Luminous Landscape Forum
Equipment & Techniques => Medium Format / Film / Digital Backs – and Large Sensor Photography => Topic started by: JdeV on March 18, 2009, 03:24:50 pm
-
Hi, I'm doing a test of a P65+ tomorrow against 4" x 5" and 8" x 10" film cameras with negative and transparency film to be drum scanned.
I want to know what the state of the art in digital backs is like now for architectural work.
I have in mind to shoot a big Paris train station interior from on high. This will provide a great deal of fine detail and difficult shadow/highlight information.
Unfortunately I have not been able to get my hands on a sliding back/digital view camera combo so have to put the P65+ on an H2 or Mamiya/Phase body.
Has anyone tested the Hasselblad/Mamiya lenses against the Schneider/Rodenstock digital lenses? Setting aside the larger image circle of the view camera lenses, are they comparably sharp?
I am thinking of using the Hasselblad with a 50mm or a Mamiya with a 45mm. Will these provide a fair demonstration of the back's abilities? I am worried that if I go wider than this (say a 28mm or a 35mm) the lens defects might limit the performance.
Any comments?
-
Hi, I'm doing a test of a P65+ tomorrow against 4" x 5" and 8" x 10" film cameras with negative and transparency film to be drum scanned.
I want to know what the state of the art in digital backs is like now for architectural work.
I have in mind to shoot a big Paris train station interior from on high. This will provide a great deal of fine detail and difficult shadow/highlight information.
Unfortunately I have not been able to get my hands on a sliding back/digital view camera combo so have to put the P65+ on an H2 or Mamiya/Phase body.
Has anyone tested the Hasselblad/Mamiya lenses against the Schneider/Rodenstock digital lenses? Setting aside the larger image circle of the view camera lenses, are they comparably sharp?
I am thinking of using the Hasselblad with a 50mm or a Mamiya with a 45mm. Will these provide a fair demonstration of the back's abilities? I am worried that if I go wider than this (say a 28mm or a 35mm) the lens defects might limit the performance.
Any comments?
Tell me where you want to shoot and I'll bring along my D3x and 24mm shift lens, and P45+ and Mamiya shift. We can do some serious comparing here.
Edmund
-
I would think if you really want to test the back, you should find a Cambo or Alpa camera to mount it on. My own experience shooting with a Cambo Wide DS next to a Hasselblead, the Cambo lenses will out perform the Hasselblad glass.
-
Hi, I'm doing a test of a P65+ tomorrow against 4" x 5" and 8" x 10" film cameras with negative and transparency film to be drum scanned.
I want to know what the state of the art in digital backs is like now for architectural work.
I have in mind to shoot a big Paris train station interior from on high. This will provide a great deal of fine detail and difficult shadow/highlight information.
Unfortunately I have not been able to get my hands on a sliding back/digital view camera combo so have to put the P65+ on an H2 or Mamiya/Phase body.
Has anyone tested the Hasselblad/Mamiya lenses against the Schneider/Rodenstock digital lenses? Setting aside the larger image circle of the view camera lenses, are they comparably sharp?
I am thinking of using the Hasselblad with a 50mm or a Mamiya with a 45mm. Will these provide a fair demonstration of the back's abilities? I am worried that if I go wider than this (say a 28mm or a 35mm) the lens defects might limit the performance.
Any comments?
Hello
I shoot architecture on a leaf Aptus 75 on an Alpa system. If you want to test this new back make sure you shoot with a lot of shift on a wide lens ie 35mm Digitar or wider. This back has a Dalsa chip which is prone to centre fold and colour casting. It will be interesting to see if Phase has truly over come these shortcomings. The other test would be to long exposures of a predominantly dark subject matter ie night shot. The chip will work well at 50 or 100 but how much noise will there be at 200 and 400. Again Phase suggest this is improved. Finally all the back manufacturers claim speeds that are not really there, Its a bit like shooting Velvia, it claimed 50 asa but in reality was 40 ish. I don't know how you could check that. The resolution of these practical problems will determine whether I will buy this back as gain adjust steps and post production time loss are killers
-
Hello
I shoot architecture on a leaf Aptus 75 on an Alpa system. If you want to test this new back make sure you shoot with a lot of shift on a wide lens ie 35mm Digitar or wider. This back has a Dalsa chip which is prone to centre fold and colour casting. It will be interesting to see if Phase has truly over come these shortcomings. The other test would be to long exposures of a predominantly dark subject matter ie night shot. The chip will work well at 50 or 100 but how much noise will there be at 200 and 400. Again Phase suggest this is improved. Finally all the back manufacturers claim speeds that are not really there, Its a bit like shooting Velvia, it claimed 50 asa but in reality was 40 ish. I don't know how you could check that. The resolution of these practical problems will determine whether I will buy this back as gain adjust steps and post production time loss are killers
Sadly I don't think I will be able to get hold of a digital view camera system. I think therefore that I will only be able to test for resolution/colour/dynamic range at this point. I will only have the back for a few hours tomorrow morning. It is €760 a day rental on its own and I am being lent it by the rental company with a body and lens for free so I can't push my luck too much!
I am not interested in bothering to test high ISOs because I've long ago concluded that except in dire emergencies there is no point going more than 1, (maybe 2 stops) above the base ISO of the back. I am very interested in long time exposures but I don't think circumstances will permit much in that direction.
Do you have any comments about the relative merits of the Hassleblad/Mamiya lenses versus Rodenstock/Schneiders when no movements are being deployed?
-
i've just test a rodenstock sironar digital 90 against an hasselblad 100 cf... quite similar in the center... on a 5D Mark II.
I will test them with a phase one next week.
-
The P65 will not be able to do long exposures,as it is a Dalsa sensor.
I would say 60 seconds tops.
The P45+ would be a better choice in this case.
Also the results from a P45+ are as good as 10x8.
Why on earth you want to shoot architecture on a 10x8?
Anyway,that my opinion anyway and I've used 10x8 in the old days,but for today maybe not.
However,if you have a big time client that wants to pay for 10x8 sheet film,go for it.
-
I'm certain the P65+ will show itself to be among the best,if not the best, current digital product
for fine detail rendering but I will look forward to your experience with how the back handles
exposures in the 30-60 second range.
If possible, try to include a significant area of continuous tone to check for banding.
There is another thread at GetDPI that has an example of of P65+ 30 second capture that exhibits quite significant
banding.
The posted sample was shot at 400 iso and 30 seconds so it would be interesting to hear if you can see similar banding
at lower iso's with long exposures
http://forum.getdpi.com/forum/showthread.p...6039&page=2 (http://forum.getdpi.com/forum/showthread.php?t=6039&page=2)
Scroll down to post #40 for the example
Mark
-
i dont think it makes any sense to test the back for architecture without using shiftable lenses .
the lesss important thing in architecture photography is resolution, once a certain level is reached,- which i would rate at 22mp. the most important thing are the lenses, and here esp. the wides from 45 down to 24/23mm as well as the shift capacities.
( one reason why a 8x10" is so nice to use is the image composing on the screen and the narrow dof one can get.
its not that it resolves more than a 4x5". )
-
i dont think it makes any sense to test the back for architecture without using shiftable lenses .
the lesss important thing in architecture photography is resolution, once a certain level is reached,- which i would rate at 22mp. the most important thing are the lenses, and here esp. the wides from 45 down to 24/23mm as well as the shift capacities.
( one reason why a 8x10" is so nice to use is the image composing on the screen and the narrow dof one can get.
its not that it resolves more than a 4x5". )
While 22mp is an abundant amount of information for most uses. I have found the P25+ to be much more prone to moiré than the P45+. I have seen this in a wood slat wall as well as in fabric covered panels on an interior. I expect the P65+ to handle this even better, but yeah.....every once in a while I want a long exposure....
-
i dont think it makes any sense to test the back for architecture without using shiftable lenses .
the lesss important thing in architecture photography is resolution, once a certain level is reached,- which i would rate at 22mp. the most important thing are the lenses, and here esp. the wides from 45 down to 24/23mm as well as the shift capacities.
( one reason why a 8x10" is so nice to use is the image composing on the screen and the narrow dof one can get.
its not that it resolves more than a 4x5". )
Hi Rainer,
Interesting what you said about 22mp being enough. If the Canon new shift lenses turn out to be top performers will that be serious competition? DR in one shot might be an issue but with 35mm it's easy to bracket the exposure.
Kevin.
-
Hi Rainer,
Interesting what you said about 22mp being enough. If the Canon new shift lenses turn out to be top performers will that be serious competition? DR in one shot might be an issue but with 35mm it's easy to bracket the exposure.
Kevin.
yes it will be for sure for many architecture shooters a bif relief to have good shift possibilities soon in 35mm. although there will also be several ( than me ) who appreciate the different way to compose images on ground glass and the still existing better output of the actual mf sensors and lenses enough to go on with mf. but resolution is the less important point herein i.m.o.
-
"The overwhelming majority of architectural clients these days are more than content with the quality from DSLRs and few care to pay more for the MFDB premium."
It strikes me that if fees are based on the quality of your camera more rather than the quality of your vision, then photographers are way out of touch with current client needs. My fees and billable expenses are much higher than they were two years ago when I was shooting only 4x5 film because I am more creative and productive with a DSLR. Even a DSLR can exceed the quality needs of most client applications. Rather than repeat the point here please read this post on my blog: Is Full Frame DSLR Good Enough? (http://kirkgittingsphotography.blogspot.com/2009/01/is-dslr-image-good-enough.html) and From 4x5 Film to DSLR (http://kirkgittingsphotography.blogspot.com/2009/01/from-4x5-to-dslr.html)
-
I tried out all the Canon T/S lenses on a 1DS before committing to my Phase Back/Arca 6x9. The biggest gripe I have is that not only would these lenses distort in the corners but they would often render perfectly straight architecture as a "wavy" line right through the center of the lens. It was just unacceptable. Also, I just find that I make more refined compositions with a view camera than a DSLR... they make me sloppier for some reason.
I dunno, after shooting for 12 years at Hedrich Blessing (prior to my resignation), maybe I'm just spoiled when it comes to good gear and visually literate clients.
Christopher Barrett
View Camera Snob
-
............ Also, I just find that I make more refined compositions with a view camera than a DSLR... they make me sloppier for some reason.
I dunno, after shooting for 12 years at Hedrich Blessing (prior to my resignation), maybe I'm just spoiled when it comes to good gear and visually literate clients.
Christopher Barrett
View Camera Snob
That IS an impressive resume!! Just on that I am inclined to believe your statements. Ofcourse there is a ring of truth in your statement @ DSLR vs View Camera. Often speed does make one shoot first & then fix later.
-
Kirk,
How do you overcome the inbuilt distortion of the canon 24mm shift lens. (I assume the newer version, and the 17mm shift also will still suffer there)
I have found a programme like dxo optics can sort out the lens correction of most lenses really well, but not shift lenses.
To me this is where the inbuilt software lens camera correction of the H 28mm (even with the hts I believe) or shift cameras with their non distorting lenses have the upper hand.
Marc
-
A Canon 21MP frame is fine but the process is soul-less.
I couldn't of said it better....
-
the big shortcoming of 35mm systems of course are the lenses, not the sensors.
not much changed here just because the digital part of both worlds became better and has more pixels.
my assumption that the upcoming 17 and 24mm tse lenses of canon will be better in terms of edge sharpness and especially of distortion still has to be prooved, after these lenses will be available. if they will be better ( because for me the current canon 24tse , or schneider 28pc, or the older 28/35m nikon shift lenses are simply out of discussion for their bottle glass qualities ),- 35mm will be a serious option, if not,- the things will remain the same than now.
its a lot of hassle and much more time consumptive to eliminate this 35mm -lens issues than to start with good corrected files coming from medium format lenses using schneider or rodenstock glass, independent of the final digital output.
if someone works a lot its the cheaper way to work with mf because its faster to deliver professional quality,- although the initial investment is much higher. i dont speak here about the lower level work or work which affords to shoot hi quantities of images and where the clients want an acceptable but not outstanding quality.
the new 5dmk2 delivers very good file quality, so i am back to use 35mm too, but for the longer focals above 100mm and for aereal shootings.
for the wideangle shift work my artec together with the rodenstocks hardly can be beaten in terms of quality and efficiency.
but as i said above, resolution is the last important factor here.
to come back to the topic: therefor it dont make much ( if any ) sense to try out a p65 with a retrofocal camera/lens system for architecture shooting.
-
A Canon 21MP frame is fine but the process is soul-less......I couldn't of said it better....
I couldn't disagree more. Understand. I made my living pretty much exclusively with a 4x5 and film for almost 30 years. To me, making images is never soulless unless you have detached yourself from the process and that is about you-not the camera being used or the quality of the architecture. I used to make my AP photography students photograph a small building near the University of New Mexico that was pretty much a stucco box, because sometimes for clients your you have to photograph crappy architecture and make it look good. Some students balked at the project but once they got into it they made the most astonishing images in amazing light. Its all about your attitude-not the camera or the architecture. At 58 years old with 31 years in the business, I am having more fun now than ever. I also continue to use the 4X5 VC pretty much exclusively for my personal work and of course for the beloved HABS projects. I love the "shift" in workflow from what I do everyday.
As per the old Canon T/S lenses. (I will buy the new ones if tests confirm they are better in the corners less with less distortion etc., the rotating tilt is of very little use to me). I work very carefully, which is easy for me having learned my trade on VCs. So, I shopped for the best copies I could find, am careful in the field to not exceed their capabilities and work with the images very carefully in PS to correct the flaws. None of my high profile clients from New Mexico to Chicago, Boston or New York, including magazines or designers have ever complained about the image quality of my work from DSLRs.
-
Nice to read all that, and not sure what will this thread discussions help us with, as beginner who never used film or MF camera in the past, and no experience in photography yet, and also can't afford many expensive gear like P45+ or P65+ alongside with some great bodies like Cambo or 4x5 or 8x10 what we can get from this all these discussion points?
Now if we will say that we will not work in business with photography then many will say ignore about gear competitions and comparisons issue, but if i want to be a serious photographer and say professional and working in Photography, should i read about all the equipments are there used for photography and test them all to see which help me in my profession and get whistles or get paid out of them?
I was looking to buy that current TS 24mm, but as Canon announced to release new TS lenses soon so i just looked at one of the new TS lens, but i am looking to try to find many job in architecture and interior design photography in my country, and you know most of them are looking for large print for posters or billboard, so what i can do if i can't afford higher MF and LF bodies?
-
My usual method is to also shoot an unshifted shot, then align the two in PS. That way I have a lens-center reference for correcting distortion and CA. I have this all running as an action, so it's pretty painless after getting it set up.
Just trying to get my head around this method...and it's not working late on a friday afternoon here in the UK!
Would you mind giving a fuller explanation?
Cheers,
Marc
-
"I was looking to buy that current TS 24mm, but as Canon announced to release new TS lenses soon so i just looked at one of the new TS lens, but i am looking to try to find many job in architecture and interior design photography in my country, and you know most of them are looking for large print for posters or billboard, so what i can do if i can't afford higher MF and LF bodies?"
I actually think there is no better way to learn AP than buying and learning to use a 4x5 VC. I agree with GB about that-it teaches discipline.
Having said that, as per DSLR and posters and billboards. I doubt most of your images will be used that large. Most in fact will probably be used on websites, magazine spreads, PowerPoint presentations or notebook size design proposals. A 21 MP camera is fine for all of that and more. But you could shoot a billboard with a 5MP camera and uprez it, many have. The viewing distance is so great on billboards that fine quality is lost. As per occasionally needing a large poster? Know your clients specific needs and learn to stitch with your DSLR.
-
Thanks Alan.
Very well presented!
Presumably you note amount of shift / rise for each shot and then align the shifted image to the unshifted image by that amount.
Marc
-
I used to before I included the zero-shift frame. The unshifted frame eliminates the need for note-taking, always a bonus!
Adding a zero shift frame, great idea.
-
So you simply always align the shifted image to the far end of the shift / rise (canvas).
Really simple idea Alan.
Thank you very much for sharing this and I'll give it a go next week to see the results.
Cheers,
Marc
-
We ran out of tripods before we ran out of cameras
Edmund
-
so with all those camera angled upwards...I guess lots of p.s. work for those converging verticals.
-
so with all those camera angled upwards...I guess lots of p.s. work for those converging verticals.
Nope. Making things look pretty wasn't the point. Since we didn't have perspective correction on the Hasselblad it wasn't used on any of the cameras for this test.
-
Nope. Making things look pretty wasn't the point. Since we didn't have perspective correction on the Hasselblad it wasn't used on any of the cameras for this test.
Fair enough, but doesn't using shift affect the final image quality from these digital backs, which is of course not the case in film which it was being tested against, especially bearing in mind the end usage..architecture?
Marc
-
In the end we shot an interior of the Gare de Lyon station with a lot of signage, cast iron and glass roof etc. and an exterior of the Frank Gehry designed Centre Cinemateque Francais in the the Parc Bercy. (Forgive my French spelling if I got any of that wrong).
We used:
P65 on Hasselblad H2 with 50mm and 100mm lenses.
P45+ on Mamiya with 55mm lens.
D3x with 24mm shift but also with 24-70mm zoom lenses.
Arca Swiss Monolith 8" x10" with 240mm lens Apo Symmar and Kodak Portra 160NC.
Toyo VX125 5" x 4" with 120mm Super Symmar HM and Kodak Portra 160NC, 400NC and 100G transparency.
All lenses were set on their optimum apertures. All exposures were exactly the same. As far as possible all shots were done at exactly the same time. However, we only had four tripods so the P45+ was shot a minute or two after. Framing was matched as closely as possible. The Nikon was shot in 5" x 4" mode. Digital files were processed in Capture One but the Nikon files were also processed in Aperture. The film was scanned by Core in London on a Screen 1045i by a very good, very experienced scanner operator.
The 8" x 10" has not been scanned yet.
What have we learned so far?
1. The P65 and the P45+ seemed to just out-resolve drum scanned 5" x4" negative film. It's something of a tie with the transparency.
The Nikon with either lens was significantly behind in resolution. There was quite bad CA on the 50mm Hasselblad lens but it was slightly sharper than the Mamiya 55mm. There didn't seem any substantial resolution advantage to the P65 over the P45+.
2. General colour and tonality of the digital files seemed broadly more accurate than the film, though the film had a pleasing 'film' look to it with the transparency probably most pleasing of all. Moire/weird colour artifacts on the LED train announcement board was horrendous with the P65 and bad on the P45+. At the station there was also a green neon cross in front of a pharmacy and a neon sign above a cafe which were bad on the medium format backs with the P65 worst. These issues would be difficult to fully resolve in retouching. The Nikon had some colour artifacts on the announcement board but managed the neon pretty well. Overall the Nikon probably provided the most neutral, accurate colour. The P65 on the Hasselblad and the P45+ on the Mamiya have a significantly different colour feel between them.
3. The difference between the 160NC and the 400NC was so miniscule that for scanning purposes you might as well only shoot 400NC. (When conventionally printed there is a contrast difference equivalent to a 1/4 stop pull going from the 160NC to the 400NC but this is an extremely minor correction to make on a scanned file).
I'm off to China for a two week architectural commission but will try to add further comments when I get back. I will have the 10" x 8" scans by then.
-
Fair enough, but doesn't using shift affect the final image quality from these digital backs, which is of course not the case in film which it was being tested against, especially bearing in mind the end usage..architecture?
Marc
You are absolutely right. As you and others have pointed out it would have been better to test with a proper digital view camera set-up.
However, I couldn't get one in the time available and because I had never shot architecture with a digital camera there was a lot that I could learn by doing so even with regular medium format bodies. (I shoot architecture/travel intermittently but at a reasonably high level. My regular practice over the last ten years or so has been mostly to shoot 5" x 4" or 10" x 8" negative and print it myself. The client being supplied with 16" x 12" or 16" x 20" prints. (I would also sometimes shoot transparency)).
What I now need to do is shoot with a digital back and a view camera with a bunch of lenses on it's own. I don't need to compare with film or 35mm digital any further at this stage having learned a great deal from this test.
-
In this test I supplied the Mamiya with 55 shift and the new gen 24mm Nikon shift lens. Jonathan who is an experienced architecture specialist chose his angle and subject matter. We may make files with scans available when the 8x10 scans come in.
In this test we don't really have an immediate "winner", because the looks differ immediately, while the resolution differences only come into play as you scale up the magnification by a HUGE amount.
All the cameras seem to have DR sufficient for this type of imaging. I'd say the upshot is that factors other than simple sensor resolution which now come into play - primarily lens quality, and tripod stability, and Raw development and sharpening. My impression is that the Mamiya and Nikon lenses are outresolved by the sensors they are attached to. One thing Jonathan and I agree on is that the P45+/Mamiya combo is doing better than we both thought it would, both in resolution and in "look".
I'd say that what we need these days is better lenses even more than better sensors. We're going to try and match the Phase back with a high-quality view-camera lens in a reshoot.
Edmund
-
Looking forward to some crops/images from the test, as well as the results from the 8x10" film...
-
I'm excited to see the results!
But no clearly visible resolution advantage 39MP->60MP and digital->4x5" ? It would have been interesting to see how the Rodenstock/Schneider-lenses perform on the 65+ (every review was made with Mamiya/Fuji-lenses till now).
Some simple math:
4x5" @ 40cycles/mm (2000ppi) = 8000x10000pixels actual resolution = 80MP
40cycles/mm are not much and can be reached at lower contrast even with faster film - when the images were focused precisely and scanned properly?
What is the scan resolution (lower resolutions like 2000ppi will cause grain alaising and lower overall contrast at higher frequencies, better scan at 4000ppi and downsample to 2000ppi).
Anyway, thanks for the test!
-
I'm excited to see the results!
But no clearly visible resolution advantage 39MP->60MP and digital->4x5" ? It would have been interesting to see how the Rodenstock/Schneider-lenses perform on the 65+ (every review was made with Mamiya/Fuji-lenses till now).
Some simple math:
4x5" @ 40cycles/mm (2000ppi) = 8000x10000pixels actual resolution = 80MP
40cycles/mm are not much and can be reached at lower contrast even with faster film - when the images were focused precisely and scanned properly?
What is the scan resolution (lower resolutions like 2000ppi will cause grain alaising and lower overall contrast at higher frequencies, better scan at 4000ppi and downsample to 2000ppi).
Anyway, thanks for the test!
In practice, its probably not so easy to print details from 40lp/mm on a 4x5 negative. I always routinely shot (have to admit the past tense here) 4x5 at f/22 or slower to get the depth of field I wanted. At f/22 the contrast at 40lp/mm will be ~35% even for a perfect lens. Multiply 0.35 by the MTFs for film, scanner, and printer (or alternatively film, enlarging lens, and paper), and you have practically zero information at 40 lp/mm. Given this, I'm not too surprised that 39mp or 60mp digital would put up a good show relative to 4x5, but like you I'm really curious to see the results of the actual test.
-
I also suspect diffraction and depth of field as crucial.
f5.6 is a good aperture in 645-digital for this kind of work, 4x5" would need f11 to gain the same amount of depth of field (MTF for most lenses at 20lp/mm 70% just 10% worse than at f5.6 with most Fuji-lenses of the H-System >50% picture height) and 8x10" would need f22 which would cause serious diffraction problems even at 40cycles/mm - in theory...
But we have seen enough horrible film vs. digital comparisons (remember "11MP-1ds beats 6x7 Velvia"!?) and comparing 4x5" with diffraction (f22) scanned at 2000ppi somewhere in the focus plane with the sharpest spot in a 60MP-digital-file would lead to similar results.
If 4x5"/8x10" is always feasible just because it propably has more resolution is a different story...
-
I'm curious about what you guys are saying,
except I'm not comprehending,
especially with the word "cycles" being used now.
-
I'm curious about what you guys are saying,
except I'm not comprehending,
especially with the word "cycles" being used now.
One cycle = one line pair.
If you're familiar with a line pair, it's a dark line followed by a light line.
Looked at mathematically, it's a graph representing density on the vertical axis which swings low (dark line), then high (light line), like a sine wave (ideally a square wave).
Regardless of shape, this repeating pattern can be referred to equivalently as either a line pair or as a cycle (one swing low, then high, and returning to the start postion).
-Brad
-
Thanks, Bradley.
However, I don't know what a line pair is.
What are you guys saying in non-scientific terms?
-
Billy,
The Imatest.com website has a lot of great tutorials on some of this technical stuff and the zeiss website also has some good info on how to read the MTF charts which also describes this.
A dark area next to a light area in a scene would be contrast - so to simplify things for testing optics dudes just print dark regions next to light regions on a chart to simulate maximum contrast situations. They then take a picture of the test chart and look at the resulting image in the area where the dark and light touch to see how clearly the lens renders the transition. Is there a bunch of grey in the middle where the lens muddied the light and dark together (not good) or a clear step from dark to light (good)? To test micro contrast they just use very fine alternating black and white lines or line pairs to see if the lens can render a clear transition from dark to light or if it fails and just sees grey. How well the lens renders the boundary between black and white can be quantified and given a percentage which is what you see on MTF curves. The cycles or line pairs is a way to quantify how small or finely drawn the light and dark test bands are, ie how many per millimeter.
Eric
-
Thanks, Bradley.
However, I don't know what a line pair is.
What are you guys saying in non-scientific terms?
Sorry, Billy.
Like Eric says above, the whole goal of MTF testing is measure how well a lens (optical system) can copy a known pattern. The pattern and the way it is measured is intended to be representative of how real world scenes are rendered (even though they look nothing alike), and the MTF is not a complete description of a lens' performance (factors such as bokeh, curvature of field, close-focus performance and other issues are not captured by the standard single MTF curve).
Still it is a useful tool in measuring lens ability to render fine contrast (what we'd call 'detail' in an image) and coarse contrast (what we'd call microcontrast in an image).
The fine detail is represented a chart with typically 30 or 40 line pairs per millimeter (40lp/mm or 40 cycles/mm). Sometimes you'll see this written as 40 lines per millimeter, but this is incorrect--they mean 40 line pairs per mm. This means that you will ask the lens to accurately reproduce a line that is 1/80mm in width, then reproduce a strongly contrasting line immediately adjacent to it, also 1/80mm in width, again and again, repeating across the frame. This is quite difficult, and even a very good lens might only reproduce around half or even less of the available contrast of the test target.
This level of performance would be shown by a line travelling across the the middle of the graph (representing 50% performance). In practice, the lines are never perfectly straight, as it is progressively more difficult to achieve the same level of performance as you move away from the center of the lens' image circle. Lines will typically curve down, indicating poorer reproduction.
The microcontrast is represented by the 'coarse' lines, at perhaps 10-15lp/mm. Because these lines are wider, they're somewhat easier to reproduce (these lines are always higher on an MTF chart).
Hope that helps,
-Brad
-
(remember "11MP-1ds beats 6x7 Velvia"!?)
I'm glad someone else remembers that farce...
-
I'm glad someone else remembers that farce...
was not a farce.
6x7cm is about 30% of the size of a 4x5".
11mp is 33% of 33mp, which is comparable to 4x5" drumscanned film in most situations.
i compared at that time also carefull my kodak slr ( 14mp) with my snacmate5000 drumscanned 6x9cm slides.
the slr didnt reached completely velvia resolution at 6x9cm filmsize, but nearly.
-
Another important thing: one cycle (line pair) needs two pixels, when you have an image with 1000pixels height you can resolve (theoretically) 500 cycles, in reality, the contrast at 500 cycles already has to be zero to avoid alaising (AA-filter) and you lose additional resolution by bayer-filtering (you usually don't photography lines that align with your pixels and are black/white, do you?). Post-processing can generate very high results when measuring those test patterns (horizontal/vertical black lines on white ground) especially when using sharpening tools.
Sorry to come up with these numbers again, but when you don't reach the resolution of a 1ds with a 6x7 (56x72mm)-Velvia, something has gone terribly wrong! To match a perfect (!) 1ds-file you just need to resolve 25lp/mm (or cycles)!
A 35mm-slide would just contain 2MP at this frequency!
My M8 (sensor is 60% smaller) is very similar to my MP loaded with Velvia resolution-wise (depends on contrast), although the "look" is entirely different.
Here is one of the few comparisons with a decent scanner (Imacon @8000ppi downsampled to 4500ppi): http://www.boeringa.demon.nl/menu_technic_..._resolution.htm (http://www.boeringa.demon.nl/menu_technic_ektar100_resolution.htm)
The lines beside the "20" are 43lp/mm. And there should be MF-(maybe even LF?)lenses that reach the MTF of a Minolta 1.7 50mm...
Bloody theory, were are the 8x10"-scans?
-
Bloody theory, were are the 8x10"-scans?
I guess we'll see them when Jonathan gets back in a couple of weeks.
Edmund
-
was not a farce.
6x7cm is about 30% of the size of a 4x5".
11mp is 33% of 33mp, which is comparable to 4x5" drumscanned film in most situations.
i compared at that time also carefull my kodak slr ( 14mp) with my snacmate5000 drumscanned 6x9cm slides.
the slr didnt reached completely velvia resolution at 6x9cm filmsize, but nearly.
A very nice MF lens (pentax 67 100 macro for example) should be able to render velvia's 80+lp/mm where as a non digital 4x5 lens should be in the 40-50 lp/mm range. I have 40x50's from Pentax67 + 100 macro scans that the original 1ds couldn't touch and even the 1ds II doesn't match it.
-
Any news?
-
Yeh he decided to buy a Nikon D3x...:+}
Snook
-
Yeh he decided to buy a Nikon D3x...:+}
Snook
We both already had Nikon D3x when we did the test. Jonathan seems to have disappeared, probably on a work trip, I was thinking about asking him whether he wanted to buy my DB
Edmund
-
Another important thing: one cycle (line pair) needs two pixels, when you have an image with 1000pixels height you can resolve (theoretically) 500 cycles, in reality, the contrast at 500 cycles already has to be zero to avoid alaising (AA-filter) and you lose additional resolution by bayer-filtering (you usually don't photography lines that align with your pixels and are black/white, do you?). Post-processing can generate very high results when measuring those test patterns (horizontal/vertical black lines on white ground) especially when using sharpening tools.
Sorry to come up with these numbers again, but when you don't reach the resolution of a 1ds with a 6x7 (56x72mm)-Velvia, something has gone terribly wrong! To match a perfect (!) 1ds-file you just need to resolve 25lp/mm (or cycles)!
A 35mm-slide would just contain 2MP at this frequency!
My M8 (sensor is 60% smaller) is very similar to my MP loaded with Velvia resolution-wise (depends on contrast), although the "look" is entirely different.
Here is one of the few comparisons with a decent scanner (Imacon @8000ppi downsampled to 4500ppi): http://www.boeringa.demon.nl/menu_technic_..._resolution.htm (http://www.boeringa.demon.nl/menu_technic_ektar100_resolution.htm)
The lines beside the "20" are 43lp/mm. And there should be MF-(maybe even LF?)lenses that reach the MTF of a Minolta 1.7 50mm...
What you're forgetting is that film has a built-in dithering pattern due to the grain structure that digital does not. It's similar to the difference between an inkjet printer and a continuous-tone dye-sub printer. The inkjet has to have a much higher nominal resolution (~1440 DPI) to match the ability of a 300 DPI dye-sub to comparably render continuous tones and smoothly blend color gradients. The inkjet can thoroughly kick the dye-sub's ass when it comes to printing fine high-contrast black-on-white subject matter such as text or your resolution test patterns, but when it comes to rendering real-world images, the resolution gap between them is insignificant due to the loss of resolution caused by the inkjet's need to dither a pattern of ink droplets to print colors that do not exactly match an ink color.
The conclusions you draw from your scan comparison are bogus because you are only considering scans of high-contrast test patterns where the dithering effect of the film grain affects resolution the least. If you were to make a low-contrast RGBCMY version of your test chart and photograph and scan that, the results would favor film much less.
-
This may sound like a simplistic question, but what of say a higher-end DSLR + (to pick one) Mamiya TS Auto Bellows M645 and choice of M645 glass a as budget option (vs. MF) as opposed to a dedicated CaniKon TS lens?
-
Excellent link provided. Thanks
Another important thing: one cycle (line pair) needs two pixels, when you have an image with 1000pixels height you can resolve (theoretically) 500 cycles, in reality, the contrast at 500 cycles already has to be zero to avoid alaising (AA-filter) and you lose additional resolution by bayer-filtering (you usually don't photography lines that align with your pixels and are black/white, do you?). Post-processing can generate very high results when measuring those test patterns (horizontal/vertical black lines on white ground) especially when using sharpening tools.
Sorry to come up with these numbers again, but when you don't reach the resolution of a 1ds with a 6x7 (56x72mm)-Velvia, something has gone terribly wrong! To match a perfect (!) 1ds-file you just need to resolve 25lp/mm (or cycles)!
A 35mm-slide would just contain 2MP at this frequency!
My M8 (sensor is 60% smaller) is very similar to my MP loaded with Velvia resolution-wise (depends on contrast), although the "look" is entirely different.
Here is one of the few comparisons with a decent scanner (Imacon @8000ppi downsampled to 4500ppi): http://www.boeringa.demon.nl/menu_technic_..._resolution.htm (http://www.boeringa.demon.nl/menu_technic_ektar100_resolution.htm)
The lines beside the "20" are 43lp/mm. And there should be MF-(maybe even LF?)lenses that reach the MTF of a Minolta 1.7 50mm...
Bloody theory, were are the 8x10"-scans?
-
"The conclusions you draw from your scan comparison are bogus because you are only considering scans of high-contrast test patterns where the dithering effect of the film grain affects resolution the least. If you were to make a low-contrast RGBCMY version of your test chart and photograph and scan that, the results would favor film much less"
Yes, film is much more contrast-dependent than digital but this test is also partly in favor of digital, because it's not color (a bayer-filtered sensor interpolates color information) and the line pattern align with the pixel-structure.
There are also some interesting degrain-algorithms coming from the movie-industry (I'm not sure if we will ever see them as consumers):
http://www.digitalfilmcentral.com/docs/ARR...s_April2009.pdf (http://www.digitalfilmcentral.com/docs/ARRI_News_April2009.pdf)
My point was that most comparisons are based on crappy scans with inferior processing methods - otherwise we wouldn't come up with the conclusion that 35mm has 2 Megapixels resolution (1ds =6x7)!
Anyway, were are the scans!?!