View Full Version : When to switch to a macro lens?

William Mortensen
15-Jan-2006, 01:21
I've been searching this forum and elsewhere, hoping to find a particular little nugget of info, but without any luck. So my question...

At what distance or magnification one should switch from a "normal" lens to a lens designed for more close-up work? How close will most "normal" lenses focus and still render an acceptably sharp subject? (I know this will vary a bit from one lens design to another, but is there any general "rule of thumb?") As usual, thanks muchly!

(And let's see if I've gotten my name back to normal after my high school students switched it...)

Louie Powell
15-Jan-2006, 06:13
Mark -

In my experience, most normal lenses will focus down to perhaps a couple of feet. This distance is a variable and is one of the things that lens makers can change to increase the marketability of their product.

Your question specifically addressed the issue of macro lenses. Macro lenses are designed for close focus.

The close focusing limit of a lens is determined by the way the focusing helix is designed. If the helix is designed to allow the lens to move further away from the camera body, then the lens will focus on subjects closer to the film plane. A helpful rule to keep in mind is that to be able to focus on subjects with a 1:1 reproduction ratio (the film image is the same size as the actual subject), you must extend the lens by a distance equal to the focal length of the lens.

But there are a number of tricks you can use to extend the close focusing capability of ordinary lenses in order to avoid the expense of purchasing special macro lenses. The simplest and least expensive trick is to use extension tubes. These are metal rings that fit between the camera body and the lens. They typically come in sets of three to allow you several steps of adjustment - and when combined with the inherent focusing capability of the lens, give quite a lot of flexibility.

Another option is a macro bellows. These give infinite flexibility in adjustment, and are especially helpful if you want to focus on very small subjects.

I have a teleconverter that has a removable cell element - to allow the metal shell to also serve as an extension tube.

One of the considerations of close focus tricks is that the camera gets close to the subject. The problem with this is that the camera can actually start interfering with the subject (this is probably a corollary of the Heisenberg principle). Experienced macro photographers tend to use longer lenses so that they have more working space. I use a 100mm lens that was originally intended as a portrait lens for macro work. Of course, using a longer lens means that longer extensions are required to achieve the same degree of magnification.

While ordinary lenses will focus on close subjects using these techniques, there will be optical issues. It's possible to improve on their performance by reversing the lens - literally attaching the front of lens to the camera (and extension device) and shooting through the back end of the lens. You can purchase reversing rings for this purpose.

Finally, another trick that is sometimes helpful is to combine a couple of lenses - for example, use a reversed wide angle lens on front of a short telephoto lens. This requires double-male adaptors to screw the two lenses together.

Whenever you reverse a lens, you need to be careful to protect the rear element from damage. You can make a lens shade for a reversed lens from a rear lens cap and a short length of cardboard tube.

There are a couple of very helpful reference books that talk a lot about various macro tricks. One is Close Up Photography by Alfred Blaker. I suspect its now out of print but you may be able to find it in a library. The other is John Shaw's Closeups in Nature.

Dan Fromm
15-Jan-2006, 07:12
Mark, it depends on the lens and I'm not aware of any fixed rules.

You asked about close focusing distances rather than magnification so you may be thinking of smaller formats. The rule of thumb for smaller formats is not to use a "non-macro" lens off its focusing mount. "Macro" lenses such as my MicroNikkors are designed to go to1:1 on their mounts or on their mounts plus an extension tube that's more or less specific to the lens. Close focusing distance depends on magnification and the lens' focal length, so it makes little sense to talk about it alone. Another rule of thumb is to use a macro lens for magnifications higher than 1:10.

Since you're posting here, I take it that you're thinking LF. There are even fewer rules for LF, and relatively fewer lenses designated "macro" for 4x5 and larger than for smaller formats.

Louise was on the right track in suggesting you read a book, wrong about the books she suggested.

I don't think Al Blaker wrote a book called Close Up Photography. IMO, the best of his books for learning about photography is Field Photography. It has an extensive discussion of close-up work that is best on working no higher than 1:1.

Lester Lefkowitz' book The Manual Of Closeup Photography isn't as good as Field Photography on photography in general -- that's not its subject -- but is stronger on working above 1:1. If it isn't clear, I have both, see them as complements.

I have a beat-up copy of the John Shaw book Louise mentioned that I bought used. It is essentially Field Photography's section on working close up with prettier pictures, poorer explanations, errors, and a lot of padding. I wasted the $5 I spent on it. Shaw is too Nikon-centric for words.

Field Photography and The Manual Of Closeup Photography are neither brand- nor format-specific, and Blaker and Lefkowitz are much better teachers than Shaw. I've given copies of both, also of Kodak Publications N-12A and N-12B (Closeup Photography and Photomacrography, respectively, both written by H. Lou Gibson), to friends.

She's rarely mentioned on US BBSs, but Heather Angel wrote a nice book on closeup photography whose title I can't remember. I don't own a copy, probably should, but when I had the chance I read it. Head and shoulders above Shaw.

All of these books are out of print, can be obtained used through, in alphabetical order, www.abebooks.com, www.addall.com, www.amazon.com, and turn up from time to time on eBay.

Good luck, have fun,

Ted Harris
15-Jan-2006, 07:25
You should find several threads in the past year here on exactly this subject and,if you search the archives you will find an excellent lengthy article on the subject by Ernest Purdam on the main part of the site. The issue is what, exactly, you mean by “close-up work.” “Normal” lenses, those designed for optimal performance at ranges of infinity to 1:10 or 1:5 also work fine as the reproduction ratio gets larger, as it approaches 1:1. Where the performance of these lenses begins to fall off is at and after you hit the actual 1:1 point. Macro lenses OTOH, those designed to perform best at ratios of 1:4 to 4:1 and larger, begin to shine when you reach the actual 1:1 ratio. This is not to say that you cannot use your standard plasmat designs in the 120mm to 210mm range for actual macro photography. You can, but there will be obvious differences in performance between the results obtained with these lenses and those obtained with lenses designed specifically for macro work. Basically, the central portion of the image produced by the ‘normal’ lens and the macro lens will be the same at 1:1 and larger but you will immediately notice significant difference in as you move toward the outer edges of the frame Of course if your subject only occupies a small central portion of the frame this won’t matter but then what is sense of moving in close? You don’t need to get deeply into the technical aspects of lens design to see the differences. For example, both Schneider’s and Rodenstock’s lens specifications show that the macro offerings have a much larger image circle than the ‘normal’ offerings.

Last year I did a real world test to see how these performance specifications translated into actual images. I tested 3 180mm lenses, the Apo Sironar N, Apo Macro Sironar and the Macro Symmar HM. My subject was a flat mesh sterling silver necklace about a 25mm across and 500+mm long. The necklace was made up of thousands of tiny silver links. I piled the necklace on the light table so that the lenses would be challenged by both layers of links and strands of necklace the covered the frame. I then racked out the bellows to close to 1:1 but not quite, and made an image with each of the lenses. I used Polaroid T55 exposing for the negative. Looking at the negatives side-by-side on the light table or scanned images across the 23” Cinema Display screen there just were no visible differences. I then moved the bellows further along the rail to just beyond 1:1 for all three lenses and again produced three images. The differences were immediately apparent. Again, in the center of the frames the images looked the same but as you moved out to the corners of the frame there were immediate differences in both the resolution and sharpness of the edges and centers of the tiny links. The two macro lenses clearly outperformed the ‘normal’ lens.

If you do a lot of tabletop work that demands edge-to-edge sharpness of objects photographed at and beyond 1:1 then a macro lens is a reasonable investment. If you do not, then perhaps it is not. The modern macro lenses are heavier and larger than their ‘normal’ counterparts and they are considerably more expensive. Nikon, Rodenstock and Schneider all have offerings.

Bob Salomon
15-Jan-2006, 09:00
At 1:3 to 1:5 the macro starts to outperform the standard design optics.

15-Jan-2006, 09:45
Or, get a G-Claron and use it for both.

15-Jan-2006, 09:47
There is a section in View Camera Technique by Leslie Stroebel that shows a pretty good comparison between macro and non-macro at a given magnification. I don't have the book here or I'd be more specific.

At larger magnifications you can also get away with using an enlarger lens without worry of vignetting.

Ted Harris
15-Jan-2006, 10:28
CXC is correct if you are photographing flat objects but a G Claron IMO, or any of the other process lenses, gives you no particular advantage on small three dimensional objects.

Richard Ide
15-Jan-2006, 10:39
Good afternoon all (and others)

Would I be correct in assuming that a process lens would have performance similar to a macro lens? Or are there differences which I am not aware of?

Thanks Richard

Mike Cockerham
15-Jan-2006, 10:51
Macro lens are corrected to a higher degree than non macro lens for flatness of field that is why macro lens on 35mm, med format and repro are used for copy work. On 3d objects with both macro and non macro stopped down the same amount there is very little if any difference.

Dan Fromm
15-Jan-2006, 12:04
Ted Harris wrote " “Normal” lenses, those designed for optimal performance at ranges of infinity to 1:10 or 1:5 also work fine as the reproduction ratio gets larger, as it approaches 1:1. Where the performance of these lenses begins to fall off is at and after you hit the actual 1:1 point."

Ted, the 50/1.4 and 105/2.5 Nikkors I used to have think you're badly mistaken. Both were terrible off their focusing mounts. The 105 was so awful that it couldn't be focused wide open at 1:1. It really depends on the lens. I know they're not LF lenses.



Bob Salomon
15-Jan-2006, 12:18
"Would I be correct in assuming that a process lens would have performance similar to a macro lens?"

No process lenses are designed to photograph 2 dimensional objects at close range at f22 (4x5 work)

Macro lenses are designed to photograph 3 dimensional objects like products or jewelery in Ted's example, at close range at various apertures. For 3 dimensional objects the macro lenses will easily outperform all others from about 1:3 or 1:5 to 3:1 or 5:1.

William Mortensen
15-Jan-2006, 12:24
Thanks for the replies! Still trying to sort it all out... I tried a slightly more than 1:1 (just larger than lifesize) with my 215mm f4.8 Caltar-S two weeks ago, and found it lacking resolution even at f/45. (But it was a twelve-minute exposure, vibration is also a possibility.) I have two 150mm process lenses (Eskofot Ultragon and Hexanon GR-II), but given their maximum f/9 aperture, the bellows extension, the critical focus at close distances, and my preference to work in softer, lower light, I don't like using them in the field. Same would go with the G-Claron.

Ted Harris' post lets me think I'd be safe approaching, but not quite hitting, 1:1. I may run tests with the 215mm Caltar-S, and my 12" Caltar II-N, Dagor, and Astrogon (Tessar). Should any of these have an advantage nearing 1:1?

Also, I have a 67mm Schneider auxilliary close-up lens, (a +3 diopter, I think). Would it be effective (or heresy!) to use this on a normal lf lens for 1:1 or more?

William Mortensen
15-Jan-2006, 12:39
Question to Bob Salomon, (while we've got you here!):

"process lenses are designed to photograph 2 dimensional objects at close range at f22 (4x5 work)"

So why do Apo-Ronars and other process lenses close down to f/256? It seems these would be the *last* lenses to need such a small stop, needing little depth of field for a flat subject. Especially considering the loss of resolution due to diffraction at that f/256. But it must be there for something... What?

Bob Salomon
15-Jan-2006, 14:07
Process lenses close down and open up to apertures that they are not recommended to be used at just as other taking lenses do because the apertures in shutters and NF mounts can be set at other apertures. How would you feel if the lens had a moveable aperture scale and it was only marked for one aperture? You can see how this would work with an Imagon. It comes in a shutter with no marked apertures since you set the aperture with the disks.

Process lenses and taking lenses may be in mounts with marked apertures but some of those apertures put you into diffraction. In the case of most process lens used on 45 the design aperture is f22.

Bob Salomon
15-Jan-2006, 14:09
"So why do Apo-Ronars and other process lenses close down to f/256?"

This is akin to your car speedometer going to 180 MPH while the car can't go that fast and the speed limit is 65 or lower in most places. Just because the lens is marked f256 doesn't mean that it is a useful aperture or that you get the best results there.

Armin Seeholzer
15-Jan-2006, 14:15
Bob still lives in the past with his own 3D problem. Today is every lens designed to get a flat field otherwise it would be a very worse one.
Today Linus gives you a special service for your APO Ronar if you like it. They optimise it for infinity wrote mister Dieter Wenzel from Rodenstock Germany.
But then I wrote to Rodenstock and asked then for which magnification the APO Ronars have been optimised which Rodenstock sold in a shutter and this is the asnwer I got:

"Im Gegensatz zu den früheren Apo-Ronaren für die Reprografie, welche für Maßstab 1:1 optimiert wurden,
Sind die Apo-Ronare in Verschluß auf einen Abbildungsmaßstab von ca. 1:20 optimiert.
Es ergibt keinen Sinn Ihre Apo-Ronare in Verschluß nachträglich auf unendlich zu optimieren, da
sich gegenüber dem Optimum bei 1:20 kein sichtbarer Vorteil ergibt.

Viele Grüße
Philipp Gschwendtner Linos Company "

So they stated that the APO Ronars which where used for Reprography, had been optimissed for 1:1, but the APO Ronars with a shutter have been optimised for 1:20. And he told also it makes no sence the shutterversion to optimese for infinity because it will not make a difference from 1:20 to infinity!
Every lens has the same problem to put a 3D picture into a 2D picture and every modern lens has to be a flatfield lens today!


Dan Fromm
15-Jan-2006, 14:35
Mark, thanks very much for the additional information about what you want to accomplish.

At 1:1, nominal f/45 is effective f/90. At f/90, you can't get much better than 17 lp/mm. And, as you pointed out, with a 12 minute exposure poor motion control -- camera and subject -- can kill you. If one of your feet has been shot, you pulled the trigger.

Consider whether you can achieve the results you want with flash, and remember that you're allowed to use more than one. It is often useful to use one or two flashes for the main subject and another one for the background.

Since you're not a beginner, get the Lefkowitz book I suggested and read it. Buy the Gibson Photomacrography book too. You need a good fright; he delivers one in the course of a long discussion on what can't be accomplished.

In my experience with GRIIs -- I routinely shoot a 210/9, had a 150/9 and sold it -- they're very very good close-up, even wide open. But you must understand that at 1:1, f/9 allows you roughly 2 mm DoF. f/22, IMO the smallest aperture usable at 1:1, gives about 5 mm.


Struan Gray
15-Jan-2006, 14:41
The very small apertures on process lenses were used when making half-toned B+W reproductions or colour seperations using a ruled-line halftone screen. In this case every dot on the page was a tiny copy of the lens aperture, and for very fine-ruled screens you need a very small aperture opening.

William Mortensen
15-Jan-2006, 18:44
Struan- The half-tone application makes sense! I knew there had to be some practical application for that f/256. (And at 1:1, the functional f/stop is f/512...)

"Buy the Gibson Photomacrography book too. You need a good fright; he delivers one in the course of a long discussion on what can't be accomplished."

Dan- I think I'm getting frightened enough to just stay on this side of 1:1! Most of the true macro lenses seem to be f/9, which gets pretty dark for my style of shooting. I guess experience will tell me just how much I can push my lenses...

Ted Harris
15-Jan-2006, 18:51
Mark, I think you are still confusing process lenses with macro lenses. All of the modern macro lenses are f5.6 lenses.

John O'Connell
15-Jan-2006, 19:06
I haven't found f/5.6 lenses to be noticeably better on the groundglass than f/9 lenses. For macro work in LF, effective apertures are so small even for focusing that the solution isn't a brighter lens but a small, bright flashlight.

15-Jan-2006, 19:39
Bob S.

If one wants to reproduce a flat field to film (another flat field), you are suggesting a process lens, vs. a macro lens, is this correct? I assume the difference is, the flat field lens is designed to have the plane of sharp focus reach into the corners of the subject, which is unlike the plane of sharp focus for a normal lens, which the plane of sharp focus is of equal distance from the lens, forming a plane of sharp focus in a semi circle, with the center of the lens being one end of the radius, the other end of the radius forming the semi circle. Is this correct?

If so, what Rodenstock lenses are designed specifically for flat field task?

Also, I want to do flat field work, such as shooting a flat 24x30" subject to MF film, 1:12 reduction. This is obviously not macro or high magnfication, but rather similar to normal photography, however, the subject is very close (similar to macro) and the desire is to have the Plane of sharp focus be flat at the subject. What lens does this best, and how does one choose the best fl for this type of work?


William Mortensen
15-Jan-2006, 20:58
"Mark, I think you are still confusing process lenses with macro lenses. All of the modern macro lenses are f5.6 lenses."

Right you are, Ted! For the little use it would probably see, I'd rather spend less for a process lens that could pass as a macro, but that's not being realistic, just cheap. Thanks for keeping my terminology straight before I confused someone else.

Michael S. Briggs
16-Jan-2006, 01:20
As JohnArs said, all modern LF lenses are designed to be flat field. This shouldn't be a problem at taking aperture.

I am having difficulty making sense of Bob's recommendation that Process lenses are intended for 2D subjects and Macro lenses for 3D subjects. The sense I can make out of this is that most (but not all) LF Process lenses are relative narrow field, while all of the current LF Macro lenses (plus the just discontinued AM-ED Nikkor) have wide coverage. For a 2D subject the photographer is likely to center the camera on the subject and not to use any movements. For a 3D subject, they are likely to want to use tilts and swings to place the plane of best focus. So for 3D subjects, a lens with extra coverage (i.e., a Macro lens) will tend to be more useful than for 2D subjects. But the Macro lens will work very well for the 2D subject, it will just have excess coverage. And the Process lens will work for 3D subjects if you don't need movements, or use a long enough focal length, or work at a sufficiently high magnification.

Ted pointed out the other obvious difference between LF Process and Macro lenses: most of the Process lenses are f9, while the LF Macros are f5.6.

As to the original question about when one should switch from a normal taking lens to a macro lens, assuming that you have both, probably around 1:5. What's confusing is that the answer is fuzzy -- there isn't a sharp boundary where one lens type becomes clearly superior -- there is a large overlap. What matters to the corrections of a lens is the angles that the light rays make in traveling from subject to lens to image. There is little difference in the angles from 1:infinity to 1:20. As you focus from 1:10 to 1:5 to 1:1 the angles start changing rapidly.

My experiment with this question compared a Fuji-W and a G-Claron doing a closeup of a flat subject. The resulting negatives were virtually identical. But wide-open and near wide-open, the Fuji-W could only be focused in the very center of the image because the rest of the image was too fuzzy.

Struan Gray
16-Jan-2006, 01:33
I would love to hear a lens designer's take on the design tradeoffs and differences between a LF macro lens optimised for 1:3 and process lenses and enlarging lenses optimised for the same ratio.

I read Bob's comment about 2D and 3D as being back-to-front. The point about the '2D' lenses like APO-Ronars and G-Clarons is that they had to have very low distortion for accurate reproduction of things like technical drawings and maps. They also had to have true apochromatic performace for making colour seperations. That means that the design will have been optimised for low distortion and near-identical performance at the center wavelengths of the standard colour seperation filters, which leaves less wiggle room for improving other aspects of the lens' performance. If you allow a little distortion, or if you improve the perforamce across the whole visible spectrum at the expense of loosing true apochromaticity, you will probably end up with a better lens for more general subjects, or one which works better at wider apertures.

Bob Salomon
16-Jan-2006, 04:32
"If so, what Rodenstock lenses are designed specifically for flat field task?"

Apo Ronars but they are out of production.

Michael S. Briggs
17-Jan-2006, 01:19
Struan, I think almost all of the LF macro and process lenses are symmetrical, and thus optimized for 1:1. From the symmetry, they will have zero distortion for 1:1 and tend to have low distortion. Schneider's datasheet for the Macro-Symmar shows distortion below 0.5%, which is very low, albeit somewhat worse then what an old datasheet shows for the 240 mm Apo-Ronar -- a price of the wider field.

The only process or macro lens that I can think of that isn't symmetrical is the Rodenstock Apo-Macro-Sironar, which is "biased" to have a intended use range of 1:5 to 2:1.

At this point the nomenclature process lens vs macro may be more of historical interest than useful in selecting a lens. I suggest looking at coverage and maximum aperture (if that is important to you). There are some wide-coverage process lenses available in shutter (e.g., G-Claron, Fuji-A).

Mark: of the lenses that you list, for working at 1:1, I'd try the G-Claron, Eskofot Ultragon, and Hexanon GR-II. At least the G-Claron (and probably the others) are symmetical and will do very well for 1:1. If focusingis difficult, try bringing on more light (perhaps a flashlight). Is your darkcloth fully opaque? As Dan pointed out, unless depth of field requires f45, it is best to avoid stopping down so far.

Dan Fromm
17-Jan-2006, 05:28
Michael, don't forget about Apo Tessars, TTH tessar type process and copying lenses, tessar type Apo Nikkors, Apo Germinars, various tessar type process lenses from the FSU, ...

I'm not sure whether Apo Skopars, which, name notwithstanding, are 5/3 heliar types, are quite symmetrical. Boyer has asserted that Apo Saphirs, also 5/3 heliar types, are symmetrical. That claim notwithstanding, their diaphragms have to be on one side or the other of the central singlet; I have two, one with the diaphragm in front, the other behind, the central singlet. Go figure ...



Ted Harris
17-Jan-2006, 06:52
Dan, I just went back over some of the earlier answers in this thread and noted yoru comments on your 35mm Nikkors. My comments here have been addressed specifically at LF lenses and their design. I also took a quick romp through Lefkowitz' book and am not sure it is that good a reference for someone interested in LF macro photography as it seems almost totally aimed at 35mm macro work and, while it is good information for that purpose I am not so sure for LF work.

Dan Fromm
17-Jan-2006, 08:41
Sorry, Ted, I responded to what you wrote, not what you meant. I'll try to read your mind better.

About Lefkowitz, he's aimed more at working above 1:1 than at a particular format. The problems don't vary with format, only with magnification.



Arne Croell
17-Jan-2006, 08:51
"I'm not sure whether Apo Skopars, which, name notwithstanding, are 5/3 heliar types, are quite symmetrical. "

Dan, they are not completely symmetric, even discounting the aperture placement. If one unscrews and checks the focal lengths of the cemented doublets at each end of the Apo-Skopar, they are not equal. They are both positive and not worlds apart in focal length, but different enough that you do not need an optical bench to see it.

ronald moravec
22-Jun-2006, 08:46
Do the shots with your regular lens stopped to 22. If you are happy, do nothing.

If the corners are too soft in flat copy or if the image is generally not sharp enough, you will need a lens like the G Claron.

With three dimensional subjects, a standard lens should be adequate. If you need to reproduce flat copy or smaller than 1:1, a macro probably is needed.