Pages: [1] 2   Go Down

Author Topic: Hasselblad Lenses - new vs. old for P45  (Read 20503 times)

bjornaagedk

  • Full Member
  • ***
  • Offline Offline
  • Posts: 133
Hasselblad Lenses - new vs. old for P45
« on: November 23, 2005, 01:39:10 pm »

My situation: Commercial Photographer, has been shooting with Hasselblad and Phase One H20 for 2 years. My lenses are from the C-series (Black T*) up to CF-lenses. Excellent quality with the H20 so far. Now I consider upgrading to the P30/P45 digital back, but what about my collection of lenses? I don't like if I have to throw them out and buy the newest lenses (maybe a complete Hasselblad H2 set) to get maximum image quality with the new digital backs.

Does anyone know - how good are the old lenses compared to the newest?? Will they loose in front of a top digital back like P30/P45 ??

Any ideas? Or where can I find more information?

B
Logged

michael

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 5084
Hasselblad Lenses - new vs. old for P45
« Reply #1 on: November 23, 2005, 05:18:55 pm »

I don't believe that the H series lenses are any better than the earlier Zeiss glass. Maybe the same, but not much better or worse. So I wouldn't suggest upgrading for the glass.

I use a P25 with Zeiss glass on a Contax 645 system, very similar lenses optically to the ones you likely have. I've compared shots taken with these and with an H1 Hassy and Fuji lenses, and see little if any difference.

But that's not the end of the story.

I believe that backs like the P25 are currently limited by their lenses in general. I recently have been shooting with my P25 on a Cambo Wide DS body with a Schneider 35mm Digitar. The improvement in resolution over any of my Zeiss glass, including the much praised 110mm Apo Makro, is crearly visible even in A3 prints.

In other words, the whole medium format industry really needs to upgrade its lenses to match just the current generation of 22MP backs let alone the new 30 and 39MP backs coming in then next few months. I've seen images shot with the P45, and that raises the bar to a whole new level again. I can only imagine what a P45 combined with a Scheider Digitar will produce. I hope to test this combination in February.

Michael
Logged

willconnor

  • Newbie
  • *
  • Offline Offline
  • Posts: 44
Hasselblad Lenses - new vs. old for P45
« Reply #2 on: December 07, 2005, 03:38:27 pm »

The large format digitars sound interesting, but I have heard they are optimized only for f5.6 to f11. Smaller than that and image quality plummets.  I don't believe f11 would always be adequate for landscape or architecture where I usually want extensive depth of field.
Logged

mikeseb

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 482
    • http://www.michaelsebastian.com
Hasselblad Lenses - new vs. old for P45
« Reply #3 on: December 07, 2005, 09:11:49 pm »

Quote
In other words, the whole medium format industry really needs to upgrade its lenses to match just the current generation of 22MP backs let alone the new 30 and 39MP backs coming in then next few months. Michael
[a href=\"index.php?act=findpost&pid=52031\"][{POST_SNAPBACK}][/a]

Given that 35mm DSLR's have just about lain waste to medium-format camera manufacturers, one doubts that "the whole medium format industry", such as it is, has either the impetus or the cash to undertake this upgrade.

Hard to imagine better glass than my contax's zeiss lenses, but I'd also love the chance to find out just how much they limit, say, my brand-new fantasy P45!
Logged
michael sebast

Concorde-SST

  • Full Member
  • ***
  • Offline Offline
  • Posts: 102
Hasselblad Lenses - new vs. old for P45
« Reply #4 on: December 08, 2005, 11:37:43 am »

Its very simple - of course you can get much
better lenses than the ones from Contax.

Most large format primes (Schneider+Kreuznach,
Rodenstock etc.) are way better - it could be
that there might be an adapter solution for using
those lenses on Contax cameras (or other brands)
I don´t know.

And - then, think military - the lenses the satellites
use - they are the most advanced ones you can
"get" !

Just a matter of how much somebody is willing
or able to pay for.

best regards,

Andreas.
Logged

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6600
Hasselblad Lenses - new vs. old for P45
« Reply #5 on: December 08, 2005, 06:53:00 pm »

Quote
Its very simple - of course you can get much
better lenses than the ones from Contax.

Most large format primes (Schneider+Kreuznach, Rodenstock etc.) are way better[{POST_SNAPBACK}][/a]
Maybe not as simple as you think, if you look at the MTF graphs for most LF lenses.
(e.g. at [a href=\"http://www.schneideroptics.com/photography/large_format_lenses/]http://www.schneideroptics.com/photography..._format_lenses/[/url] )

They might well give more relative resolution in "lines per picture height", but if you want to use them with smaller format film or sensors, it is absolute resolution in lines per mm that count, and on this measure, LF lenses fall behind good medium format lenses and even further behind good 35mm format lenses (with even smaller format lenses like Olympus ones for FourThirds even better). This is a simple basic optical trend: lenses covering the same angular filed of view but at shorter focal lengths and over smaller image circles generally  have higher absolute resolution (lp/mm).

LF lenses typically have MTF at  published at 5, 10 and 20 lp/mm, with 20 lp/mm and even 10 lp/mm not overall better than for good 35mm lenses at 30 or 40 lp/mm (or Olympus Four Thirds lenses at 60 lp/mm).

The exceptions that prove the rule are the Schneider Digitar lenses, which are view camera lenses, but designed for smaller formats and smaller image circles, for digital use. Schneider gives MTF at 20, 40 and even 60 lp/mm (twice what most 35mm lens makers give!) and the numbers are very impressive. This matches Michael's practical experience: he expresses a very high opinion of these lenses at http://luminous-landscape.com/reviews/came...ensive-ps.shtml


About quality being a matter of how much you are willing to pay: resolution limits are determined more by inherent design optimization rather than manufacturing tolerances, so a dominant cost is design cost, and the lenses for which customers are willing to spend the greatest total amount on design are the ones that have a very large number of customers to share that once-off cost. A design that sells hundreds of thousands or even millions in its lifetime can easily justify many millions of dollars in design effort. That tends to give higher volume sectors (35mm film, mainstream digital SLR formats) an advantage that grows in time over lower sales volume larger formats (large and medium format film, medium and 35mm format digital.)
Logged

Anon E. Mouse

  • Full Member
  • ***
  • Offline Offline
  • Posts: 197
    • http://
Hasselblad Lenses - new vs. old for P45
« Reply #6 on: December 08, 2005, 08:48:15 pm »

Resolving power is a matter of diffraction and is dependent on the effective aperture in use. Digital "large-format" lenses are optimized to work at f/8 or f/11. 4x5 lenses are designed to work at much smaller apertures. If you compare the data of the 90mm film and digital Schneider lenses, you will not see much of a difference. There is no mystery to "digital" lenses. They are simply more corrected for the larger working aperture which they require (this correction has to do with field flatness and abberations). Rodenstock and Schneider redesigned their camera lenses to correct abbrerations for those apertures. (They need to work at larger apertures because of the resolution of the sensor.) Large-format lenses are corrected for smaller working apertures because they can afford the loss of resolution at the film plane, not because they actually have less resolution.

BTW, resolving power is always giving at lines/mm at the image plane. That is what resolving power is. Different specification for resolving power are published for different formats not because the lenses resolve less, but because of the size of the circle of confusion required to make an image that appears sharp. A large-format lens can resolve 60 lines/mm, but what is the point of giving that data when the film image is only enlarged 2x to make an 8x10 print?
Logged

BernardLanguillier

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 13983
    • http://www.flickr.com/photos/bernardlanguillier/sets/
Hasselblad Lenses - new vs. old for P45
« Reply #7 on: December 08, 2005, 10:38:28 pm »

Anon,

LF lenses are not just optimized for smaller apertures, they are also designed so as to deliver a very uniform image over a large image circle, as opposed to digital lenses being optimized for delivering  a smaller image circle, but with a higher overall sharpness, less CA etc...

The comparison of the two 47 mm Schneider sells appears to be quite telling in this regard. My widest LF lens being a Schneider 58 mm, I have never had the chance to use either myself though.

Regards,
Bernard
« Last Edit: December 08, 2005, 10:39:01 pm by BernardLanguillier »
Logged

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6600
Hasselblad Lenses - new vs. old for P45
« Reply #8 on: December 09, 2005, 02:31:27 pm »

Quote
Resolving power is a matter of diffraction ...

Different specification for resolving power are published for different formats not because the lenses resolve less, but because of the size of the circle of confusion required to make an image that appears sharp.
[a href=\"index.php?act=findpost&pid=53088\"][{POST_SNAPBACK}][/a]

Firstly, there is a lot more to resolution performance than diffraction limits. If diffraction were the whole story, MTF graphs for particular lenses would be redundant: there would be a single MTF graph for each combination of aperture, focal length and lp/mm level.

Secondly, you really need to look at some MTF graphs, as I suggested before. Even when data is available at the same lp/mm level, and even if one looks at data out to the same distance from the center, LF lenses consistently have lower MTF than 35mm lenses, and so on down to smaller formats. That is basic optics or engineering: at a given focal length, designing for larger angular coverage and correcting aberrations over a larger image circle involves additional design trade-offs that tend to reduce resolution in the central portion needed by a smaller format.

One more time: even excellent LF lenses typically have distinctly lower MTF at 20lp/mm than good 35m lenses at 30p/mm or FourThirds lenses at 60lp/mm. If we had MTF curves at say 60lp/mm for all three, the performance gaps would increase. I choose 60 lp/mm is about where resolution needs are going as pixel spacing approaches 5 microns, as Schneider clearly thinks since they give 60lp/mm curves for their Digitar lenses at 60mm and longer.
The Digitar lens MTF at 60lp/mm is impressive, but distinctly worse than any smaller format lenses I have seen 60lp/mm curves, all of which are FourThirds. Even the modestly priced, ultra-wide ranging Olympus Zuiko Digital 18-180 zoom (probably made by Sigma) outdoes Digitars at 60lp/mm.
Logged

Anon E. Mouse

  • Full Member
  • ***
  • Offline Offline
  • Posts: 197
    • http://
Hasselblad Lenses - new vs. old for P45
« Reply #9 on: December 11, 2005, 09:55:09 pm »

Here you can get the MTF data for a Nikkor 85mm f/1.8 D:

http://www.photodo.com/nav/prodindex.html

Here you can get the MTF data for a Super Angulon 90mm lens:

http://www.schneideroptics.com/photography.../super-angulon/

Hardly a significant difference even taking the larger image circle into account.
Logged

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6600
Hasselblad Lenses - new vs. old for P45
« Reply #10 on: December 12, 2005, 03:26:54 pm »

Seeking out one example rather than looking at overall trends is bad science, but let us look at it. The Super Angulon data are for 5, 10 and 20 lp/mm while those for the Nikon are at 10, 20 and 40, so I will compare at the common value of 20lp/mm. For the Super Angulon the MTF values are at best about 70% on axis, fall to below 50% by the edge of the 35mm format image circle radius, below 40% at the limits of the 645 image circle, and continue on down to 30%. The Nikon is about 83% on axis, and stays around 80% all the way out to the 35mm image circle radius: substantially better, I would say. And that is with your single hand-picked example.

Remember, your original claim was that the larger format Schneider lenses would be better than Contax lenses when used on a Contax 645 body, so MTF graphs showing one to be noticably worse than a smaller format lens do not support your cause.
« Last Edit: December 12, 2005, 03:29:10 pm by BJL »
Logged

Concorde-SST

  • Full Member
  • ***
  • Offline Offline
  • Posts: 102
Hasselblad Lenses - new vs. old for P45
« Reply #11 on: December 12, 2005, 03:34:15 pm »

How about the idea of the absolute amount
of light falling through the lens (or rendered
by the lens?) I think the glass quality and
colour rendition makes a difference too.

Hope thats not too far off from the topic.

best,

Andreas.
Logged

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6600
Hasselblad Lenses - new vs. old for P45
« Reply #12 on: December 12, 2005, 05:08:43 pm »

Quote
How about the idea of the absolute amount of light falling through the lens
[a href=\"index.php?act=findpost&pid=53364\"][{POST_SNAPBACK}][/a]
It might be a shift of topic, but I for one think that this is an excellent point, relating to my frequent observation that lenses of equal absolute aperture size with different formats gives equal total light gathering rate, equal light gathering rate per pixel if pixel counts are equal, as well as equal DOF and equal diffraction limitation on prints of equal size. Equal absolute aperture size (diameter) meaning that relative aperture (a.k.a. f-stop) is proportional to focal length and image format size.

This is especially significant with medium format, where aperture size choices quite clearly stay the same (or even get smaller) as format size increases, probably because aperture choices are so often dictated by considerations of appropriate DOF and adequate diffraction control. For example, lenses for larger medium formats have higher minimum f-stops (a.k.a. relative aperture). And all have distinctly higher minimum f-stops than 35mm.

With equal absolute aperture, light is gathered from each part of the subject at the same rate, so with larger formats like 6x6 or 6x7 it is spread over into a somewhat larger, dimmer image than with 645 or the new digital options 36x48mm and 33x44mm.

That means that when one uses the same exposure time and pixel count, the same amount of light is gathered per pixel on the sensor. A larger sensor and focal length combination wins only if and when it can use an exposure time so long that the smaller photo-sites of a smaller sensor would suffer blown-out highlights: the advantage in highlight headroom and dynamic range typically expected for larger photo-sites and sensors. However, the current choice is either older 9 micron photo-sites or newer 7 micron (Dalsa) or 6.8 micron (Kodak) photo-sites which have about the same dynamic range, due to technological progress.

Yes, the new technology applied to the old 9 microns photo-site size could probably give even better DR, but that is hardly being offered: only in the forthcoming 18MP Kodak sensor in the smaller 33x44mm format, for the forthcoming Phase One P20 and the planned Pentax integral digital MF body. The smaller sensor size and lower pixel count leaves that sensor at an effective  DR disadvantage to the new 36x48mm options.
Logged

Anon E. Mouse

  • Full Member
  • ***
  • Offline Offline
  • Posts: 197
    • http://
Hasselblad Lenses - new vs. old for P45
« Reply #13 on: December 12, 2005, 09:18:09 pm »

BJL, are you sure you understand MTF data? You do realize a wide-angle large format lens is not designed to work wide open? You do understand the difference between an f/8 and f/22 curve, which gives a possible difference in resolving power of 225 l/mm and 82 l/mm based on diffraction.

You can go to those links I provided and look at any of the MTF data. I chose those lenses simply because of focal length. I would recommend you also try some of the 35mm zoom lenses as well as some other large-format designs. Take care in which curves you are comparing and the aperture used to generate the curve.

However, I am interested in your cites or links to "overall trends." Can you provide some? I am interested in your information. Your claim is "always better" and "significant" for the differnance between the 35mm and large-format.

Quote
Remember, your original claim was that the larger format Schneider lenses would be better than Contax lenses when used on a Contax 645 body...

Please don't misquote me. I never claimed any format lenses would be "better." I am saying there is not much of a difference. Fortunately, you have some science to back up your claim and I am looking forward to a link.
Logged

Anon E. Mouse

  • Full Member
  • ***
  • Offline Offline
  • Posts: 197
    • http://
Hasselblad Lenses - new vs. old for P45
« Reply #14 on: December 12, 2005, 09:24:38 pm »

Quote
With equal absolute aperture

BJL, do you mean entrance pupil?
Logged

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6600
Hasselblad Lenses - new vs. old for P45
« Reply #15 on: December 13, 2005, 12:54:41 pm »

Quote
BJL, do you mean entrance pupil?
[a href=\"index.php?act=findpost&pid=53405\"][{POST_SNAPBACK}][/a]
Yes: different authors seem to use "effective aperture diameter", "entrance pupil" and "absolute aperture size to mean the same thing.
Logged

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6600
Hasselblad Lenses - new vs. old for P45
« Reply #16 on: December 13, 2005, 03:40:48 pm »

Quote
BJL, are you sure you understand MTF data? You do realize a wide-angle large format lens is not designed to work wide open? You do understand the difference between an f/8 and f/22 curve, which gives a possible difference in resolving power of 225 l/mm and 82 l/mm based on diffraction.
Yes I understand all of that (except that shorter focal length LF lenses like 90mm are more likely to be used at apertures significantly larger than f/22). My description applied to all the aperture ratio data provided for the Super Angulon, both f/5.6 and f/22. Since you introduced those data, can you point to any choice of aperture ratio where that larger format lens does not have distinctly lower MTF within the 35mm format image circle than the smaller format lens?

Quote
Please don't misquote me. I never claimed any format lenses would be "better." I am saying there is not much of a difference.
My mistake: I confused you with the poster Concorde-SST that I was originally responding to, who claimed that
Quote
... of course you can get much better lenses than the ones from Contax.

Most large format primes (Schneider+Kreuznach, Rodenstock etc.) are way better - it could be
that there might be an adapter solution for using those lenses on Contax cameras
My posts relate to this claim of advantages to using larger format lenses (like Schneider's) with the heavy "crop" a smaller format body (like a Contax 645) instead of using lenses designed for the smaller format. No disparagement from me of larger format lenses when used with the larger formats and generally lower degrees of enlargement for which they were designed!

My specific claim is that if you compare the lenses of the same focal length over the same image circle size at the same resolution level (l/mm), a lens designed specifically for that image circle will usually have higher MTF than one designed to cover a larger image circle, and thus corrected over a wider angular field of view.

Your chosen LF/35mm format pair corroborates this.

Since MTF graphs are available at the same 20l/mm for Olympus Zuiko Digital FourThirds format lenses (image circle radius about 11mm) and for Schneider view camera lenses, I will compare the three "ZD" primes to Schneider lenses of the same or similar focal length at 20l/mm (lower  curves on Schneider graphs, upper curves on Olympus graphs.)
Olympus data available at http://www.olympusamerica.com/e1/sys_lens_spec.asp

I) Olympus Zuiko Digital f=50 f/2; Schneider Super-angulon f=47 f/5.6
The smaller format lens has MTF at f/8 nearly a flat line 85% across the field, and wide open at f/2, MTF is about 80% across the image circle (i.e. out to radius 11mm).
The larger format lens has MTF at f/5.6 of 80% on axis, down to 40% to radius 11mm; at f/22, MTF is 70% on axis and falls to between 40% and 60% within radius 11mm.

Or compare that same Schneider lens to the Olympus 14-54 mid-priced zoom lens at 54mm: it has MTF at either f/8 or wide open f/3.5 of about 90% at the optical axis and above 80% out to radius 11mm.

II) Olympus Zuiko Digital 150mm f/2, Schneider APO-Symmar "L" 150mm f/5.6
The smaller format lens stays at about 90% at both f/8 and wide open f/2;
the larger format lens has MTF 70% or under at all three apertures given, f/5.6, f/11, and f/22.

III) Olympus Zuiko Digital 300mm f/2.8, Schneider APO-Symmar "L" 300mm f/5.6.
Much as in the previous case.

In fact, in these pairings, the smaller format lens has similar or better MTF at 60l/mm to what the larger format lens has at 20lp/mm.


The few cases I can find of "not much difference" is comparing the Schneider Digitars to Olympus Zuiko Digitals. In particular, the Digitar f=47, f/5.6 (40mm image circle radius, or just enough for 6x6 medium format) and Olympus f=50mm f/2 macro have quite close MTF at the one common resolution level of 20lp/mm, a good 80-85%. The Digitar f=150mm, f/5.6 (75mm image circle radius) is also fairly close to the ZD f=150, f/2. But those Digitars do have the design advantage of far smaller maximum apertures than the Zuiko Digitals.
Logged

Anon E. Mouse

  • Full Member
  • ***
  • Offline Offline
  • Posts: 197
    • http://
Hasselblad Lenses - new vs. old for P45
« Reply #17 on: December 16, 2005, 12:25:43 am »

Sorry I did not reply sooner. I just got back from a business trip.

Quote
Yes I understand all of that (except that shorter focal length LF lenses like 90mm are more likely to be used at apertures significantly larger than f/22).

No, not at all. f/22 is a resonable aperture to use a wide-angle large-format lens. Because of the image circle, you are not going to be free of mechanical vignetting until f/11. Using tilts to optimize depth of field is not always practical and the camera is working at much less depth of field than a smaller format with equal angle of view. f/22 would be a good working aperture.

Quote
My description applied to all the aperture ratio data provided for the Super Angulon, both f/5.6 and f/22. Since you introduced those data, can you point to any choice of aperture ratio where that larger format lens does not have distinctly lower MTF within the 35mm format image circle than the smaller format lens?

This is way I am asking if you understand MTF data. Different format lenses are not designed under the same parameters. 35mm are designed to be used wide open, large-format lenses are not. Image circle changes the requirements further. You are going to have to understand optics the way they work and will require you to be able to compare data made under different conditions. If you are just going to say that number is bigger than the other number, then it is a pointless conversation.

We can both go back and forth finding MTF results to support an opinion - I did notice your Zuiko Digital link. I can bring out 35mm zoom lenses that have awful MTF curves. The other problem is the definition of "significant." Is the 5%, 10%, 20% difference? MTF suggests the performance of the lens and is a very useful tool to diagnose a problem, but it does not tell how images appear. A 20% difference in MTF becomes smaller as subject contrast goes down. But even with lower values, the lens is resolving. Then it is just "my opinion" over "yours" on whether it is "significant."

I often see reviews at LL bashing test results as not being relavant in the "real world." That is not true. Neither is simply sticking to numbers as absolute values. Using either position does not help anyone. When lenses and cameras are tested, both MTF and actual image data are used to evaluate them. There is good reason for it. Neither can give the whole picture (no pun intended). Modern lenses are fine optics regardless of format and are close in quality.

While I am guilty of it, I am getting tired of photographers bickering over technical minutiae. Most of it seems as mine-is-better-than-yours spats with each side cherry-picking arguments. Personally, I see nothing that says lenses of a particular format is "significantly better" than another format. If you don't agree, than fair enough.

Peace.
Logged

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6600
Hasselblad Lenses - new vs. old for P45
« Reply #18 on: December 16, 2005, 02:19:44 pm »

Maybe there is still a misunderstanding here. I started by solely addressing the claim that larger format lenses (e.g. from makers like Schneider) would give superior quality on a smaller format body (e.g a Contax 645) compared to using good lenses designed for that smaller format.

I then went a bit further, towards the general idea that the best tools developed for a particular purpose (like a particular format) are in general superior to tools designed for different purposes (larger image circles, wider angular FOV, far higher f-stops, etc.) I am solely concerned with the question of what are the best tools for a particular particular job (originally, MF).

I am not knocking Schneider for making truly excellent hammers that perform poorly when used as screw-drivers.


And I have certainly not biassed my selection of examples against the larger format lenses. For example, I scoured the Schneider web site for the best lenses at each focal length, I used your chosen exampes, and I even handicapped the smaller formats in some comparisons by using 4/3 format zoom lenses.


Since your recent responses contain no positive evidence, but only attempts to discredit the data and arguments I have presented, the discussion seems to have run its course.
Logged

stevem

  • Newbie
  • *
  • Offline Offline
  • Posts: 1
Hasselblad Lenses - new vs. old for P45
« Reply #19 on: February 20, 2006, 06:11:47 pm »

Just wanted to know because you guys seem like the ones to ask... but does anybody know if schneider or rodenstock are planning on improving the current digital lenses (scheider)?

I just shot a job using a P45 and upon learning final output sizes from client (happens a lot in advertising so don't knock me for not knowing earlier), suddenly wished that I had shot 8 x 10 or even 4 x 5 (not that I would have had the time to do this anyway, given how crazy the industry is now a days).

My problem, is that I need the swings and tilts for focus, but the image quality is just not there compared to our Hasselblad systems, and even RZ lenses. I know you guys love these MTF charts and Shneider doesn't lie about about this kind of stuff, but I'm just not getting the same kind of resolution that I am with say, a hasselblad 80mm, or RZ 65mm. Not to mention, at F11, even though it's a smaller image area to capture on, there just isn't enough depth of field.

So when is the improvement in lenses coming? Why purchase a slow P45 when there is very little practical difference between this and a H25 back?
Logged
Pages: [1] 2   Go Up