil

 

[jlm]

if i recall, alpa publishes it's focus distance charts for a variety of lenses, showing focal distance for every couple degrees of barrel rotation. they hold the rotation increment constant and determine the actual focal distance. you could plot this, make a best fit curve, and interpolate any distances you want from the curve. my guess is they derive the distances empirically, and to be sure, it is lens specific.

 

[f8orbust]

A bit OT, but because of the presence of the bayer array, other than in an achromatic back, a scanning back or on a foveonesque-style sensor, it's better to use 2 x pixel pitch as your CoC in any DoF calculations as this will roughly correspond to the maximum detail extractable.

 

[dougpeterson]

Hi

Wish Cambo with come out with "Alpa" style rings for their lens with tilt shift.

My project for the weekend is setting up two tripods, distro and focus aid (lens align) and a tape measurez on my floor, tethered to get at least the missing numbers on my focus ring for my 90 and 120.

I did the 40 and 60 which came out good. I just hope my little stickers do not come off. It is a bit tedious -- good job for a rainy day.

And............

I also have a question - Is there a chart which would have variables of focal length and distance (and I guess sensor size although I sorta think that does not matter if we correspond in 35mm terms), that shows when the image goes off axis.

For example if you were shooting a honeycomb, the dead center (on axis) would be two dimensional. As you move away from dead center the sides of the honeycomb start to appear, the further you move away towards the outside of the image circle the more the sides appear.

If interested I can explain why I am asking - :chug:

Thank you for the indulgence.

Phil

You're thinking in terms of lens length, which will confuse the issue. Instead just think about [distance-to-subject] vs distance between [subject-center] and [subject-edge]. Then it's a simple triangle, and you can use an arc tan to find the angle to the edge, because you know the distance of the opposite side ([subject-center] to [subject-edge]) and adjacent side ([distance to subject]).

For instance if the object is 10cm across (i.e. extends 5cm to the left of the center and 5cm to the right of the center) and you are 5cm away and you choose a lens that fills the frame then the angle to the edge of the object is 45 degrees. If, with the same subject you are 500cm away and fill the frame then the angle will be 0.6 degrees.

To determine what the highest angle you can accept for your project is something I think only you can tell us from practical experiments. But if you determine that angle you can easily replicate it with objects of different sizes by maintaining that same angle.

This is true even if you don't fill the frame. But if you take a wide angle lens and step back 50 meters then the subject will only be a few pixels across .

If your not a math guy, just click this link and change the distances to match your scenario.

 

[alajuela]

You're thinking in terms of lens length, which will confuse the issue. Instead just think about [distance-to-subject] vs distance between [subject-center] and [subject-edge]. Then it's a simple triangle, and you can use an arc tan to find the angle to the edge, because you know the distance of the opposite side ([subject-center] to [subject-edge]) and adjacent side ([distance to subject]).

For instance if the object is 10cm across (i.e. extends 5cm to the left of the center and 5cm to the right of the center) and you are 5cm away and you choose a lens that fills the frame then the angle to the edge of the object is 45 degrees. If, with the same subject you are 500cm away and fill the frame then the angle will be 0.6 degrees.


To determine what the highest angle you can accept for your project is something I think only you can tell us from practical experiments. But if you determine that angle you can easily replicate it with objects of different sizes by maintaining that same angle.

This is true even if you don't fill the frame. But if you take a wide angle lens and step back 50 meters then the subject will only be a few pixels across .

If your not a math guy, just click this link and change the distances to match your scenario.

Thank you Doug

Makes sense, I will think about this as I go thru my day, to form a mental image. Then see if this works for me.

I need to visualize how much the image changes from two dimensional (dead center / on axis) to "perspective" as you move away from the center to the sides of the image (circle).

Here is a hint "Todd Webb" his street shot from your neighborhood, I really would not call it a typical panorama ( I have shot many). I have tried this type of shot before and believe me alot of variables and set up. Makes a regular pano a "walk in the park"

Thanks again

Phil

 

[jagsiva]

...so....the iPhone version of the calculator is almost ready. It will be based on Numbers (think free in App Store). I decided to do it this way and not as a stand alone app to get away from the crazy rules Apple has for distributing an app via the Appstore. Of course, there really is no other easy way of getting and executable into an IOS device.

Just doing some final testing. Have tested the 23/32/40HR. Will not be able to test 43/35XL. Just in the process of testing the SK120ASPH. This is taking a little longer as the baseline chart numbers can be improved upon. So I am actuality taking specific measurements. The lens also has a significant amount of focus shift at near, mid, and far subject distances. So I am pretty sure, different formulas for the different ranges will give more accurate results. Just testing this now. Hope to get it posted for the weekend.

The distance entry fields may look a little goofy, but I wanted a way to enter them without using a keyboard, yet still have a way to enter numbers at very fine intervals.

You can also keep your disto distance fixed and allow the calculator to account for rear tubes, Rotamount etc. Before I release it on the weekend, I'll also add a field for disto off-set so you can keep the disto on the tripod somewhere and just dial in a distance.

Sorry for the teaser, but thought I'd give an update.

 

[stngoldberg]

Thanks for this

Stanley

 

[Paul2660]

+1

Paul