PDA

View Full Version : Picking ideal lens and fl, for flat copy work



bglick
6-Feb-2006, 16:04
I need a lens to perform flat copy work. I will photographing very sharp prints about 32 x 40" down to 4x5 film. So about 1:8 . I am trying to pick a lens (and a lens fl) to perform this task which will yield the sharpest images to the entire film area. I tried researching flat field lenses, as this task is surely a flat field task. After reading this article by Ron Wisner, I am starting to beleive there is no real benefit of a flat field lens over a traditional lens. (except maybe for wide angle flat copy, which can be avoided by choosing a longer fl lens)

http://www.wisner.com/myth.htm

Here is my thinking so far, please feel free to correct or add to any of my thoughts.... I have no experience at doing repro work.

1. Lens fl - First I take the image diagnol of the film, 150mm, and desire a fl lens which is at least equal to this. My theory is, the larger the angle of view of the lens, the more obstacles a lens must overcome, which prevents the lens from being as sharp as a normal fl lens. (normal fl being defined as, lens fl = film diag.) Of course, the same applies to camera lenses.

2. MTF values - Next, I looked at the MTF curves of lenses within this "normal fl", and within the 1:8 magnfication working area.

3. The obvious items, such as image circle, light fall off, distortion, etc.



Now, my first thought was that a darkroom lens (not camera lenses) would be most suited for this task, and it's possible they are. But after reviewing the MTF values of Schneiders enlarging lenses, I began to second guess this. I found the Schneider 150 mm Digitar lens (camera lens) to be superior to the enlarging lenses in just about every aspect at this magnfication ratio??

Then, I began to think it was the "flat field issue" that made the darkroom lenses superior for this task, but Ron Wisner article convinced me otherwise. (despite Bob Solomons small mention of this a few weeks back) Then I began to think, possibly the latest digital lenses simply trumped the previous lens technologies, whereas in the past, darkroom lenses would produce superior results vs. camera lenses, but possibly the latest lens designs and glass types may have given the digitar camera lenses an advantage, even in reproduction / darkroom applications? Of course, this only applies when the digitars specs can accomplish the reproduction requirements, which is not always the case.

Or, I am missing something completely? Maybe when the digitars are suitable for reproduction, they are the best option available, even though they were designed for camera use? What criteria would you use to pick such a lens?

If the digitar becomes the lens of choice, I am considering a vacuum back, as the 150mm digitar is optimized for this magnfication ratio at f8, which can make the ol "depth of focus" an un-welcome deal breaker.

TYIA

bglick
6-Feb-2006, 16:45
One last thing I forgot to mention.

The digitar lense I proposed to use, is designed to take a larger image and compress it onto the smaller film. Hence the MTF values should reflect such.

However, the enlarging lenses I mentioned above are designed to take a small image and project it larger, therefore there MTF values reflect such, which is NOT the task I am requiring the lens to perform. The Schneider web site did not have any "copy lenses" listed on their web site.

Now, I assume a "copy lens" is an enlarging lens designed to be used in the reverse manner, i.e. compressing an image vs. enlarging it. If this is correct, I could not find any MTF charts for "copy lenses". Can anyone provide a link to such? I would assume, if any lens is a tough competitor to the Digitar, it must be one of these "copy" lenses? Right?

Dan Fromm
6-Feb-2006, 18:45
Look, Bill, all decent lenses have high MTFs near the center of the field at all apertures. So use a lens that's longer than needed to avoid an MTF that falls off towards the edge of the field you need. Incidentally, that's why MTF at the edge increases with magnification.

Both taking lenses and enlarging lenses are optimized for "large in front", "small behind." The lenses don't care whether light passes from front to rear, as when taking a picture, or from rear to front as when printing. That's why some of us use enlarging lenses for closeup photography.

Enlarging lenses, except those made for printing murals, are typically optimzed for more-or-less the magnification you need.

Just grit your teeth and ante up for a 210 Apo El Nikkor.

Oh, and by the way, how do you plan to align subject and film plane?

bglick
6-Feb-2006, 19:38
> So use a lens that's longer than needed to avoid an MTF that falls off towards the edge of the field you need.

Dan, the 150 Digitar has better MTF throughout the image circle, vs. any of the other suitable lenses I could find, including the Schneider enlarging lenses?

Why grit my teeth and buy 210 Apo Nikkor? I can't find any information on it? Why do you suspect it being better then the digitar?

As for alignment, I will save that possibly for a new thread, as this has been discussed in several threads.

Jack Flesher
6-Feb-2006, 21:37
The digitars, while exceedingly sharp, have relatively small image circles; the 120 won't fully cover 4x5 film. Hence a copy lens is probably going to be your best bet. Personally I would consider a Schneider G-Claron 210 -- they can be found used in shutters for under $400...

Henry Ambrose
6-Feb-2006, 21:49
This will likely do a better job for your repro work:

Chromatics Art Scanning Camera (http://chromatics.com/Serv-ArtCopy.htm" target="_blank)

For the smaller subjects a flatbed scanner will work wonderfully. Unless having this work on film is the final desired output you won't beat scanning the original art.

If film shot in your camera is a must then, in general, longer lenses will make your job easier. How about a repro lens? They are cheap, readily available and made for doing the kind of work you have in mind until you get down to the 4x5 size objects. At that point a duplicating lens is in order for the very finest result.

Just because the 150 Digitar is best across its circle does not mean that its better than any other lens with a much larger circle. The larger circle lens MTF may fall off at 200mm but be extremely sharp over the area you'll be using.

But again - why not scan this stuff? For any method of repro I can think of, scanning wil give a better result.

bglick
6-Feb-2006, 22:41
Jack the image circle of the digitars is 150mm, enough for 4x5.... the sharpness is extraordinary even at the edges....

Henry, I failed to mention, the film is the final product.... I am coming from a digital file to film, not the reverse. I can not find a better MTF over the 150mm image circle I need.... no LF lenses even come close to the digitar, at least for all the modern Schneider offerings. I would use a longer digitar, but 150mm is the longest they currently make. I think Rodenstock may make one digital lens which is longer, but I have not seen its MTF chart.

Paul Moshay
6-Feb-2006, 23:09
WG, I have only been doing art copy work in my studio for 14 years now and I use the GClaron line of lenses, 150, 210, 270 and 305. They are perfect for the job from 4x5 to 8x10 transparencies. They are low in cost and easy to use and produce images that can be printed billboard size at highest quaity. Quit grousing over MTF's and go to work, you will like them. Paul

bglick
6-Feb-2006, 23:19
Paul, unless the entire MTF principle is bunk, the digitar MTF's out perform the G Claron MTF's by a factor of about 3x, yep, 300% better, not 30% better. If the MTF's are only marginaly accurate, well, for that type of improvement, I am willing to do some grousing (whatever that means)

Tadge Dryja
7-Feb-2006, 04:10
You seem pretty set on the digitar, eh? Why bother asking if you've already made up your mind..?

That said, why use a lens at all? You want to go from a digital file to a piece of 4x5 film. Why have a print in between? Use a film plotter. There are various labs that will do that for you, for much less than the cost of the equipment you're considering.

bglick
7-Feb-2006, 06:23
> Why bother asking if you've already made up your mind..?

I don't have my mind made up, hence why I am asking. I just can't find any lenses that seem to out perform the digitar.

I agree with the film recorder concept, but only a CSI film recorder can acheive 40 lp/mm on 4x5, the others perform very inferior, and sharpness is of the upmost criteria, hence why re photographing a back lit trannie will provide the sharpest 4x5 film. CSI film recorders have been discontinued for several years, and no parts are made, so not a good long term solution. In addition, this procedure will be done often, so I need a reliable and consistent method. The only 4x5 film recorders made today, CCG, write at 30 lp/mm to the film, which again, is much inferior to what a sharp lens can deliver to film.

Jack Flesher
7-Feb-2006, 08:35
You won't find any lenses that outperform the digitars... As I said above they are exceedingly sharp.

The real question is do you need that resolution and are the IC's capable of allowing you to generate the image you desire? You seem to believe that the 150 will cover 4x5, and it may, but I suspect you will be left with little to zero avaialble extra IC for movements -- movements that are often critical in copy work to get everything lined up perfectly.

Of course you can back off and crop, but now you are giving up some of the gross resolution you gained by going with the digitar to begin with.

FWIW, I tested a bunch of LF lenses recently using my friend's scanning back. One of the best lenses we tested was a 90mm Digitar -- it made just over 65 LPMM at the center, but fell off significantly 2/3rds the way out. Interestingly, my Rodenstock 150 APO S also made just over 65 LPMM in the center and its resolution also falls off significantly 2/3rds the way out -- but 2/3rds the way out on the APO S is already way beyond the edge of a sheet of 4x5 film. This is what Bill was trying to explain to you above.

It really does sound like your mind is made up, as Tadge said, so why not buy the Digitar and then get back to us with how it works out -- IOW is the IC really big enough and is the MTF really accurate?

;),

bglick
7-Feb-2006, 08:54
Jack, thanks for the post....

My only hesitation on the digitars is the Image circle and light fall off.... and of course, what you mention, just how accurate are the MTF's are to begin with? I appreciate you sharing the tests you performed.

After further investigations, I found the digitar information to be misleading, as the Schneider USA site uses the same MTF charts as the German site, however, the German states the MTF charts are for 80mm image circle, whereas the USA site states its for a 150mm image circle. Buried deep in a Schneider brochure, it states the image circles are quoted at max. resolving areas, not the max size image circle the lens can produce. I think the USA site got this information crossed, not surprising. Based on this, and your findings, maybe the Rodenstock APO S is a better choice, as this also has demonstrated some excellent capabilities.

Jack Flesher
7-Feb-2006, 09:46
WG: The APO S is probably not going to perform as well at copy distances -- under 1:10 -- as some of the lenses already mention specifically designed for copy work, like the G-Clarons.

The fact is, a lens is generally designed to generate best performance at over 1:10, between 1:2 and 1:10, 1: 1 and 1:2 or 4:1 through 1:1. Some Rodenstock and Schneider copy lenses are labeled 1x or 2x corresponding to the latter two ranges above. Most of the modern plasmats are better from 1:10 through infinity, and are not true flat-field lenses. Other lenses are optimized for copy work in the 1:2 through 1:10 range -- like the Schneider G-Clarons and Rodenstock Rodagons.

Paul has been doing copy work for several years and can probably tell you more about this than I can since I don't do it for a living. Suffice it to say, there is more to selecting a copy lens than its MTF or image circle -- and it may save you a lot of grief to listen to what folks that do this for a living have to say...

Cheers,

Jack Flesher
7-Feb-2006, 09:59
I should add that I chose a G-Claron as my 210 landscape lens: It is not as hyper sharp in the center as my 150 APO S, but I don't need a 210 often for the way I see landscapes and the G-Claron is relatively light and compact -- especially compared to the 210 plasmats! So it is adequate as a "normal" lens (especially stopped down), but then serves beautifully as my "macro" lens when I want to capture rock patterns or other naturally-ocurring detail with my 4x5 in that sub 1:10 zone...

bglick
7-Feb-2006, 10:16
> Other lenses are optimized for copy work in the 1:2 through 1:10 range -- like the Schneider G-Clarons and Rodenstock Rodagons.

Jack, I agree with this premise, hence why I started this thread. I would have suspected the "copy" lenses would have been better suited for this task, but the MTF data of the digitars clearly pointed me in the opposite direction, which baffled me. Trust me, if the MTF data was even close, I would go with the lens that was designed specifically for this task. Remember, the MTF data I am reading for the digitars is in the same magnfication ratio I plan to use them for, 1:8. (some listed 1:10, some 1:5)

I just checked the Rodagons MTF values at 150mm fl, and even they are much inferior to the Digitars. Either I am missing something, or the new digitar lens design have really trumped older lens designs....or, the MTF data is not reality, but rather a marketing aide. Although, when you look at the lenses with excellent MTF values, they are typically the ones people rave about.

I am not discounting any contributions from people who do copy work. The problem is, unless someone is comparing one lens vs. another, its hard to evaluate their comments. I used to think the SS XL lenses are the sharpest lenses made, till I saw the Mamiya 7 lenses.....then, I saw the digitar 80mm lens and it slightly trumped the M7 lenses. Get my drift? Now copy is a bit of a different animal, but at 1:8, its not too far removed from normal, or non copy lenses... at least that is how I am seeing it......

Jack Flesher
7-Feb-2006, 10:37
Then get the Digitar. Just be sure to report back to us after use it the first time.

;),

paulr
8-Feb-2006, 10:36
"unless the entire MTF principle is bunk, the digitar MTF's out perform the G Claron MTF's by a factor of about 3x,"

I don't know how these would compare in real life, but remember that the G Claron is optimised for 1:1 reproduction (with an acceptable range of 1:5 to 5:1). The digitar is an all purpose lens. Those MTF curves don't reflect its performance way outside it's intended magnification range.

it's an interesting question, because the digitar is a much more modern design, but you're also asking it to do something it wasn't designed for.

I'd suggest you call Schneider USA and ask a tech rep. It's been a while since I've done this, but years ago, I got hour long lectures on the phone about optics (and these weren't sales pitches ... it was real information).

If there's no one that helpful there now, shoot me an email and i'll pass along the email address of a product specialist I know at Schneider in Germany.

bglick
8-Feb-2006, 10:57
Paul, I agree that one of the weakness of MTF data is, it is often not supplied at the magnfication ratios you plan to use the lens for, which in most cases, those magnifications would represent an inferior MTF vs. what they offer you. (remember, they are trying to sell lenses) However, I disagree with your comments on the digitars, as the MTF's listed are at the image magnfication I plan to use them for, so therefore, I don't think I will be using it for a task they were not designed for?

I did shoot an email off to Schneider Germany, i will keep you posted.

paulr
8-Feb-2006, 11:28
the other question is what is the final use of these negatives ... how many times will they be enlarged. the digitar lenses, showing mtf values for up to 60 lp/mm, are clearly designed for high multiples, like 4X to 12X enlargements. it's likely that they're great at smaller enlargements too, but they may not be the best. i'd definitely be interested to hear what the schneider people have to say about this.

Jim Galli
8-Feb-2006, 16:26
I tend to take the seat of the pants approach. Dan Fromm gave you the best advice. Longer than you need will solve resolution, mtf falloff, and light fall off. Kenro used to supply a 10 3/4" Goerz copy lens (dagor type) that is simply stunning. Maybe all the modern numbers games would slaughter it, I dunno. Ask the Marlboro Man up on the billboard. In my profession we've yet to do better than a 127 Ektar for painting light on 1/2 frame 35mm film. Before you throw up, think about it. That little cheap tessar will resolve 80lppm in that little sweet spot we need to cover that small film. Just a slightly different approach than leanest meanest I guess.

paulr
8-Feb-2006, 21:49
"Longer than you need will solve resolution, mtf falloff, and light fall off. "

longer gives you flatter performance from center to edge, but all else being equal it will be less sharp in the center than a shorter normal lens. more of a tradeoff than a solution.