The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

IQ180/P45/8x10/4x5 camera test

timparkin

Member
First of all I'd like to thank the people on the forum who have helped make this test work. We took on board nearly all of the suggestions and ended up using the IQ180 on four different camera systems (Linhof Techno, Alpa SWA, Cambo Wide and Phase 645DF). The different platforms each performed well with their respective lenses (40mm Digaron W, Schneider 35mm XL, 45mm Phase f/2.8 AFD) and we also used a Digraon W 70mm and Phase 80mm f/2.8 D.

All of the systems acheived the maximum resolution the sensor was capable of although the Digaron lenses proved to have the greatest contrast and judging by the moire beyond the sensor resolution they were easily sharp enough to cope with a 200Mp sensor, never mind an 80Mp one.

The post processing of the photos was handled as recommended by a few readers here and by the Phase One regional support team and Joe Cornish (landscape photographer). Obviously the IQ180 files have had to be uprezzed to compare with 4000dpi scans of 8x10 film but hopefully it hasn't been done too insensitively. We also tried adding noise to the IQ180 files to compare them with film scans.

I've added my own commentary in a separate article on the homepage under Editor's Conclusion - in summary, I want one but I want to keep my film cameras too :)

http://www.landscapegb.com/issues/lgb-0028/

Tim parkin
http://www.timparkin.co.uk
 

Jack

Sr. Administrator
Staff member
Very nice work Tim, thank you first for the incredible effort and then for sharing it with us!

One thing I would like to see added, is the test group's combined impressions of viewing say 30x40 or 40x50 prints made from each file. Ideally this would include traditional wet output as well as digital from the film scans. Obviously it's not necessary to make actual 30x40's, but perhaps pick one section, like maybe a portion of the Imatest target and one of the pieces of camera equipment, and print up comparative 8x10's from it. While it won't be "scientific," I think hearing from actual photographers involved on what their honest impressions of the perceived differences are would be interesting and maybe add a nice topping to an already excellent test.

Kudos,
 

Shashin

Well-known member
I am always fascinated by folks who try to count the number of angels dancing on the head of a pin. Great test. I have nothing against the test at all. But to look under a microscope for all possible structure than will never be reproduced or even seen, is like making 10m circles buy using Pi to the 20th place.

Why do photographers take resolving power so personally? This is what the argument comes down to--my camera is better than yours and a fear of being caught out with "inferior" equipment. Personally, I worry about whether my images are interesting. Resolution will not add interest. (Face it, that is the genesis of the film/digital discussion. It has been know for a very long time all these system produce great images.)

But it is a great test. I have no idea what practical information you got from it--I bet the images made by each of these systems still look just as good.
 

georgl

New member
Is it just me or are the Portra 400 scans particular noisy/grainy compared to Portra 160?
I'm asking because I currently try to decide which film to choose and scanned Ektar 100 vs. Portra 400 on my Eversmart, you can see the unfiltered/unsharpened result below. The difference in Mr. Parkins test seems extreme!?


I am always fascinated by folks who try to count the number of angels dancing on the head of a pin. Great test. I have nothing against the test at all. But to look under a microscope for all possible structure than will never be reproduced or even seen, is like making 10m circles buy using Pi to the 20th place.
It's a simple method to find out if the scanner is seriously limiting the potential of final IQ - like it happens often with these comparisons.
 

timparkin

Member
I am always fascinated by folks who try to count the number of angels dancing on the head of a pin. Great test. I have nothing against the test at all. But to look under a microscope for all possible structure than will never be reproduced or even seen, is like making 10m circles buy using Pi to the 20th place.

Why do photographers take resolving power so personally? This is what the argument comes down to--my camera is better than yours and a fear of being caught out with "inferior" equipment. Personally, I worry about whether my images are interesting. Resolution will not add interest. (Face it, that is the genesis of the film/digital discussion. It has been know for a very long time all these system produce great images.)

But it is a great test. I have no idea what practical information you got from it--I bet the images made by each of these systems still look just as good.
Well the microscope was the closest I could get to seeing what a perfect darkroom enlarger could image. People still use them. It also confirmed whether it was worth trying an 8000dpi scanner. It also confirmed that film *can* record more details than you can scan with a drum scanner. It also proved that Mamiya medium format lenses can resolve over 100 lines per mm. Personally I think a good enlarger set up should be able to get a little more detail out of the medium format than the drum scan (I've yet to borrow a 12,000dpi scanner - that comes in a future personal test).

Resolution can add interest. People see a veracity in a very finely detailed large print and can engage with it in a different way. the whole Tableau movement in contemporary fine art photography relies to some extent on resolution .

Also - the test talks about resolution but that is only one aspect. The images are also compared for tonality and colour which I think it more important personally. However, the 20x24 prints that were made looked different comparing DSLR's with MFDB's and LF - hence resolution made things look different. I've recently made some 20x24 prints from 35mm film and, if resolution isn't important, the they would have looked just as good. They didn't, they looked pretty awful in comparison.

As for what practical information - how about just the information about diffraction and optimum apertures? The difference in moire dependent on aperture? The quality of the A900 colour? The difference in rendering of detail from film to digital? If you are using 8x10 for resolution, how aperture affects it's advantage over other formats? How chlorophyll challenges digital sensors and to test using foliage to asses digital colour?

Anyway - it's more than resolution, hence why used real world images.

p.s. I worry about whether my images are interesting *and* the quality of them.
 

timparkin

Member
Is it just me or are the Portra 400 scans particular noisy/grainy compared to Portra 160?
I'm asking because I currently try to decide which film to choose and scanned Ektar 100 vs. Portra 400 on my Eversmart, you can see the unfiltered/unsharpened result below. The difference in Mr. Parkins test seems extreme!?
It's a trade off of detail and sharpness. I scanned at an aperture of 6 micron as a 13 micron scan will lose some fine detail. However, scanning at 6 micron does cause a lot more noise. However, scanning Portra 160 at 6 micron seems to result in some impressive scans, not much noise reduction needed at all.

I've shot a fair bit of 6x17 of army groups and Portra 160 has produced the sharpest, smoothest images. As an example, one of the shots of a group of 400 had a guy in the front row with an analogue watch on which not only could you see the second hand but you could also make out the roman numerals that indicated the hours.

Personally I use Portra 160 for as much as possible unless high dynamic range is needed or high iso. I can expose Portra 160 with the shadows placed at -2 however I can shoot the Portra 400 with the shadows places at -3 and get equivalent results. Hence Portra 400 can really be rated at 800 and still get equivalent results to Portra 160. Portra 400 does have a bit of colour crossover, needs some greens in the shadows and some yellow/green removed from the highlights. Portra 160 colour is just astonishingly good..

Tim
 

Shashin

Well-known member
Is it just me or are the Portra 400 scans particular noisy/grainy compared to Portra 160?
Seriously? Yes, a 400 speed film has more/larger grain than a 160 speed film. Grain size is one attribute that increases speed/sensitivity.
 
...One thing I would like to see added, is the test group's combined impressions of viewing say 30x40 or 40x50 prints made from each file. ...
+1, this would be the icing on the cake for the test. I could contribute a few large digital prints to 40x50, but would be limited to a few and unmounted (as must be shipped from HK).
 

Shashin

Well-known member
As for what practical information - how about just the information about diffraction and optimum apertures? The difference in moire dependent on aperture? The quality of the A900 colour? The difference in rendering of detail from film to digital? If you are using 8x10 for resolution, how aperture affects it's advantage over other formats? How chlorophyll challenges digital sensors and to test using foliage to asses digital colour?

Anyway - it's more than resolution, hence why used real world images.

p.s. I worry about whether my images are interesting *and* the quality of them.
Tim, thanks for your reply. I just don't see it.

For example, the information on diffraction and optimal apertures is not really practical. If f/22 is the optimal aperture for resolving power, what difference would it make if it gave me too much or not enough depth of field? No imaging system can reproduce color accurately and to break it into film and digital does not make any sense as both media have lots of variation in spectral response and color management in themselves. And I am not going to shoot an 8x10 image in regard to another format.

I too am interested in quality, but that is a broad category. If I want grainy, low rez gritty images, I will use TriX in a 35mm camera and mess with the developing. And the quality will be perfect and so will the resolution--resolution is not an absolute frame in which to judge (I know, I know. I am a Getdpi member and I don't think the more resolution the better, but it take all sorts). (And print size will not affect the image--standard viewing conditions and the theory behind it work really well.) I take imaging systems and use them for their strengths and the results the process gives me. I understand how imaging systems work so I understand the compromises of my choices. An 8x10 is not "better" than a 35mm camera, it is just different.

(BTW, A microscope really does not make an ideal enlarger. I have work in darkrooms a long time and professionally--remember dye transfer? I am not going to get the detail out of a print like you can get the detail out of a microscope (I also work with microscopes). That is not a real world system.)

I think you test is a fine piece of work (it really is impressive), but it is just another in the film/digital debate. The only "practical" thing it showed is all of those systems work really well and make really nice images. But we already know that. It does quantify the differences in the system, in that regard, it is interesting, but some of your choices have already thrown in unrealistic biases.

I am sorry if I sound like the Grinch that stole Christmas, but I am really disturbed by how photographers beat each other up over technical trivia. This film/digital thing was/is really ugly--this is not a comment of what you did, but the poison that has been floating around the intertrons for a long time. Unfortunately, your test will just be more fodder in the ongoing war.
 

timparkin

Member
Tim, thanks for your reply. I just don't see it.

For example, the information on diffraction and optimal apertures is not really practical. If f/22 is the optimal aperture for resolving power, what difference would it make if it gave me too much or not enough depth of field? No imaging system can reproduce color accurately and to break it into film and digital does not make any sense as both media have lots of variation in spectral response and color management in themselves. And I am not going to shoot an 8x10 image in regard to another format.
I'm afraid you haven't seen the point of this comparison at all. I didn't break it down into film vs digital. I chose representative imaging systems of the major types. You have just interpreted this as a film vs digital comparison.

As far as colour reproduction goes, you are correct, you can't reproduce it accurately. However, you can compare the different reproductions and make an aesthetic choice. Hence picking the two major types of film that affect tonality. You could easily call this an "IQ180 vs Phase P45" comparison or a "Medium format vs DSLR" comparison. Or a "Drum scan vs Epson flatbed" comparison. Lots of these comparisons exist in the test results but you've chosen to characterise this as film vs digital.

And if f/22 is the optimum aperture, you know where the compromises are. e.g. A lot of people say 'I'll stop down an extra stop to make sure I have enough depth of field'. If you know what effect this may have on image quality you may decide not to. Also, people may choose to shoot 8x10 over 4x5 because of resolution but unless they know the relationship based on aperture, they may be getting no extra resolving power. That's useful information in my book

I too am interested in quality, but that is a broad category. If I want grainy, low rez gritty images, I will use TriX in a 35mm camera and mess with the developing. And the quality will be perfect and so will the resolution--resolution is not an absolute frame in which to judge (I know, I know. I am a Getdpi member and I don't think the more resolution the better, but it take all sorts). (And print size will not affect the image--standard viewing conditions and the theory behind it work really well.)
Judging print size by standard viewing conditions misses real world behaviour. People do walk right up to prints and check out the detail. Some people do have better vision than 20/20.

I take imaging systems and use them for their strengths and the results the process gives me. I understand how imaging systems work so I understand the compromises of my choices. An 8x10 is not "better" than a 35mm camera, it is just different.
Correct and I've just helped qualify what the difference is to people. If you haven't shot 8x10 or 4x5 before and only learned about it from what is available on the internet, you will have an incomplete idea about it's merits. It sounds like you have already shot with all of these platforms so you know the answers.


(BTW, A microscope really does not make an ideal enlarger. I have work in darkrooms a long time and professionally--remember dye transfer? I am not going to get the detail out of a print like you can get the detail out of a microscope (I also work with microscopes). That is not a real world system.)
Agreed, however it is representative of what information is actually stored on film and hence an indication of what may be retrieved.


I think you test is a fine piece of work (it really is impressive), but it is just another in the film/digital debate. The only "practical" thing it showed is all of those systems work really well and make really nice images. But we already know that. It does quantify the differences in the system, in that regard, it is interesting, but some of your choices have already thrown in unrealistic biases.
No - they all make different images and knowing the differences is useful if someone is thinking of spending a lot of money. I'd like to know what you mean by unrealistic biases as well please.

I am sorry if I sound like the Grinch that stole Christmas, but I am really disturbed by how photographers beat each other up over technical trivia. This film/digital thing was/is really ugly--this is not a comment of what you did, but the poison that has been floating around the intertrons for a long time. Unfortunately, your test will just be more fodder in the ongoing war.
If you know everything that is in the report already then well done. If you don't but just choose to ignore it as irrelevant then it doesn't really matter. What you are doing is choosing to say it is irrelevant to you and trying to tell everyone else it is irrelevant as well. I just don't know why you would do that? Can you honestly say that there is nothing that people may find useful in the report?

I don't really want a response to anything else as you have made your position clear but If you can enlighten us of what these unrealistic biases are I would truly appreciate it otherwise it does sound a little like name calling.

Sincerely

Tim
 

georgl

New member
Tim, thanks for your comments on Portra 160 vs. Portra 400.

Here are two samples that irritated me, based on that I would never choose Portra 400 but it seems to be a function of the Howtek (mine doesn't work properly, I cannot compare it myself) - especially the "sparkling" noise in the shadows:

HTML:
http://static.timparkin.co.uk/static/tmp/cameratest-2/studio-hassleblad/studio-hassleblad_0008_Mamiya%207%20Portra%20400.jpg
HTML:
http://static.timparkin.co.uk/static/tmp/cameratest-2/studio-hassleblad/studio-hassleblad_0009_Mamiya%207%20Portra%20160.jpg
It's extreme compared to my sample of Ektar vs. Portra 400

@Shashin
Many tests have been seriously flawed and irritated many photographers I know - a <800ppi-drum-scan of 8x10 is soft, a 11MP-Canon beats a 6x7-slide. Yes, it's still the old "film vs. digital"-debate and it's tiring - but I don't see the point of letting faulty tests dominating the internet.
 

timparkin

Member
Tim, thanks for your comments on Portra 160 vs. Portra 400.

Here are two samples that irritated me, based on that I would never choose Portra 400 but it seems to be a function of the Howtek (mine doesn't work properly, I cannot compare it myself) - especially the "sparkling" noise in the shadows:

HTML:
http://static.timparkin.co.uk/static/tmp/cameratest-2/studio-hassleblad/studio-hassleblad_0008_Mamiya%207%20Portra%20400.jpg
HTML:
http://static.timparkin.co.uk/static/tmp/cameratest-2/studio-hassleblad/studio-hassleblad_0009_Mamiya%207%20Portra%20160.jpg
It's extreme compared to my sample of Ektar vs. Portra 400
Looks like I've missed the noise reduction on the 4x5. If you look at the twenty note one that is more representative
 

timparkin

Member
Looks like I've missed the noise reduction on the 4x5. If you look at the twenty note one that is more representative
What I meant to say was the Portra 160 looks like it has noise reduction and the 400 hasn't. They were post processed separately. I am post processing them again together and will upload the whole 160 vs 400 layers for you (plus an example of different apertures whilst scanning)
 

Shashin

Well-known member
@Shashin
Many tests have been seriously flawed and irritated many photographers I know - a <800ppi-drum-scan of 8x10 is soft, a 11MP-Canon beats a 6x7-slide. Yes, it's still the old "film vs. digital"-debate and it's tiring - but I don't see the point of letting faulty tests dominating the internet.
I absolutely, agree. There have been some really awful stuff done. Lots of baloney has been said. And so this is a really fine test, not perfect, but really nice. But many (not all) are more interested in winning than learning and one side will hold this up as a "victory" and the other side will find some way of discrediting it on some technically and so the whole thing goes on.

Folks who use a particular media or system use it for the results and workflow, both of which are personal. This won't get medium-format shooters working in 8x10 and MFD photographers are not going start selling their equipment. Prints from any of these systems will be fine.

I am a big fan of the variety of formats and processes in photography. Just as I like routers and molding planes, I am happy to shoot a MFD and a Lomo. I am finding that photography is becoming more a technical exercise where resolving power is king in an unbending absolute scale where photography is actually a bunch of relative and subjective criteria that impact an image in different ways and the photographer needs to learn these compromises in order to create an image suitable for the subject. I mean, seriously, what are we, the US Congress? (I, for one, need to get things done.)

This is just how I feel about this stuff. I think the test is great. I just don't understand the competitive nature among photographers. I find it sad because I think it has been destructive for the field.

Whirled Peas...
 

timparkin

Member
What I meant to say was the Portra 160 looks like it has noise reduction and the 400 hasn't. They were post processed separately. I am post processing them again together and will upload the whole 160 vs 400 layers for you (plus an example of different apertures whilst scanning)
I've checked the original files and depending on how you scan it there can be a big difference. However, if you want smooth results, you can get them with both but Portra 160 does end up sharper. I've zipped up the files for you.

http://static.timparkin.co.uk/static/tmp/portra.zip

Tim
 
Top