90mm lenses and their need for a center filter.

[Wayne Crider]

I just passed up on a Caltar II N F6.8 because some archives I have read talked about it needing a center filter, and I'm not in a position to buy a filter in a ddition to the lens. I'm wondering if this is true and if 90 F8's are better in this respect. Also, since my needs aren't critical in terms of enlargements over 11x14, would a 100mm Wide Field Ektar be a good compromise for a 90, and save m e a couple of bucks?

[David Karp]

Wayne,

I have a Rodenstock Grandagon-N 90mm f/4.5. I don't use a center filter and have been happy with images taken in both black and white and transparency film. If you are shooting black and white, then you can always correct for any fall off by dodging and burning, but like I said, I have never had a problem, even when shooting transparency film on a clear day with an even blue sky.

[Ellis Vener]

Wayne,

The 90mm f/8 SW Nikkor supposedly has a very large (235mm) image circle @ f/22 and infinity focs, compared to the other f/6.8 & f/8 90mm lenses. There have been other posts on this. I use the 90mm f/4.5 Caltar II-N (AKA Rodenstock Grandagon) and for 4x5 usage i don't need to use the center filter unless I am using Velvia and using an extreme amount of rise and/or shift. For 6x17cm the CWF definitely goes on, but that is to cover an additional inch at both ends of the long dimension.

[bob moulton]

I echo the other posts. I have used the nikkor and a Shneider 90mm without finding a center filter a must. By the way in Large Format Landscape Photography Jack Dykinga, who does use center filters explains how one can consider the composition of the scene carefully when using a WA lens without center filter. Bob

[Bob Ashford]

In B/W I can usually do not need the centre filter with my Super Angulon 90mm f/8 but when using colour trannies I DO, especially when they are light coloured objects around the edge of the picture area such as light sand etc. So composition is very important.

Bob

[Mike Mahoney]

There have been quite a number of posts related to center filter requirements, and there is much valuable info in them. Opinions vary as may be expected, but a search of the archives should give you a fair sample.

You can measure the light falloff of your particular lens by shooting an evenly lit white wall at various apertures and measuring in PS. Then create a circular grad layer in PS to compensate.

This is fairly simple, and Photoshop Elements can do it, and it's an older version, but is not expensive, in fact about a third of the cost of most center filters. In fact, PS LE is a free bundle with many scanners and digital cameras.

[Mark Sampson]

I use a 90/8 Nikkor-SW for architecture and have never needed a center-filter.The 100mm WF-Ektar should be a fine lens (I love my 135mm version) but does not have the covering power of any of the modern 90s. So where a modern 90 might start to fall off a little, the Ektar would be cutting off entirely. WF-Ektar coverage= 80 degrees, modern 90= 100-105 degrees.

[Wayne Crider]

Well the info on the Caltar at Calumet advertises one for it, and from what I read in various threads, some use a filter and some don't. Maybe the Nikon is a better way to go instead.

[Navy Moose]

I just got a Caltar II-N 90mm f/6.8 lens and I had the very same questions concerning the need for the center filter. I paid $499 for this lens at KEH and the center filter at Calumet was more than what I paid for the lens.

I shoot black and white landscapes and from what I've read here, I won't be getting this center filter.

Yet another thread re-awakened from past the expiration date by me. :-D

[windpointphoto]

Why not call Calumet and see if they will rent you one. If they won't I have one I'll loan you. Use different films and then you'll know. Problem solved. Forever.

[IanG]

Never heard of anyone using a centre filter on a 90mm before, assuming that it's being used on a 5x4 camera. On a 5x7 then it's probably wise for critical E6 work.

I've been using the f6.8 Grandagon 90mm (same as the Caltar) for over 20 years and have never seen or had the slightest problem with fall off at the corners. Rodenstock don't recommend one for 5x4 work.

Ian

[Bob Salomon]

Never heard of anyone using a centre filter on a 90mm before, assuming that it's being used on a 5x4 camera. On a 5x7 then it's probably wise for critical E6 work.

I've been using the f6.8 Grandagon 90mm (same as the Caltar) for over 20 years and have never seen or had the slightest problem with fall off at the corners. Rodenstock don't recommend one for 5x4 work.

Ian

Yes Rodenstock does offer the CF for use on 45 with the 90mm if the photographers work and results require the filter.

Many users like the WA look of the fall off and live with it. Many users use enough movement that the fall off is greater in one area then another and require the filter. Many users simply over light the edges of their set or dodge and burn their prints to even out any fall off.

The use of the center filter is a personal choice. If one feels that they want more even results and want to reduce (the cf does not totally eliminate the fall off) then they will need to get one. If one feels that they would rather spend time correcting the fall off in the scene with PS if they scan the result then they can do that. If one wants to solve the problem at the camera then they need the filter. Heliopan also makes a CF that solves the problem and costs less then Rodenstock's. Like the Rodenstock they are just a neutral color wise (but all CF will have a cast with some subjects and lighting conditions). Just make sure that if you get one it is the proper density.

As for do you have fall off or not? Pick a white or gray wall that will fill the frame or point the camera up at an even sky and shoot some shots at optimal aperture and compare. If the fall off is there (and it is with all modern wa lenses from 35mm up) then that will show it to you. The fall off is there with Nikon as well as German lenses - Nikon just never sold or offered the cure to the effect.

[Navy Moose]

Hi Bob,

Thank you for the information. I will make my decision after I take some shots using the lens. Some of the stuff I've read said Rodenstock solved the problems which caused light fall off. I've been looking for documentation from Rodenstock concerning this lens. It is too bad PopPhoto or the other magazines don't review or test these lenses.

Navy Moose

[Leonard Evens]

I hate to (partially) disagree with the others, but there are certainly situations where it is helpful to use a center filter. I have the f/6.8 90 mm Grandagon N which, I think, is essentially the same lens as the Caltar. Like the others I got along fine without a center filter, but when I started doing architectural photography in color I did sometimes encounter a problem. There is definite fall-off as you move from the center of the field, particularly if you use extensive rise. In b/w this might not be too noticeable, but in color it can be associated with a color shift. (I use Portra VC 160.) On a building facade which is supposed to be white or gray stone or concrete, such a slight shift can be objectionable. Using a center filter doesn't eliminate the problem entirely, but it makes it more manageable. Actually, the effect came strongly to my attention with my f/4.5 75 mm Grandagon N, but it takes the same filter, so I also use it with the 90 mm lens, although the effect, while still apparent, is not as obvious.

[Bob Salomon]

Hi Bob,

Some of the stuff I've read said Rodenstock solved the problems which caused light fall off. I've been looking for documentation from Rodenstock concerning this lens.

Navy Moose

First, the statement about the problem had been solved was a very misleading statement in the older brochures from the early 80's and earlier that was made by the former Rodenstock distributor - Berkey Photo Marketing - who is long out of business. What they stated was that the Grandagon had unsurpassed evenness of illumination. This led many to assume that it solved the CF problem but all it really said was that no modern WA lens was more even in illumination. This, coupled with the fact that Berkey did not promote the availability of the CF compounded the problem.

Rodenstock does print info regarding illumination as well as center filters. We would be happy to mail it to you if you are in the USA. If you are not the local distributor will supply it to you.

[David Crossley]

I've used my Nikkor 90 f8 sw feeling comfortable not using the centre filter for up to 4x5 black and white films. However, the use of the appropriate centre filter (111b) is absolutely necessary for the 6x17 and 5x7 formats.


David Crossley/Crossley Photography....

[Alan Davenport]

Certain lens designs vary from the cos^4 light falloff. IINM, the Super Angulon design is closer to cos^3, while the Biogon is actually worse. Or maybe I have that backward? I'm sure someone will correct me if I'm wrong; it's happened before...

I have the 90/8 SA and I've never even considered that I might need a center filter with it. YMMV.

[Leonard Evens]

Certain lens designs vary from the cos^4 light falloff. IINM, the Super Angulon design is closer to cos^3, while the Biogon is actually worse. Or maybe I have that backward? I'm sure someone will correct me if I'm wrong; it's happened before...

I have the 90/8 SA and I've never even considered that I might need a center filter with it. YMMV.

If I am remembering correctly the Grandagons also have cosine cubed fall off. My f/6.8 90 mm Grandagon N has an image circle of about 220 mm. That corresponding to an angle of coverage of about 100 degrees. The cube of cosine(50) is about 0.25, which is four stops. So if you used maximum rise, you would have quite a bit of fall off at the edge of the field. On the other hand, if the image is centered, the drop off in the corner would just be about two and one quarter stops, and it would be significantly less for most parts of the scene.

So whether or not a center filter is helpful can depend on just what you are doing. In my case, I am taking pictures of architecture close up so I have to use maximum rise and the edges of the scene are important.

[Navy Moose]

If I am remembering correctly the Grandagons also have cosine cubed fall off. My f/6.8 90 mm Grandagon N has an image circle of about 220 mm. That corresponding to an angle of coverage of about 100 degrees. The cube of cosine(50) is about 0.25, which is four stops. So if you used maximum rise, you would have quite a bit of fall off at the edge of the field. On the other hand, if the image is centered, the drop off in the corner would just be about two and one quarter stops, and it would be significantly less for most parts of the scene.

So whether or not a center filter is helpful can depend on just what you are doing. In my case, I am taking pictures of architecture close up so I have to use maximum rise and the edges of the scene are important.
It was my understanding there would be no math in this debate. Can someone please explain what these calculations mean? I think I understand what the image circle is, this determines the size of the circle of light hitting the film plane. The cosine stuff just about made me cross eyed.

Thank you.

[Ron Marshall]

It was my understanding there would be no math in this debate. Can someone please explain what these calculations mean? I think I understand what the image circle is, this determines the size of the circle of light hitting the film plane. The cosine stuff just about made me cross eyed.

Thank you.

Image circle is the diameter of the circle of light that a lens will project. If the diagonal of the film is smaller than the image circle, the lens will cover that format. For 4x5, image circle must be at least 153 mm.

The image formed by a lens is brightest at the center of its image circle, and gets dimmer toward the edges. How quickly this falloff ocurs and how much depends on lens design. That is expressed in terms of cosine to the fourth power or cosine to the third power. I won't bother explaining the geometry, what is important is that some lens designs have more falloff than others.

[Navy Moose]

Hi Ron,

Thank you for clarifying the image circle for me. The math I've been seeing in my study of large format is pretty daunting, especially to math challenged people like me.

I'm familiar with the inverse square law. Are you talking about vignetting with the cosine ^ 3? I was digging through a reference and saw mentioned there is a calculation for that.

[Ron Marshall]

Hi Ron,

Thank you for clarifying the image circle for me. The math I've been seeing in my study of large format is pretty daunting, especially to math challenged people like me.

I'm familiar with the inverse square law. Are you talking about vignetting with the cosine ^ 3? I was digging through a reference and saw mentioned there is a calculation for that.

That's correct; cosine^3 or cosine^4.

The cosine is taken of the angle between a line on the axis of the lens (straight through) and the point on the image circle where you want to calculate the light intensity.

[Ben Chase]

I also have the Rodenstock Grandagon-N 90mm f/4.5 and shoot landscapes as well as interiors with both color transparency film and black and white.

I don't have the center filter for this lens and really don't see the need to get it, even with about 20 degrees of rise in some of the shots.

[Leonard Evens]

It was my understanding there would be no math in this debate. Can someone please explain what these calculations mean? I think I understand what the image circle is, this determines the size of the circle of light hitting the film plane. The cosine stuff just about made me cross eyed.

Thank you.

Sorry to confuse you. It might have been better just to give the results of the calculations: that even with the best design, if you use maximum rise, you may have as much as 4 stops drop off in light intensity from the center to the edges. But if your image is centered and you don't care that much what happens at the corners, you may not notice much difference.

You should take the trouble to understand what the image circle is. Ron gave a good explanation. The reason it is important is because different lenses of the same focal length can have different size image circles, and this can be important when deciding which lens to buy.

For example, the f/ 5.6 72 mm Super Angulon XL has an image circle of about 226 mm, where my 75 mm f/4.5 Grandagon N has one of only about 196 mm. That allows for considerable more movement, and in principle the Super Angulon would be a better choice where a larger amount of rise is necessary. I've encountered the limitations of my 75 mm Grandagon in this respect when trying to photograph some nearby Chicago skyscrapers from Millenium Park. Unfortunately the Super Angulon XL is much more expensive and much heavier, and also it wouldn't work with my camera.

It is important to distinguish the angle of coverage from the anglue of view. The former refers to the total image circle where the resolution is good enough for a decent picture. The latter refers just to that part subtended by the film frame.

[Nick_3536]

Image circle is the diameter of the circle of light that a lens will project.

That's circle of illumination.

The image circle is inner part of that circle that provides good image definition. Some lenses casts a large light circle with edges that aren't worth much. At least not if you're enlarging. Might be okay with a contact print.

[Navy Moose]

Sorry to confuse you. It might have been better just to give the results of the calculations: that even with the best design, if you use maximum rise, you may have as much as 4 stops drop off in light intensity from the center to the edges. But if your image is centered and you don't care that much what happens at the corners, you may not notice much difference.

You should take the trouble to understand what the image circle is. Ron gave a good explanation. The reason it is important is because different lenses of the same focal length can have different size image circles, and this can be important when deciding which lens to buy.

For example, the f/ 5.6 72 mm Super Angulon XL has an image circle of about 226 mm, where my 75 mm f/4.5 Grandagon N has one of only about 196 mm. That allows for considerable more movement, and in principle the Super Angulon would be a better choice where a larger amount of rise is necessary. I've encountered the limitations of my 75 mm Grandagon in this respect when trying to photograph some nearby Chicago skyscrapers from Millenium Park. Unfortunately the Super Angulon XL is much more expensive and much heavier, and also it wouldn't work with my camera.

It is important to distinguish the angle of coverage from the anglue of view. The former refers to the total image circle where the resolution is good enough for a decent picture. The latter refers just to that part subtended by the film frame.
Hi Leonard,

Thank you again. I am doing reading on the image circle. I bought the Caltar II-N 90mm because it had a decent image circle and was faster than the f/8 lens I was looking at.

I have Steve Simmons' book and Jack Dykinga's book as well that I'm reading.

Navy Moose