IQ180 and 8x10 assessment

[timparkin]

Hi Folks,

I'm about to rerun some of the testing of the 8x10 and the IQ180 and I wanted to start a thread to get help in making a test protocol. This is really a comparitive and subjective test but here is what I have planned so far

1) IQ180 + Alpa SWA and Linhof Techno + 40 and 70 Digaron W lenses

2) Ebony 810 and Toyo 810MII + Nikkor 240 or Fujinon 240A and a 360 Sironar S hopefully

3) Ebony 45SU + 180 Sironar S and a Shneider or Rodenstock 90

We're also going to fire off a Mamiya 7 + 55mm and 80mm and a bunch of top end DSLRs (sony,canon,nikon) and also test a Phase 45+

These will be shot on an infinity target and also a fixed target with repositioning to fill frame (100m or so away).

8x10 will be shot using a 5 series Gitzo plus BH55 or Manfrotto 405 geared head plus a Gitzo 3 series as a front support.

Film used will probably be Fuji Velvia and Fuji Pro160S (I might buy some Ektar but I figure we're probably lens limited and I have a bunch of Pro160S I can use already).

Film will be scanned on an Epson then on a Drum Scanner (Howtek 4500) and then a sample of it will be scanned on a Heidleberg Tango 11,000 dpi and finally we'll make a print from it (ciba and ra4) perhaps only a section depending on cost.. ? and an inkjet and get some non-photographers to assess the results.

We'll also verify the detail by photographing the work on a microscope (see latest reply on the previous post for a sample)

I was going to use taking aperture of f/22 for the 8x10 f/16 for 5x4 and f/5.6 for the Phase (let me know if I've worked these out wrong - I made a spreadsheet and then lost it)

I was hoping to also run a side test with the 5x4 and Linhof techno with the cameras on full shift and tilt to see how movements affects things. I was pondering about buying some nice textiles with stripes and coloured specks to throw around at various distance to have a resolution target (some indian sari fabrics would be wonderful for it!).

Any thoughts?

[Ken Lee]

1) When testing outdoors, changes in lighting can introduce unwelcome variables, and force us to make different exposures. Having to adjust things in haste can also make us error-prone. So if you can test indoors, you can banish those issues.

2) Some lenses are corrected for infinity (Nikkor M), while others are corrected for 1:5 (Fujinon A) or other ratios, like 1:10 for the Sironar S. These influences don't affect normal prints made by typical shooters, but if you're using a Microscope and scanning at many thousands of SPI, you might want to test each lens at close range and at infinity.

3) Some lenses exhibit focus shift when they are stopped down. It might be helpful to use a strong loupe and focus the image at the actual taking aperture, rather than wide open.

[Oren Grad]

Assuming you're not using adhesive or vacuum film holders, unfortunately, you should should do multiples of each film test - at least three - to get some sense of how much variation is introduced by slop in the film holders (or frame-to-frame flatness variation in the rollfilm camera).

You should check beforehand to assure that the GG is properly aligned in the specific view cameras you're going to use, and that the RF is properly aligned in the Mamiya 7 if you use that. With the Mamiya, during the tests, you might run a focus bracket to be sure. Given the limitations of the sort-of-live-view in the IQ and the absence of it in the P45+, you may want to run a focus bracket on the digital captures as well.

By all means throw in some fabrics or other targets with regular patterns, and add evaluation of aliasing as another performance parameter.

"Infinity" is tricky when comparing systems with such disparate "format" sizes and focal lengths. I would suggest something at studio range as well as something at middling distance outdoors, taking into account Ken's caveat re changing conditions outdoors.

Before making the test exposures, perch a laser pointer or other sensitive vibration indicator on top of each camera/lens/tripod setup and make sure that there is no detectable camera shake induced by the cable release or other shutter release action.

[cosmicexplosion]

shoot both cameras at the same time to avoid different lighting conditions.

[TheDeardorffGuy]

I really hate to ask this because you have obviously thought long and hard on this project. But Why? What are you trying to prove? Are'nt all those cameras built square? Why not make a contact print to evaluate the results? Introducing digital scanning brings a new factor that I do not feel was designed into the taking lens in the first place. Just curious. I tested dozen of lenses in the 80s and found I could get a great print from lenses that had the worst reputation. Example: 10-18-24 Turner Reich. They have huge focus shift. The instructions tell you that. refocus and they are nearly Artar sharp. again just curious. I photograph trains and old farm equipment, what do you shoot?

[Oren Grad]

Why not make a contact print to evaluate the results?

Because the question under investigation is the strengths and limitations of the different image capture methods when the desired output is a very large print.

[Rory_5244]

This sounds like a lot of work! Thank you, Tim, for pursuing this irrespective of whatever the results may be.

[mdm]

Can you try some ASA100 B&W film such as Ilford Delta 100 or Kodak TMX? I would back a 210 Sironar S on 5x7 to get very close to the best an 8x10 can do. Flatter film.
I can donate a box of Delta 100 if you want it (ordered from Harman Direct).

[Robert Jonathan]

Wow, looks like an awesome test. Good luck with everything.

Can you try some Provia 100F and the new Kodak 160 neg as well?

Just throwing that out there. :)

[TheDeardorffGuy]

After I posted my comment I thought "BIG PRINT". The largest I ever did was a 48x60 on a custom built horizontal room enlarger I designed for a customer. There was some grain at a viewing distance of 2 ft. But at 8 feet none. But with printing methods and scanners now it should be nearly "3D". Good luck



Because the question under investigation is the strengths and limitations of the different image capture methods when the desired output is a very large print.

[John NYC]

Awesome.

<<getting popcorn>>

[Asher Kelman]

Great! Will follow with anticipation.

But do you have the best lenses? How did you make your choice for the 8x10?

Asher

[Nathan Potter]

I'd be inclined to place a resolution target in the center of the field of view for all the images. This would provide a target from which analytical data can be gathered directly from the final print. The black and white EV values can be measured with the densities recorded on film for an assessment of processing consistency for the film and - well I'm not sure how you would assess the digital exposure consistency - have to think about that a bit.

Simply print out a copy of the US Air Force 1951 target at the appropriate size. Seems if you're shooting at 100 meters or so and using a 240 mm lens the demagnification factor is about 40X so to produce a 10 um line on film would require a 400 um line on paper at 100 meters. All very doable. For consistency using different focal lengths you would scale the size of the target to end up with about a 10 um line at the image plane. Hopefully you would cover at least the 5 um to say 200 um linewidth sizes at the film and sensor plane using the USAF or equivalent resolution target .

Indeed, stability of the setup is quite crucial. Think about the laser pointer being reflected off part of the white target at 100 meters for a critical assessment of vibration.

Nate Potter, Austin TX.

[B.S.Kumar]

Tim,

If you have access to roll film holders for the Alpa and the Techno, you might consider shooting those as well. And if you can get hold of an Arca RM3DL, you could shoot 4x5 film and the IQ180 on the same camera with the same lenses. While the film will be smaller, it will eliminate differences due to camera and lens combinations.

Or if the test is solely for 8x10 film, how about using a 4x5 reducing adapter, and then an adapter for the IQ180?

Thanks for taking the time and effort.

Kumar

[Frank Petronio]

I'd shoot a newspaper and a grey step wedge taped to a brick wall indoors at 25 feet. Infinity outdoors just introduces atmosphere and haze as larger variables. If a lens is any good at 25 feet then it will be good at infinity.

I'd set all the cameras as low as their tripods practically allow. Or no tripod. Whatever is going to be the most rock solid. How about ditching the ball head and just mounting to the tripod itself?

I'd get some better 8x10 lenses that match the modernity of those fancy digital lenses. At least the Sironar-S and not the 20-year old Fuji Nikkor field cameras lens. Glass haze alone could affect the results.

And I'd bracket the focus by actually racking the front in and out a mm each way. It's only three sheets.

I don't see the point of adding all these other cameras into the mix, which only confuses the issue. Let your friends run those on the side and compare those results later or as a sidebar. Otherwise you'll just have too much information and not interpret it as clearly. Rather, do the very best IQ180 and 8x10 shots, do a bracket, pick the best and follow best workflow practices through out.

Civilians won't be good judges, they'll focus on some aspect that might be irrelevant. Pick photographers who are unaware of this project. Only show them abstract crops so biases about the subject don't come into it.

The probable flaws in the LuLu contest were that the 8x10 was not optimal - flimsy tripod, poorly focused, poorly scanned. Also it was a funky target, changing light, etc. Address those and keep it really basic and simple.

OK one more variable. You're shooting the digi at f/5.6 and f/22 for 8x10. Given the same ISO between film and digital, the digi is only shooting at, say, 1/125th second versus the poor 8x10 having to shoot at 1/8th.

Make a few shots at the same shutter speed, throw a ND over the digital. Not that it proves anything practical, since the digital will always have that advantage of working at a wider aperture and shorter shutter speed... but it would be a reference as to how much the slower shutter speed degrades the image.

Finally, if you projected a laser level beam from on top of the cameras, a spotter could tell when they stop shaking.

Sorry to be so terse and bossy. Thank you for doing this and good luck!

[Robert Jonathan]

I second Frank's advice on just mounting the cameras (or just the 8x10) directly on the tripod legs/base.

I have a Gitzo G1500 5 series. I also have an RRS BH55 that I use with my DSLR.

The BH55 is a great ballhead, but when I use my Horseman 8x10 or 4x5, a camera that is probably heavier (and sexier :)) than any camera you're using in this test, I do NOT use the ballhead. The giant, heavy camera goes directly on the legs.

It's probably the next best thing to a studio camera stand.

[Ken Lee]

Tim - If you incur substantial expenses, I don't think it would be inappropriate if some of us could help out.

If you are set up for PayPal or its equivalent, I'm sure that some of us would probably be very happy to contribute $5. Not a lot of money, but if 20 or more people do it, it can help cover the cost of processing. Fortunately, you'll be doing the scanning, but all those 8x10 chromes gotta cost something to purchase and process.

If you send us a link, we can contribute anonymously. Just be sure to take down the link when your expenses have been met :p

[jb7]

Good luck with the comparison-
It looks like a lot of work, and you won't like my suggestion-

If diffraction is deemed to be a major constraint to the maximum print size for any given format, then your choice of aperture will have a major role to play. Different scene types require different depth of field, this is a real world consideration. A landscape with close up foreground stretching as far as the horizon will need to be stopped down more than a portrait, maybe, for example.

It should be possible to make the largest prints using a sharp lens around optimal to wide aperture, diffraction limitation suggests this. But this is not how most people use their cameras-

It might be good if the test could cover three different scene types- to cover the more common aperture ranges- on 8x10, perhaps an example at wide open, like around f/8, another stopped down to around optimal for the format, say f/22 or so- and another set for maximum depth of field- say, f/45 or f/64. Something measurable near the centre and edge can provide the line pairs for counting...

A lot more work, perhaps three separately run tests really- but it might serve to show up the differences in resolution provided by the same sensor, but on different scene types- determined by aperture size- in addition to the test between formats and technologies.

A single test at a single aperture might provide data that could accidentally favour one system over another when using one set of chosen parameters, and so I think it might be seen to be a valuable, if slightly inconclusive and potentially biased test-*

Sorry, just being difficult...

[mortensen]

Thanks, Tim, for taking the time and thoughts to do this. Your film comparisons and articles in general has provided me a lot of information and thoughts!

[timparkin]

Hi Folks - I've just come back from a walk (which I was actually getting paid to photograph! Bonus part of doing National Parks commissions). I'll make sure I get back to these questions in the next few hours but in the meantime, thank you very much for you suggestions. Keep them coming and I'll do my best to accomodate - Tim

[Will Frostmill]

Hi Tim,

I second Frank Petrino's suggestions above, particularly reducing the number of cameras. If I was going to add a third camera to the mix, I'd pick an inexpensive 4x5 with an inexpensive lens. Like a Calumet CC-400 with a bog-standard 150mm Caltar, on a Tiltall. Just to demonstrate how big (or small!) the gap is between 'ordinary' equipment and extraordinary equipment and technique.

Will

[Nathan Potter]

I'd be inclined to place a resolution target in the center of the field of view for all the images. This would provide a target from which analytical data can be gathered directly from the final print. The black and white EV values can be measured with the densities recorded on film for an assessment of processing consistency for the film and - well I'm not sure how you would assess the digital exposure consistency - have to think about that a bit.

Simply print out a copy of the US Air Force 1951 target at the appropriate size. Seems if you're shooting at 100 meters or so and using a 240 mm lens the demagnification factor is about 40X so to produce a 10 um line on film would require a 400 um line on paper at 100 meters. All very doable. For consistency using different focal lengths you would scale the size of the target to end up with about a 10 um line at the image plane. Hopefully you would cover at least the 5 um to say 200 um linewidth sizes at the film and sensor plane using the USAF or equivalent resolution target .

Indeed, stability of the setup is quite crucial. Think about the laser pointer being reflected off part of the white target at 100 meters for a critical assessment of vibration.

Nate Potter, Austin TX.

Well of course I missed the demagnification factor by 10X. So at 100 meters a 10 um line at the image would be about .160 inch (4064 um). Using a USAF target indoors as others have suggested is difficult due to the small linewidths required at the target.

Nate Potter, Austin TX.

[timparkin]

1) When testing outdoors, changes in lighting can introduce unwelcome variables, and force us to make different exposures. Having to adjust things in haste can also make us error-prone. So if you can test indoors, you can banish those issues.


Probably not really possible unless we can hire a church hall or something similar. Daylight would be useful too. We will have multiple people there to fire off all cameras at the same time.



2) Some lenses are corrected for infinity (Nikkor M), while others are corrected for 1:5 (Fujinon A) or other ratios, like 1:10 for the Sironar S. These influences don't affect normal prints made by typical shooters, but if you're using a Microscope and scanning at many thousands of SPI, you might want to test each lens at close range and at infinity.


It's not a lens test as such and unless someone can give me an idea that this will introduce a significant error, I think shooting at 1:10 - 1:30 ratios would correlate with most real world usage? I'll keep it in mind though, thanks.



3) Some lenses exhibit focus shift when they are stopped down. It might be helpful to use a strong loupe and focus the image at the actual taking aperture, rather than wide open.

Good point! Hard to work out point of focus when stopped down though - I'm worried this may introduce further errors. Is this likely with Sironar S lenses?

Tim

[timparkin]

Assuming you're not using adhesive or vacuum film holders, unfortunately, you should should do multiples of each film test - at least three - to get some sense of how much variation is introduced by slop in the film holders (or frame-to-frame flatness variation in the rollfilm camera).



Will a bit of thin double sided affect focus? Quite happy to add some if it makes sense. I shall probably use the Chamonix holder as it seems a snugger construction than my fidelitys (again, unless there is a noted issue with Chamonix holders). Will probably focus bracket three shots anyway.




You should check beforehand to assure that the GG is properly aligned in the specific view cameras you're going to use, and that the RF is properly aligned in the Mamiya 7 if you use that. With the Mamiya, during the tests, you might run a focus bracket to be sure. Given the limitations of the sort-of-live-view in the IQ and the absence of it in the P45+, you may want to run a focus bracket on the digital captures as well.



Difficult to check the alignment but I know I've shot 8000dpi capable images on my 45SU and 4000 capable on the Toyo 810M (see the photograph at the botton of this post).




"Infinity" is tricky when comparing systems with such disparate "format" sizes and focal lengths. I would suggest something at studio range as well as something at middling distance outdoors, taking into account Ken's caveat re changing conditions outdoors.



Agreed.. about infinity - will keep away from that one.




Before making the test exposures, perch a laser pointer or other sensitive vibration indicator on top of each camera/lens/tripod setup and make sure that there is no detectable camera shake induced by the cable release or other shutter release action.

Great idea..

[timparkin]

shoot both cameras at the same time to avoid different lighting conditions.
Agreed

[timparkin]

I really hate to ask this because you have obviously thought long and hard on this project. But Why? What are you trying to prove? Are'nt all those cameras built square? Why not make a contact print to evaluate the results?
I'm afraid I don't understand the general flow of those questions and where they are heading but..

1) "What am I trying to prove?" - nothing, just making a comparison
2) "aren't all cameras built square?" - Typically yes, apart from roundshot ones :-)
3) "Why not contact print to evalulate?" - You can't contact print an IQ180 file




Introducing digital scanning brings a new factor that I do not feel was designed into the taking lens in the first place. Just curious. I tested dozen of lenses in the 80s and found I could get a great print from lenses that had the worst reputation. Example: 10-18-24 Turner Reich. They have huge focus shift. The instructions tell you that. refocus and they are nearly Artar sharp. again just curious. I photograph trains and old farm equipment, what do you shoot?

Scanning is the only way I know of to digitise a transparency and I don't think most lenses were designed with enlargement in mind as well. They were typically designed to get the best results on film. Some old ones were designed to get the best results on metal. If the transparency or negative produced a sharp result for printing, it will produce a sharp result for scanning I would imagine (unless there is some interaction between taking lens and enlarging lens I didn't know about).

I shoot Ghurkas and trees mostly with the occasional niece...

[timparkin]

This sounds like a lot of work! Thank you, Tim, for pursuing this irrespective of whatever the results may be.

Ta! :-)

[timparkin]

Can you try some ASA100 B&W film such as Ilford Delta 100 or Kodak TMX? I would back a 210 Sironar S on 5x7 to get very close to the best an 8x10 can do. Flatter film.
I can donate a box of Delta 100 if you want it (ordered from Harman Direct).

That would be greatly appreciated and I agree Delta 100 might be a good idea. Is that a good sharp film to use (I have not don't much black and white at all)?

I don't have a 5x7 camera but will be testing on 4x5 too just to see what the lenses will put onto film (using a 150 Sironar S which should match a 210 I would imagine - we could scale up the results and give a good guess for 5x7 results I imagine)

Tim

[timparkin]

Wow, looks like an awesome test. Good luck with everything.

Can you try some Provia 100F and the new Kodak 160 neg as well?

Just throwing that out there. :)

Thanks Robert.. I think Velvia 50 has the greater on film resolution to make sure we get a good result. Surprisingly for many people, Velvia has a very similiar dynamic range to provia, it's just very difficult to get at because of the dmax. I have some Portra 400 I can use for the test which is supposedly sharper than Portra 160 (although I might try one Portra 160 just to compare as I love the stuff and would like a box of 8x10).

[timparkin]

Great! Will follow with anticipation.

But do you have the best lenses? How did you make your choice for the 8x10?

Asher

I'm hoping to use Sironar S lenses as from my experience and from the MTF's I've seen they are exceedingly sharp and contrasty. Does anybody have any other non-esoteric suggestions? I may be able to borrow from a UK supplier.. The choice of lens focal length was just through looking at the standardised format small dimension to ensure a similar aspect ratio (i.e. pretend we are cropping everything square and then compare dimensions).

[timparkin]

I'd be inclined to place a resolution target in the center of the field of view for all the images. This would provide a target from which analytical data can be gathered directly from the final print. The black and white EV values can be measured with the densities recorded on film for an assessment of processing consistency for the film and - well I'm not sure how you would assess the digital exposure consistency - have to think about that a bit.

Simply print out a copy of the US Air Force 1951 target at the appropriate size. Seems if you're shooting at 100 meters or so and using a 240 mm lens the demagnification factor is about 40X so to produce a 10 um line on film would require a 400 um line on paper at 100 meters. All very doable. For consistency using different focal lengths you would scale the size of the target to end up with about a 10 um line at the image plane. Hopefully you would cover at least the 5 um to say 200 um linewidth sizes at the film and sensor plane using the USAF or equivalent resolution target .

Indeed, stability of the setup is quite crucial. Think about the laser pointer being reflected off part of the white target at 100 meters for a critical assessment of vibration.

Nate Potter, Austin TX.

I got your second comment as well and will see what I can do to include one but this is mostly about comparative results so we can remain subjective. Looking at resolution targets and comparing bayer convolution with fine contrast detail from LF results is fright with differences of interpretation and I would rather not go there. If it's easy to include, I'll give it a go though. Agree with laser pointer...

[timparkin]

Tim,

If you have access to roll film holders for the Alpa and the Techno, you might consider shooting those as well. And if you can get hold of an Arca RM3DL, you could shoot 4x5 film and the IQ180 on the same camera with the same lenses. While the film will be smaller, it will eliminate differences due to camera and lens combinations.

Or if the test is solely for 8x10 film, how about using a 4x5 reducing adapter, and then an adapter for the IQ180?

Thanks for taking the time and effort.

Kumar

I'd love to try a film holder on the back of these cameras - I'll have a chat with my supplier but the results will be limited by the maximum resolution achievable on film which is something I think has been tested previously?

I like your idea about using a reducing back for the IQ to measure absolute resolution of 8x10 camera/lens system! If we have time!

[timparkin]

I'd shoot a newspaper and a grey step wedge taped to a brick wall indoors at 25 feet. Infinity outdoors just introduces atmosphere and haze as larger variables. If a lens is any good at 25 feet then it will be good at infinity.

I'd set all the cameras as low as their tripods practically allow. Or no tripod. Whatever is going to be the most rock solid. How about ditching the ball head and just mounting to the tripod itself?

Yep - using the camera directly attached is a good idea although from experience, most of the flex in my system seems to come from the camera system, hence using a second tripod fixes most vibration. I'll look at doing this for the 10x8 as we have a levelling head for the 5 series tripod.



I'd get some better 8x10 lenses that match the modernity of those fancy digital lenses. At least the Sironar-S and not the 20-year old Fuji Nikkor field cameras lens. Glass haze alone could affect the results.

And I'd bracket the focus by actually racking the front in and out a mm each way. It's only three sheets.


Sironar S's if possible. the Sironar S 210 doesn't cover 10x8 though - suggestions?




I don't see the point of adding all these other cameras into the mix, which only confuses the issue. Let your friends run those on the side and compare those results later or as a sidebar. Otherwise you'll just have too much information and not interpret it as clearly. Rather, do the very best IQ180 and 8x10 shots, do a bracket, pick the best and follow best workflow practices through out.


I have enough people to make sure the other cameras won't affect the veracity of the main results. It would be silly not to take the opportunity to test the range of cameras someone may be interested in and will make the tests a lot more useful to a larger range of people (even if we don't really care much about the DSLR results personally)



Civilians won't be good judges, they'll focus on some aspect that might be irrelevant. Pick photographers who are unaware of this project. Only show them abstract crops so biases about the subject don't come into it.


Yep, you are probably right sadly - might be interesting to see how they interpret the difference between digital and analogue aesthetic though.



The probable flaws in the LuLu contest were that the 8x10 was not optimal - flimsy tripod, poorly focused, poorly scanned. Also it was a funky target, changing light, etc. Address those and keep it really basic and simple.

I've got the original 10x8's here and the main issue was that the focus was nowhere near the area that was compared (combined with the depth of field of the 8x10 system was less than the IQ180). Here is a quick shot showing a microscope photo of the number plate compared with a paint splash on the oil drum in the foreground.

http://static.timparkin.co.uk/static/tmp/810-test-focus.jpg






OK one more variable. You're shooting the digi at f/5.6 and f/22 for 8x10. Given the same ISO between film and digital, the digi is only shooting at, say, 1/125th second versus the poor 8x10 having to shoot at 1/8th.

Make a few shots at the same shutter speed, throw a ND over the digital. Not that it proves anything practical, since the digital will always have that advantage of working at a wider aperture and shorter shutter speed... but it would be a reference as to how much the slower shutter speed degrades the image.


I know where you are coming from here but in real world tests this is the advantage of the MFDB system already? I'll make a note of this and if we have time.



Finally, if you projected a laser level beam from on top of the cameras, a spotter could tell when they stop shaking.

Sorry to be so terse and bossy. Thank you for doing this and good luck!
No problem with terseness, better in many ways and I've worked online for long enough to take it for what it is :-)

[timparkin]

I second Frank's advice on just mounting the cameras (or just the 8x10) directly on the tripod legs/base.

I have a Gitzo G1500 5 series. I also have an RRS BH55 that I use with my DSLR.

The BH55 is a great ballhead, but when I use my Horseman 8x10 or 4x5, a camera that is probably heavier (and sexier :)) than any camera you're using in this test, I do NOT use the ballhead. The giant, heavy camera goes directly on the legs.

It's probably the next best thing to a studio camera stand.
I would really like a linhof tilt head :-) But I agree - I'll use a levelling head directly attached as discussed earlier.

[timparkin]

Tim - If you incur substantial expenses, I don't think it would be inappropriate if some of us could help out.

If you are set up for PayPal or its equivalent, I'm sure that some of us would probably be very happy to contribute $5. Not a lot of money, but if 20 or more people do it, it can help cover the cost of processing. Fortunately, you'll be doing the scanning, but all those 8x10 chromes gotta cost something to purchase and process.

If you send us a link, we can contribute anonymously. Just be sure to take down the link when your expenses have been met :p

Thanks Ken! I'll check out how much it's costing and get back to you :-)

I normally develop my own 10x8 and 5x4 but can't repeat this so might send them off for reassurance.

Really appreciate the suggestion and offer!

[timparkin]

Good luck with the comparison-
It looks like a lot of work, and you won't like my suggestion-

If diffraction is deemed to be a major constraint to the maximum print size for any given format, then your choice of aperture will have a major role to play. Different scene types require different depth of field, this is a real world consideration. A landscape with close up foreground stretching as far as the horizon will need to be stopped down more than a portrait, maybe, for example.

It should be possible to make the largest prints using a sharp lens around optimal to wide aperture, diffraction limitation suggests this. But this is not how most people use their cameras-

It might be good if the test could cover three different scene types- to cover the more common aperture ranges- on 8x10, perhaps an example at wide open, like around f/8, another stopped down to around optimal for the format, say f/22 or so- and another set for maximum depth of field- say, f/45 or f/64. Something measurable near the centre and edge can provide the line pairs for counting...

A lot more work, perhaps three separately run tests really- but it might serve to show up the differences in resolution provided by the same sensor, but on different scene types- determined by aperture size- in addition to the test between formats and technologies.

A single test at a single aperture might provide data that could accidentally favour one system over another when using one set of chosen parameters, and so I think it might be seen to be a valuable, if slightly inconclusive and potentially biased test-*

Sorry, just being difficult...

No I can see where you are coming from - however it turns out that the sharpest point in these systems is at the same aperture for equivalent depth of field (just about) and I can't afford to bracket focus, aperture and run multiple scenes. With the aim of minimising the issues, I'll add it to the list though. Does anybody think I'm thinking of the right aperture to test at?

Tim

[timparkin]

Thanks, Tim, for taking the time and thoughts to do this. Your film comparisons and articles in general has provided me a lot of information and thoughts!

No problems! I'm hoping this ends up useful to people and will allow them to make informed choices on platforms based on real world results.

[Frank Petronio]

In the old and new days, many studio photographers, especially shooting "down", would and do use double-stick tape to hold the middle of their large film from bowing. Just process normally and remove from the non-emulsion side post. It leaves one less variable and if you bracket the focusing physically (one in, in out, one per loupe) you'll nail one of the shots for sure.

[timparkin]

In the old and new days, many studio photographers, especially shooting "down", would and do use double-stick tape to hold the middle of their large film from bowing. Just process normally and remove from the non-emulsion side post. It leaves one less variable and if you bracket the focusing physically (one in, in out, one per loupe) you'll nail one of the shots for sure.

Will do!! :-)

[John NYC]

Thanks Ken! I'll check out how much it's costing and get back to you :-)

I normally develop my own 10x8 and 5x4 but can't repeat this so might send them off for reassurance.

Really appreciate the suggestion and offer!

I would contribute as well, so if you are having lots of out-of-pocket expenses do consider that.

[Oren Grad]

Sironar S's if possible. the Sironar S 210 doesn't cover 10x8 though - suggestions?

The 210 Apo-Sironar-S and Apo-Symmar L are specified to barely eke out 8x10 at infinity. But I don't think that's what we really want here - better to be comfortably within spec. The 210 Apo-Sironar(-W) and Super-Symmar HM are specified to cover 8x10 with some movement, but unfortunately both are long discontinued, and neither is easy to find on short notice. If 240 is OK instead of 210, the Apo-Sironar-S should be fine.

[Frank Petronio]

Ideally the 70mm on the Alpa and a 360 Sironar-S or Symmar-L on the 8x10, in a perfect world. Slightly longer than normal, sharp lenses.

[John NYC]

Ideally the 70mm on the Alpa and a 360 Sironar-S or Symmar-L on the 8x10, in a perfect world. Slightly longer than normal, sharp lenses.

I don't believe there is a 360 L.

[Oren Grad]

I don't believe there is a 360 L.

That's right. Unfortunately, 135, 240 and 360 were dropped with the L series.

[Brian K]

Why is there talk of using a 210mm lens? If you are shooting 8x10 for the purpose of a resolution comparison, then a 300 or 360mm is the more appropriate choice. Use a 300 or 360mm Sironar -S and you can't fault the optics.

It is quite obvious that the focus on the teat was noticeable off. Tim should make some tests related to the assurance that his GG and film focus on the same exact place, so unless he simply messed up on focus he needs to fix this.

Also if you are really serious about testing then Frank's notion of a newspaper or something else with small type at about 25 feet can be a good choice. To further eliminate operator imperfections, do this indoors and with studio electronic flash as the light source. This eliminates user error and flimsy tripods, wind, etc. But this shows more about what each system can do under the best of circumstances.

However realistically one really needs to see the comparison in real world situations. because wind, user error, poor focusing are part of the equation. Sorry to say Tim the poor results you got with 8x10 does not indicate view camera expertise. And the test you did indicates more about your best resolving system, which in your case is the IQ180.

[timparkin]

Why is there talk of using a 210mm lens? If you are shooting 8x10 for the purpose of a resolution comparison, then a 300 or 360mm is the more appropriate choice. Use a 300 or 360mm Sironar -S and you can't fault the optics.

We're talking about using a 210 because it's the right focal length to compare with the 40mm Digaron W. We're also going to be using a 360 Sironar S to compare with the 70mm Digaron W.



It is quite obvious that the focus on the teat was noticeable off. Tim should make some tests related to the assurance that his GG and film focus on the same exact place, so unless he simply messed up on focus he needs to fix this.

Could you tell me how to do this? If I've managed nr. 4000dpi of detail out of my 10x8 already would that imply my focus is OK?




Also if you are really serious about testing then Frank's notion of a newspaper or something else with small type at about 25 feet can be a good choice. To further eliminate operator imperfections, do this indoors and with studio electronic flash as the light source. This eliminates user error and flimsy tripods, wind, etc. But this shows more about what each system can do under the best of circumstances.

I agree but that isn't how most people will use the systems and we won't be taking photographs if it's really windy and we'll be using strong tripods as discussed.



However realistically one really needs to see the comparison in real world situations. because wind, user error, poor focusing are part of the equation. Sorry to say Tim the poor results you got with 8x10 does not indicate view camera expertise. And the test you did indicates more about your best resolving system, which in your case is the IQ180.
Ermm... I didn't do the original tests and even though the original test had problems, the 8x10 still outresolved the IQ180 (more on that later).

I would really appreciate your input but it would help if you knew who did what. Markus Zuber did the original test and I have uploaded various images to show resolution beyond 4000dpi on some 617 and 4x5 shots and nr. 4000dpi results for 8x10. The original results were fine as far as wind and stability goes but there was a focus inaccuracy and depth of field was less than the IQ180.

Tim

[carverlux]

Tim,

First - a very sincere thank you for the courage of taking this on. Not only because it is a LOT of work but the potential follow-up you'd have to deal with can indeed be daunting!

If the quest for these tests are looking to objectively evaluate the highest possible potential of each approrach - IQ180 or 8x10 - as many variables as possible must be eliminated. I would like to make 2 suggestions that will eliminate the two most troublesome variables and provide a much more even playing field for both setups:

Get away from taking pictures outside, do it indoor and expose the tests only using a studio strobe or a flash. Why?

1. Shooting a studio strobe at about 1/1000 sec will give you definitive results that are free from the influence of any discernable movement - subject, camera, lens, operator intervention, cable release bounce...etc.

2. To use the lenses of each setup at their most advantageous f/stops so the full potential can be shown, the strobe can be moved closer or farther away from the test subject so the best aperture can be used.

In my own lens testing, I use this method as dealing with the absolutely never-ending and unpredictable camera and subject movement over the years simply drove me nuts. A mild breeze or even a footstep can affect the outcome and color what should be a purely objective observation of facts. I came by this method the hard way: using my 8x10 p2 outdoors had been a constant challenge. It was hard to get it still enough to do exposures any slower than 1/60 - and I have used 1 and 2 Rail Clamps and tripods of all types of Sinar clamps - the model 1, model 2, the BLOCK rail clamp with the micrometer drive that must weigh over 8 lbs by itself. Just to keep the camera from moving in the great outdoors, I carried no less than 20 lbs of Sinar stuff!

I would NEVER attempt this test with a single-point-anchored ballhead of ANY size as per the LuLa setup simply because the camera in this scenario has no chance to truly be still enough to eliminate the effect of movement.

Hope this make sense and is helpful to your setup.

carver

[carverlux]

That's right. Unfortunately, 135, 240 and 360 were dropped with the L series.

I would suggest the sharpest lenses that can be used for this test may be from the Process world - Apo Artars, Apo Germinars, Apo Ronars and Apo Tessars. Not Dagors, Sironars or Dagors which were all optimized for maximum coverage and good sharpness.

For an 8x10 test to match a 70mm Alpa on the IQ180, a properly shuttered and aligned 19" Red Dot Artar or 480mm Apo-Ronar MC would be my vote.

carver

[aluncrockford]

I am starting to be confused as to the point of this test. Film in its very structure has a granular base and when enlarged will eventually break up into grain . Digital will enlarge up to 100 % of the capture then it will pixelate . The digital capture will always be a cleaner file as that is the nature of the system . If a 10x8 file is compared to a high res digital back up to the maximum enlargement of the digital back then the digital file will be cleaner and by logical conclusion will have a higher res . The other thing to bear in mind is the size of the chip means that the lenses are shorter and therefore have a much wider depth of field, helping the perception of sharpness . When I have done a direct comparison with the P45 and 5x4 Velvia the digital file was cleaner and produced a crisper image. The difference between 10x8 and digital is apparent when the print size goes beyond say 24x30 then 10x8 provides a far better outcome.
The point that I think needs to be made is the selection of either film or digital should be made with consideration of the feel and look of the image and not the level of sharpness as tested by a chart . In my opinion it is the sharpness of digital which I find to be less than splendid and that is why for my personal work everything is shot on film. To get too hung up on sharpness and resolution might suggest a lack of confidence in the visual quality of the images produced , the justification of the image that it is sharp is hardly the highest of recommendations

[timparkin]

Tim,

First - a very sincere thank you for the courage of taking this on. Not only because it is a LOT of work but the potential follow-up you'd have to deal with can indeed be daunting!

If the quest for these tests are looking to objectively evaluate the highest possible potential of each approrach - IQ180 or 8x10 - as many variables as possible must be eliminated. I would like to make 2 suggestions that will eliminate the two most troublesome variables and provide a much more even playing field for both setups:

Get away from taking pictures outside, do it indoor and expose the tests only using a studio strobe or a flash. Why?

1. Shooting a studio strobe at about 1/1000 sec will give you definitive results that are free from the influence of any discernable movement - subject, camera, lens, operator intervention, cable release bounce...etc.

2. To use the lenses of each setup at their most advantageous f/stops so the full potential can be shown, the strobe can be moved closer or farther away from the test subject so the best aperture can be used.

In my own lens testing, I use this method as dealing with the absolutely never-ending and unpredictable camera and subject movement over the years simply drove me nuts. A mild breeze or even a footstep can affect the outcome and color what should be a purely objective observation of facts. I came by this method the hard way: using my 8x10 p2 outdoors had been a constant challenge. It was hard to get it still enough to do exposures any slower than 1/60 - and I have used 1 and 2 Rail Clamps and tripods of all types of Sinar clamps - the model 1, model 2, the BLOCK rail clamp with the micrometer drive that must weigh over 8 lbs by itself. Just to keep the camera from moving in the great outdoors, I carried no less than 20 lbs of Sinar stuff!

I would NEVER attempt this test with a single-point-anchored ballhead of ANY size as per the LuLa setup simply because the camera in this scenario has no chance to truly be still enough to eliminate the effect of movement.

Hope this make sense and is helpful to your setup.

carver


Hi Carver - I appreciate the reasoning behind taking pictures inside with a strobe but I don't have the equipment or knowledge of how to do this (nor do my colleagues). Additionally, I would imagine most people using the equipment wouldn't either (unless they were doing studio production work, in which case they're probably better off with the IQ180 if they can afford it).

I'll definitely be using more than one tripod for the 10x8 because I have seen the same issues that you mention. I can assure you we'll be taking more than 20lb of gear for the 10x8 out :-)

Anyway - I agree it would help but I'm unable to use strobes for the test. We'll be taking photographs with multiple tripods with spikes on concrete under no wind conditions in midday light hopefully. This should be enough to get good resolution (if it isn't then most people can't get good resolution in most conditions and it will still be representative).

What I may be able to do is to do a comparitive test with the 8x10 and a P45+ at a future date if I can beg a strobe set up.

Tim

[ic-racer]

And we are doing this because Phase One does not supply MTF data or any other resolution data for this product?

[carverlux]

Hi Carver - I appreciate the reasoning behind taking pictures inside with a strobe but I don't have the equipment or knowledge of how to do this (nor do my colleagues). Additionally, I would imagine most people using the equipment wouldn't either (unless they were doing studio production work, in which case they're probably better off with the IQ180 if they can afford it).

I'll definitely be using more than one tripod for the 10x8 because I have seen the same issues that you mention. I can assure you we'll be taking more than 20lb of gear for the 10x8 out :-)

Anyway - I agree it would help but I'm unable to use strobes for the test. We'll be taking photographs with multiple tripods with spikes on concrete under no wind conditions in midday light hopefully. This should be enough to get good resolution (if it isn't then most people can't get good resolution in most conditions and it will still be representative).

What I may be able to do is to do a comparitive test with the 8x10 and a P45+ at a future date if I can beg a strobe set up.

Tim

Tim,

OK - I would have thought that strobes might have been easier for you but I should not have assumed so - sorry! In any case, for as long as you can use a shutter speed at close to the accepted "1/focal length" as possible then the risk of movement should be minimized.

BTW, all else being equal, the P45+ has NO chance against a properly exposed, processed and scanned 8x10.

carver

[timparkin]

And we are doing this because Phase One does not supply MTF data or any other resolution data for this product?

No, I'm doing it to see what it looks like... and then to share what it looks like with other people..

[timparkin]

Tim,

OK - I would have thought that strobes might have been easier for you but I should not have assumed so - sorry! In any case, for as long as you can use a shutter speed at close to the accepted "1/focal length" as possible then the risk of movement should be minimized.

BTW, all else being equal, the P45+ has NO chance against a properly exposed, processed and scanned 8x10.

carver

Neither does the IQ180. I know that, you know that but it will be interesting to see the real results and let other people know that.

[PaulSchneider]

Tim, this is really, really great. I think you can contribute a valuable perspective with a second test.

A few suggestions:

1. Try to contact a big, professional rental studio and do a controlled shoot of a complex scene with studio flashes. Maybe the might sponsor your test and offer their location for free in return of a honorable mention?

2. Print large, museum quality, under glas: Try to contact a professional print lab that specializes in large-format, highest quality diasec prints for galleries and museums. After all, what we want to know is whether 8x10 or IQ180 are to be used for huge, high-quality prints. In my view this is the last niche where these resolutions figures still mater print-wise. Of course artistic choice of aesthetics of a chosen medium are excluded from this perspective. In this case Grieger labs from Düsseldorf in Germany come to mind. They are the ones who regularly do 100 inch prints for the likes of Andreas Gursky and Thomas Struth. They are also considered to be one of the absolute best in large-size printing. Since I'm a native German speaker I could try to establish contact.

Since this interests me personally a lot too, I would offer to film a visit to their facilities in Germany or to a compareable lab in Düsseldorf (there's another one specializing in prints for Candida Höfer) and maybe they might be interested in a bit of publicity and maybe all this might lead to a few free tests prints so one can judge the results in vivo.

3. I would shoot a few sheets of every subject matter and send the results to maybe 2 different professional high end scanners. Schwarzzeit, who's a member of this forum, apparently has an ICG 380, a scanner arguably considered to be the best of the world.
Lenny comes to mind too with his Atzek.

Have fun! This will be epic!

Kind regards

Paul

[timparkin]

Tim,

OK - I would have thought that strobes might have been easier for you but I should not have assumed so - sorry! In any case, for as long as you can use a shutter speed at close to the accepted "1/focal length" as possible then the risk of movement should be minimized.

BTW, all else being equal, the P45+ has NO chance against a properly exposed, processed and scanned 8x10.

carver

1/focal length might be difficult for the 8x10 under any natural light I imagine. I'm typically working with exposure times from 1/15 to a few seconds. Out of interest, how much difference did you see between shooting under strobe and shooting under natural light?

[carverlux]

2. Print large, museum quality, under glas: Try to contact a professional print lab that specializes in large-format, highest quality diasec prints for galleries and museums. After all, what we want to know is whether 8x10 or IQ180 are to be used for huge, high-quality prints. In my view this is the last niche where these resolutions figures still mater print-wise. ...

Tim,

I agree with Paul wholeheartedly about high-quality prints. It is the final arbiter.

The only thing about printmaking now is that many professional printers scan and print which injects dubious variables into the final print. If done through "regular" enlarging, enlarging lenses, even VERY famous ones, are seldom as good as a good process lens that was used to take the picture.

So the print chain must be meticulously scrutinized to eliminate all possible detractors. Good luck!

carver

[timparkin]

Tim, this is really, really great. I think you can contribute a valuable perspective with a second test.

A few suggestions:

1. Try to contact a big, professional rental studio and do a controlled shoot of a complex scene with studio flashes. Maybe the might sponsor your test and offer their location for free in return of a honorable mention?



If I can get some strobes I will try to have a go at this.




2. Print large, museum quality, under glas: Try to contact a professional print lab that specializes in large-format, highest quality diasec prints for galleries and museums. After all, what we want to know is whether 8x10 or IQ180 are to be used for huge, high-quality prints. In my view this is the last niche where these resolutions figures still mater print-wise. Of course artistic choice of aesthetics of a chosen medium are excluded from this perspective. In this case Grieger labs from Düsseldorf in Germany come to mind. They are the ones who regularly do 100 inch prints for the likes of Andreas Gursky and Thomas Struth. They are also considered to be one of the absolute best in large-size printing. Since I'm a native German speaker I could try to establish contact.

Since this interests me personally a lot too, I would offer to film a visit to their facilities in Germany or to a compareable lab in Düsseldorf (there's another one specializing in prints for Candida Höfer) and maybe they might be interested in a bit of publicity and maybe all this might lead to a few free tests prints so one can judge the results in vivo.

3. I would shoot a few sheets of every subject matter and send the results to maybe 2 different professional high end scanners. Schwarzzeit, who's a member of this forum, apparently has an ICG 380, a scanner arguably considered to be the best of the world.
Lenny comes to mind too with his Atzek.


Now I'm happy for anybody to take a go at scanning these and I think Lenny has offered his help in doing this. If Schwarzzeit can help with a 12,000dpi scan that would be really good to get right down to the grain level on a couple of small sections just to see what is happening for high contrast, very fine edges. I have a Howtek 4500 and a friend has a Heidleberg Tango 11,000 dpi scanner. The more confirmation we can get on some of the results the better. I'm also going to scan on an Epson V750 just to see what your average consumer can manage.

As for prints - now there is a minefield. I was hoping it might be possible to hand over the sourcing of prints to the community in terms of raising a little money for them and seeing if we have resources within the forum to get these things managed (or if we can raise the money or talk Grieger into doing this - do you think you could help with this?). I can get inkjets done of a certain size but for wet prints I'm in the dark (or rather I'm not, which would be one of the issues).

Tim

[carverlux]

1/focal length might be difficult for the 8x10 under any natural light I imagine. I'm typically working with exposure times from 1/15 to a few seconds. Out of interest, how much difference did you see between shooting under strobe and shooting under natural light?

Tim,

1/15 second would not produce results that the majority of semi-skeptics can embrace as having eliminated all possibilities of movement. And honestly, nor would I. I have plenty of lens testing exposures nixed from careful consideration exactly because of this. This is the reason why I suggested using strobes / studio flashes.

The difference - at least to me - is night and day. If focused properly, you can count on 100% of your strobe exposures to be movement free, especially if the ambient light is low. With a 1/15 second or longer, my personal experience is admissability of no better than 1 out of 5 or 20%.

Sounds like Paul had the same idea as well!

carver

[timparkin]

Tim,

1/15 second would not produce results that the majority of semi-skeptics can embrace as having eliminated all possibilities of movement. And honestly, nor would I. I have plenty of lens testing exposures nixed from careful consideration exactly because of this. This is the reason why I suggested using strobes / studio flashes.

The difference - at least to me - is night and day. If focused properly, you can count on 100% of your strobe exposures to be movement free, especially if the ambient light is low. With a 1/15 second or longer, my personal experience is admissability of no better than 1 out of 5 or 20%.

Sounds like Paul had the same idea as well!

carver

Intriguing... Is this an 8x10 thing because I've shot 4x5 in a 50mph gusts gale and taken one shot in a lull, protected by my body and one when the wind gusted and unprotected. Both were equally sharp and maxed out my 4000dpi scan (I'll get some microscope pictures). I do use spikes on my tripod and ran it low to the ground. I was very surprised by the results as my goal was to measure how much worse the second photo would be. I presume I must have been unlucky/lucky - depending on your point of view..

Out of interest - if I have to take pictures in landscape, how can I get consistent sharp photographs?

[Ken Lee]

If the results will be shown over the web, why introduce printing? It just introduces a another set of variables. And it's not cheap.

This whole discussion got started because someone compared some questionable scans of some questionably exposed film and drew some questionable conclusions. Some skillful scans of some good exposures, ought to be enough.

Let's not make this into something so complicated that it never happens. :)

[Asher Kelman]

Tim, this is really, really great. I think you can contribute a valuable perspective with a second test.

A few suggestions:

1. Try to contact a big, professional rental studio and do a controlled shoot of a complex scene with studio flashes. Maybe the might sponsor your test and offer their location for free in return of a honorable mention?

2. Print large, museum quality, under glas: Try to contact a professional print lab that specializes in large-format, highest quality diasec prints for galleries and museums. After all, what we want to know is whether 8x10 or IQ180 are to be used for huge, high-quality prints. In my view this is the last niche where these resolutions figures still mater print-wise. Of course artistic choice of aesthetics of a chosen medium are excluded from this perspective. In this case Grieger labs from Düsseldorf in Germany come to mind. They are the ones who regularly do 100 inch prints for the likes of Andreas Gursky and Thomas Struth. They are also considered to be one of the absolute best in large-size printing. Since I'm a native German speaker I could try to establish contact.

Since this interests me personally a lot too, I would offer to film a visit to their facilities in Germany or to a compareable lab in Düsseldorf (there's another one specializing in prints for Candida Höfer) and maybe they might be interested in a bit of publicity and maybe all this might lead to a few free tests prints so one can judge the results in vivo.

3. I would shoot a few sheets of every subject matter and send the results to maybe 2 different professional high end scanners. Schwarzzeit, who's a member of this forum, apparently has an ICG 380, a scanner arguably considered to be the best of the world.
Lenny comes to mind too with his Atzek.


Paul,

First, what are the taking lenses that starts the chain?

Film has to be color to compare with the IQ180 and that and its development needs to be chosen and specified to give the best resolution, color and contrast. Then, when it comes to prints, one only needs to sample from about 5 places in the film to determine quality. That way one only has to print 4x5 samples and so it becomes much less expensive.

Asher

[timparkin]

If the results will be shown over the web, why introduce printing? It just introduces a another set of variables. And it's not cheap.

This whole discussion got started because someone compared some questionable scans of some questionably exposed film and drew some questionable conclusions. Some skillful scans of some good exposures, ought to be enough.

Let's not make this into something so complicated that it never happens. :)

Hence why I'm not proposing this as part of the initial work. It's simple enough to get this done afterward so don't count it out yet.

[Asher Kelman]

Tim, this is really, really great. I think you can contribute a valuable perspective with a second test.

A few suggestions:

1. Try to contact a big, professional rental studio and do a controlled shoot of a complex scene with studio flashes. Maybe the might sponsor your test and offer their location for free in return of a honorable mention?

2. Print large, museum quality, under glas: Try to contact a professional print lab that specializes in large-format, highest quality diasec prints for galleries and museums. After all, what we want to know is whether 8x10 or IQ180 are to be used for huge, high-quality prints. In my view this is the last niche where these resolutions figures still mater print-wise. Of course artistic choice of aesthetics of a chosen medium are excluded from this perspective. In this case Grieger labs from Düsseldorf in Germany come to mind. They are the ones who regularly do 100 inch prints for the likes of Andreas Gursky and Thomas Struth. They are also considered to be one of the absolute best in large-size printing. Since I'm a native German speaker I could try to establish contact.

Since this interests me personally a lot too, I would offer to film a visit to their facilities in Germany or to a compareable lab in Düsseldorf (there's another one specializing in prints for Candida Höfer) and maybe they might be interested in a bit of publicity and maybe all this might lead to a few free tests prints so one can judge the results in vivo.

3. I would shoot a few sheets of every subject matter and send the results to maybe 2 different professional high end scanners. Schwarzzeit, who's a member of this forum, apparently has an ICG 380, a scanner arguably considered to be the best of the world.
Lenny comes to mind too with his Atzek.


Paul,

First, what are the taking lenses that starts the chain? Next, what's the target size at which judgement will be made; 4x, (32x40) or what? It should be at the useful limit of that film and lens system. Then the Phase One IQ180 digital file would be printed at that size too.

Film has to be color to compare with the IQ180v and that and its development needs to be chosen and specified to give the best resolution, color and contrast. Then, when it comes to prints, one only needs to sample from about 5 places in the film to determine quality. Printing giant is too expensive for a test like this.

Asher

[carverlux]

Intriguing... Is this an 8x10 thing because I've shot 4x5 in a 50mph gusts gale and taken one shot in a lull, protected by my body and one when the wind gusted and unprotected. Both were equally sharp and maxed out my 4000dpi scan (I'll get some microscope pictures). I do use spikes on my tripod and ran it low to the ground. I was very surprised by the results as my goal was to measure how much worse the second photo would be. I presume I must have been unlucky/lucky - depending on your point of view..

Tim,

It all follows the law of physics - mass, center of gravity and distribution etc. 4x5 cameras are dimensionally smaller and their bellows extension significantly shorter so it is conceivable that you can knock off exposures in 50mph gusts if your camera is secured well enough, your camera & tripod combo has sufficient mass and is low enough to the ground - and you are very lucky.

With an 8x10 camera, using a monorail outside is completely hopeless as I already described. Even the very best flatbed 8x10 folding camera needs a fair bit of help such as a large tripod head platform (many US tripods made for 8x10 have platforms that measure 6"x8"), and a potential brace for the lens depending on its weight and how far it is extended to make things stable enough. You also know that the 8x10 bellows is much bigger and extends much farther so it acts like a sail in the wind and contributes to camera movement like you would not believe.

Again, on the topic of running the ultimate comparison test, a 1/15 second exposure in the outdoors will just raise questions as persons familiar with using 8x10 or larger cameras will easily point to that as a disadvantage stacked against the 8x10 which can easily contribute a skewed outcome. This is the reason why I scratched my head when I saw the test setup in LuLa - I wonder why they did what they did in the setup.

I know 1/focal length does not sound possible but with HP5 at ISO 400, you can use 1/400 at f/16 on a perfectly sunny day based on the Sunny 16 rule. And it just so happens there are plenty of VERY sharp lenses for 8x10 are in the 400mm range - the 16½" Red Dot Artar, the 420mm Apo-Ronar just to name a couple. If you want less grain and slightly higher resolution films like Ektachrome or Ektar at ISO 100, then you may not have a choice but to move the test indoors.

carver

[timparkin]

It all follows the law of physics - mass, center of gravity and distribution etc. 4x5 cameras are dimensionally smaller and their bellows extension significantly shorter so it is conceivable that you can knock off exposures in 50mph gusts if your camera is secured well enough, your camera & tripod combo has sufficient mass and is low enough to the ground - and you are very lucky.

With an 8x10 camera, using a monorail outside is completely hopeless as I already described. Even the very best flatbed 8x10 folding camera needs a fair bit of help such as a large tripod head platform (many US tripods made for 8x10 have platforms that measure 6"x8"), and a potential brace for the lens depending on its weight and how far it is extended to make things stable enough. You also know that the 8x10 bellows is much bigger and extends much farther so it acts like a sail in the wind and contributes to camera movement like you would not believe.

I agree with all of this, I was just pondering that if I can manage this on a 40-50mph gusty day with a single tripod, would I be able to get a good result on 8x10 with two tripods on concrete on a windless day?



Again, on the topic of running the ultimate comparison test, a 1/15 second exposure in the outdoors will just raise questions as persons familiar with using 8x10 or larger cameras will easily point to that as a disadvantage stacked against the 8x10 which can easily contribute a skewed outcome. This is the reason why I scratched my head when I saw the test setup in LuLa - I wonder why they did what they did in the setup.

I know 1/focal length does not sound possible but with HP5 at ISO 400, you can use 1/400 at f/16 on a perfectly sunny day based on the Sunny 16 rule. And it just so happens there are plenty of VERY sharp lenses for 8x10 are in the 400mm range - the 16½" Red Dot Artar, the 420mm Apo-Ronar just to name a couple. If you want less grain and slightly higher resolution films like Ektachrome or Ektar at ISO 100, then you may not have a choice but to move the test indoors.


Well I'll fire off a Portra 400 shot just to make sure :-) should be able to get max shutter speed with this (even if I have to underexpose it a couple of stops).

Thanks for the advice! I'm getting there :-)

[rdenney]

It does not seem to me that this is a comparison of a medium-format digital back against 8x10 film. Rather, it seems to me a test of what can be achieved using a MFDB versus 8x10 film.

Each system has strengths and weaknesses in practical use, and those have to be part of the equation.

But before any of that, the question of what we hope to learn must be asked. Let's put it in the form of a resolution, debate-style:

RESOLVED: Using the IQ180 makes it possible to achieve superior representation of fine detail in a six-foot-wide digital print compared to using 8x10 film.

Asking the question that way has two effects: 1.) it keeps the test aimed at what those who might really be considering a choice want to know, and 2.) it keeps us from being tempted to isolate the question so narrowly that it loses its real-world value. Limiting it to representation of fine detail takes advantage of what is easy to measure objectively, but one can still add subjective evaluations to address other issues such as tonal smoothness. Asking it this way takes cost out of the equation, which is reasonable: Once we understand the state of the art, we can apply our own value to the results.

That's why it is not a lens test. We already know how the lenses perform, or at least we think we do. If we are going to test what is possible, though, then we should use lenses that are actually available to those seeking the best possible performance. My RESOLVED statement also keeps us in the digital domain, imposing a scanning step on the film. That puts the film at a disadvantage, of course, but most photographers, even those who use 8x10 every day, will have to use a drum scan and a digital printer to make a print that large. Again, these are processes that photographers desiring the best performance they can get in a really big print will most often use. Having articulated those processes makes it easier to address issues that emerge when each process is put into practice.

Testing small prints won't get at the answer, because both have to be tested to the point where the breakdown of fine detail becomes obvious. So what if the digital back achieves the same apparent sharpness in a 30" print? People get excellent 30" prints, at least in the one measure of the representation of fine detail, from 4x5. The print needs to be big enough so that in the question of fine detail the use of 8x10 is the preferred film choice. Nobody will be persuaded unless the MFDB achieves better fine detail in 8x10's home turf. Of course, an actual six-foot print does not have to be made, just portions at that degree of enlargement. And what gets display on the monitor doesn't have to be the actual print, just the file that made the print, as long as both are at the same printed pixel density by whatever means.

And choices concerning aperture should be related to achieving maximum fine detail in a real situation. It has been argued that one advantage of the MFDB is better depth of field from the smaller format. Okay, so compared samples that are assuredly on the focus plane, but compared also samples within the depth of field but not in the focus plane. It may be that the degradation from overenlarging the medium format is not as bad as the loss of depth of field from 8x10 in an actual comparison. Or the reverse.

Buld the test scenario around that question, and make sure the question is stated clearly. Otherwise, people will argue endlessly about the validity of the test, instead of arguing about the validity of the question the test proposed to resolve.

I happen to like the idea of throwing in a 4x5 camera and run-of-the-mill lens. That would give people a way to relate what the MFDB and 8x10 options provide that a really inexpensive solution does not provide.

Rick "thinking the question is as important as the answer" Denney

[Asher Kelman]

Rick,

We are considering that the Phase One IQ180 digital camera system might be better than 8x10 for making those large fine prints, where we currently rely on film.

A 6ft wide print? Really? Shouldn't one One use a multiplication factor for 8x10 color film that is within the bounds of the largest one would actually commonly use to make fine prints? So choose the film and then tell us what the mag factor will be. 6ft wide seems arbitrary!

Asher

[timparkin]

It does not seem to me that this is a comparison of a medium-format digital back against 8x10 film. Rather, it seems to me a test of what can be achieved using a MFDB versus 8x10 film.

Each system has strengths and weaknesses in practical use, and those have to be part of the equation.

But before any of that, the question of what we hope to learn must be asked. Let's put it in the form of a resolution, debate-style:

RESOLVED: Using the IQ180 makes it possible to achieve superior representation of fine detail in a six-foot-wide digital print compared to using 8x10 film.

I understand where you are coming from but I want to get something different out of this. I want to find out what the best digital output looks like compared to the best film output. Resolution is only part of that and things like colour texture rendering, tonal qualities etc. I don't plan to tell people that one platform is better than another but will look at the results and compare certain attributes I can see and will request feedback from people about what they see also.

The goal is to allow people to have a resource that lets them do a proper side by side comparison of the various film formats and also the various digital formats. The main goal is to retest the original comparison but whilst we have the hardware, put a few more side by side (give the human resources to avoid errors induced by cranial overload).

I know where you are coming from though and these questions can be asked given the raw output from the test. i.e. If we ensure the best possible raw captures then we can pose questions against this resource in terms of lots of different criteria.

Thanks for your input and I know it will be useful once we start to see what the output from the test can answer..

Tim

[timparkin]

Rick,

We are considering that the Phase One IQ180 digital camera system might be better than 8x10 for making those large fine prints, where we currently rely on film.

A 6ft wide print? Really? Shouldn't one One use a multiplication factor for 8x10 color film that is within the bounds of the largest one would actually commonly use to make fine prints? So choose the film and then tell us what the mag factor will be. 6ft wide seems arbitrary!

Asher

10x enlargement? 100"x80"?

Tim

[Asher Kelman]

10x enlargement? 100"x80"?

Tim

Is that your experience? That seems very optimistic. If one can do that and still retain excellence, great!

Asher

[timparkin]

Is that your experience? That seems very optimistic. If one can do that and still retain excellence, great!

Not my experience at all but I've seen 4x5's at 40x50" and they've been pretty damned good.

[Noah A]

Reading this thread has pointed out, in my opinion, the futility of such tests.

People (including myself) questioned the recent Luminous Landscape test because the methodology seemed flawed and the results on the film end didn't match up to the quality many of us know is possible from 8x10. I do think the test was flawed but I didn't lose sleep over it.

But the main reason such tests are of limited value is because they try to appeal to a broad audience. For example, after proposing the test Tim received a bunch of great suggestions--to try different films, different lenses, tripods, shooting conditions, etc. Each suggestion was put forward by a well-meaning photographer who wanted to know how the digi setup would compare to his or her current setup.

So that brings me to my point. I propose, with all due respect to Tim who is about to embark upon quite a project; and one that will surely be controversial, that it's really better for each photographer who is considering the switch to digital to do his or her own tests.

I've thought about switching, and may be forced to consider it again (see the Kodak threads:( ).

If I were to do a test, I'd shoot with my current tripod, camera, lenses and film stocks. I'd shoot the test under the conditions I regularly work under. And I'd make test prints in the sizes I make my print editions.

There are so many variables here that this test will be a nightmare that will annoy people on all sides. Meanwhile those who are really thinking about switching to different photographic tools will just talk with the fine folks at Capture Integration or some other local shop and arrange for a rental or test.

Good luck with the test, tim. Can't wait to see the results:D .

[Asher Kelman]

Not my experience at all but I've seen 4x5's at 40x50" and they've been pretty damned good.

Well, if one is sampling not printing the giant size, then 5x and 10x would seem to test the system well if that fits in with the magnifications current for use of 8x10 film for those who print large.



So that brings me to my point. I propose, with all due respect to Tim who is about to embark upon quite a project; and one that will surely be controversial, that it's really better for each photographer who is considering the switch to digital to do his or her own tests.

Still, Tim's test is going to be useful for a lot of us!

Asher

[timparkin]

Reading this thread has pointed out, in my opinion, the futility of such tests.

People (including myself) questioned the recent Luminous Landscape test because the methodology seemed flawed and the results on the film end didn't match up to the quality many of us know is possible from 8x10. I do think the test was flawed but I didn't lose sleep over it.

But the main reason such tests are of limited value is because they try to appeal to a broad audience. For example, after proposing the test Tim received a bunch of great suggestions--to try different films, different lenses, tripods, shooting conditions, etc. Each suggestion was put forward by a well-meaning photographer who wanted to know how the digi setup would compare to his or her current setup.

So that brings me to my point. I propose, with all due respect to Tim who is about to embark upon quite a project; and one that will surely be controversial, that it's really better for each photographer who is considering the switch to digital to do his or her own tests.

I've thought about switching, and may be forced to consider it again (see the Kodak threads:( ).

If I were to do a test, I'd shoot with my current tripod, camera, lenses and film stocks. I'd shoot the test under the conditions I regularly work under. And I'd make test prints in the sizes I make my print editions.

There are so many variables here that this test will be a nightmare that will annoy people on all sides. Meanwhile those who are really thinking about switching to different photographic tools will just talk with the fine folks at Capture Integration or some other local shop and arrange for a rental or test.

Good luck with the test, tim. Can't wait to see the results:D .

I'm glad you can't wait to see the results. I don't know many people who are thinking of moving to digital who would be in a position to make a test like this. For instance, they would have to upgrade their current film system to 'state of the art' because this would be the alternative to buying the digital system.

They certainly wouldn't be allowed to walk out with a full linhof and alpa system and P45+ and IQ180 systems, never mind the set of lenses. They would inevitably end up making a compromised test under the tutelage of the supplier.

Also, most people would then probably not be able to make all of the relevant scans (if they knew which people to trust for instance - how many bad scanning houses are there out there). Reaching state of the art in both digital and film is out of the question for most people. The only people who can probably do this is people who have already worked with those systems, hence why we'll have a Phase supplier, a photographer who has worked intensively with both film and digital and someone who has the engineering background to conduct a decent test.

The results won't satisfy everyone but they will provide 95% of relevant results to 95% of people and that is enough for me.

I know where you are coming from and I know a small percentage of people will have a problem with it. Having worked with online communities for a long time, I know the 'public' reaction I will get (I've dealt with worse, I assure you!). However I also know the worth the results will have for a very large section of the photographic public.

Tim

[Brian K]

We're talking about using a 210 because it's the right focal length to compare with the 40mm Digaron W. We're also going to be using a 360 Sironar S to compare with the 70mm Digaron W.


Could you tell me how to do this? If I've managed nr. 4000dpi of detail out of my 10x8 already would that imply my focus is OK?


I agree but that isn't how most people will use the systems and we won't be taking photographs if it's really windy and we'll be using strong tripods as discussed.


Ermm... I didn't do the original tests and even though the original test had problems, the 8x10 still outresolved the IQ180 (more on that later).

I would really appreciate your input but it would help if you knew who did what. Markus Zuber did the original test and I have uploaded various images to show resolution beyond 4000dpi on some 617 and 4x5 shots and nr. 4000dpi results for 8x10. The original results were fine as far as wind and stability goes but there was a focus inaccuracy and depth of field was less than the IQ180.

Tim

Sorry about the comments about the 8x10 image quality, I assumed the samples were yours.

I am currently scanning 120 Portra 160 film on My IQSmart3, shot with a Rolleiflex 6008i and a 80mm 2.8 Schneider. I'm finding that, much to my surprise, it's showing an ultimate resolution of about 5000dpi. On a chimney roughly 150 feet from my house, with some chicken wire type wire under the chimney cap, that I can see that wire quite clearly at more than 50x. Put in another way, a .75mm section of film blown up to 39mm is rendering fine detail. And this is NOT a wet scan and there is minimal in scanner sharpening.

As for using tripods outdoors, I carry either sandbags or lead shot bags with me. I loop a bungee cord around the tripod legs, hang another bungee from it, and then anchor that to about 50 pounds of sand/lead setting below. (It's not fun when I have to hike) This added mass really deadens the vibrations especially with a heavy carbon fiber tripod. I shoot the majority of my images this way.

I also use a large umbrella to shield the camera from any wind.

Many GG's are out of alignment, or have been replaced with fresnels or whatever that were never accurately positioned. By Linhof came shimmed to match the GG to the film plane, my Sinars were also aligned and verified.

[timparkin]

Sorry about the comments about the 8x10 image quality, I assumed the samples were yours.

I am currently scanning 120 Portra 160 film on My IQSmart3, shot with a Rolleiflex 6008i and a 80mm 2.8 Schneider. I'm finding that, much to my surprise, it's showing an ultimate resolution of about 5000dpi. On a chimney roughly 150 feet from my house, with some chicken wire type wire under the chimney cap, that I can see that wire quite clearly at more than 50x. Put in another way, a .75mm section of film blown up to 39mm is rendering fine detail. And this is NOT a wet scan and there is minimal in scanner sharpening.


Agreed, I'm looking at 80x through my microscope and seeing detail way beyond 4000dpi.. e.g. take a look here (http://static.timparkin.co.uk/static/tmp/4000dp-vs-microscope.jpg) left hand side microscope ~30,000dpi , right hand side 4000dpi



As for using tripods outdoors, I carry either sandbags or lead shot bags with me. I loop a bungee cord around the tripod legs, hang another bungee from it, and then anchor that to about 50 pounds of sand/lead setting below. (It's not fun when I have to hike) This added mass really deadens the vibrations especially with a heavy carbon fiber tripod. I shoot the majority of my images this way.

I also use a large umbrella to shield the camera from any wind.

Again, I know where you are coming from. I've found my spikes the most useful addition yet. Embedded an inch or so into the ground, it gives pretty good stability. Additional weight can't hurt at all!



Many GG's are out of alignment, or have been replaced with fresnels or whatever that were never accurately positioned. By Linhof came shimmed to match the GG to the film plane, my Sinars were also aligned and verified.
OK - in which case my Ebony is still stock using a maxwell screen and I'm happy with the resolution. The Toyo 810MII is pretty mint and I've been told is unlikely to have a problem. With focus bracketing we'll find out a bit more though. I'll also choose a subect that isn't a flat planar wall so even if the focus is out a little, we will still show a sharp area on the focal plane.

Tim

[rdenney]

A 6ft wide print? Really? Shouldn't one One use a multiplication factor for 8x10 color film that is within the bounds of the largest one would actually commonly use to make fine prints? So choose the film and then tell us what the mag factor will be. 6ft wide seems arbitrary!

A 6-foot pring is 72 inches, or a hair over 7x. That seems unreasonable?

Yes, it's arbitrary, but not completely. If the magnification is too small, you won't be able to evaluate fine detail. Pick a different size--I don't care. But if you evaluate with a print where the digital may well outperform 8x10 (because of the lack of generational scanning loss, as LuLa did), then expect people to respond, "yeah, but what if you were making a really big print?". The point of the test is to provide a useful comparison, after all. Do we really need a test to show that a MFDB is as good as 8x10 at representing fine detail for, say, a 16x20 print? If not, then there is a line below which the print is not large enough. The test just needs to be on the other side of it.

Were it me, I'd test both to destruction, so that they are both completely broken down when viewed closely (because of grain, lack of lens resolution--whatever). Then, I'd put them side by side, and start backing up, to compare them based on minimum viewing distance to achieve visually sharp results. The maximum magnification could easily be calculated from that. Maybe that would take 10-foot prints. Maybe you'd get to it with four-foot prints, but I rather doubt it.

Rick "just sayin..." Denney

[Frank Petronio]

As for using tripods outdoors, I carry either sandbags or lead shot bags with me. I loop a bungee cord around the tripod legs, hang another bungee from it, and then anchor that to about 50 pounds of sand/lead setting below. (It's not fun when I have to hike) This added mass really deadens the vibrations especially with a heavy carbon fiber tripod. I shoot the majority of my images this way.

I also use a large umbrella to shield the camera from any wind.

And then in the next thread we'll read about someone using their 8x10 with a #2 Gitzo CF ;-)

[Brian K]

And then in the next thread we'll read about someone using their 8x10 with a #2 Gitzo CF ;-)

a gorilla pod

[Brian K]

Agreed, I'm looking at 80x through my microscope and seeing detail way beyond 4000dpi.. e.g. take a look here (http://static.timparkin.co.uk/static/tmp/4000dp-vs-microscope.jpg) left hand side microscope ~30,000dpi , right hand side 4000dpi


Again, I know where you are coming from. I've found my spikes the most useful addition yet. Embedded an inch or so into the ground, it gives pretty good stability. Additional weight can't hurt at all!


OK - in which case my Ebony is still stock using a maxwell screen and I'm happy with the resolution. The Toyo 810MII is pretty mint and I've been told is unlikely to have a problem. With focus bracketing we'll find out a bit more though. I'll also choose a subect that isn't a flat planar wall so even if the focus is out a little, we will still show a sharp area on the focal plane.

Tim

If you want to test your film plane v GG there's a simple way. Take a holder the same brand as you usually use, hopefully ALWAYS use, cut a hole in the center of the holder, say 2x2.5" On the side that faces the lens glue a GG, clear side toward lens, mat side toward film, use 4x5 film to act as precision spacers, so the mat side sits exactly on the film. You now have a GG that sits exactly on the film plane, without ANY doubt. But if you use a variety of holders you'll need to make damn sure that they are all exactly the same thickness.

[Brian K]

Agreed, I'm looking at 80x through my microscope and seeing detail way beyond 4000dpi.. e.g. take a look here (http://static.timparkin.co.uk/static/tmp/4000dp-vs-microscope.jpg) left hand side microscope ~30,000dpi , right hand side 4000dpi


Again, I know where you are coming from. I've found my spikes the most useful addition yet. Embedded an inch or so into the ground, it gives pretty good stability. Additional weight can't hurt at all!


OK - in which case my Ebony is still stock using a maxwell screen and I'm happy with the resolution. The Toyo 810MII is pretty mint and I've been told is unlikely to have a problem. With focus bracketing we'll find out a bit more though. I'll also choose a subect that isn't a flat planar wall so even if the focus is out a little, we will still show a sharp area on the focal plane.

Tim

Tim, I have a Leica Stereomicroscope, I'd like to hook up a camera like you have, what do I need?

[rdenney]

I understand where you are coming from but I want to get something different out of this. I want to find out what the best digital output looks like compared to the best film output. Resolution is only part of that and things like colour texture rendering, tonal qualities etc. I don't plan to tell people that one platform is better than another but will look at the results and compare certain attributes I can see and will request feedback from people about what they see also.

Agreed on all counts. The point of a debate question is not to presuppose the answer, but to be clear about why the test is being conducted. It also recognizes that there are two sides to this question, each with its biases and prejudices. Instead of pretending to have no bias, it addresses the bias by recognizing it and then constructing the test to be sure that the bias doesn't leak into the method. My expectation (i.e., bias) is that my resolution as I proposed will not be able to be defended, but I'd be happy to end the presentation with "you decide" and leave the conclusion to the reader.

But if you choose to frame it in a less confrontational way, that's fine with me.

I do think you need to make the representation of fine detail a principal factor, though. That's what drives most people to seek out large format, and probably that's what makes this $50,000 back different from, say, a $10,000 Pentax 645D. If you don't make it your principal evaluation, what will you be able to display on the web? Tonality, color, and smoothness--whatever--are mighty hard to see on a computer monitor, and it's really the print that must display it. Are you going to send prints out to people? Of course not.

But you would certainly want to include subjective evaluations from experts. This is where the debate-style question really comes into play--it allows your reviewers to admit their biases without disengenuously disclaiming them. And it would make the resulting article a more interesting read.

In my line of work, there was at one time a long-standing debate about the use of two different types of equipment. The debate had pitted agency against agency and had assumed religious proportions. I was putting together a technical program for a conference, and the manufacturers of one type wanted a session devoted to the question. What they wanted was a sales pitch from their hacks. What I gave them was a formal debate, with teams, times, rebuttals, and judges. In the end, I told the judges not to award a prize--let the attendees decide for themselves--but the judging was there to enforce the debate rules. Each team had an implementer and a manufacturer, and both were selected because of their passion for their point of view (and also for their ability to maintain their good humor in this sort of a situation). I submit that the debate actually transmitted far more real and valuable information about the choices than any sales-type presentations could have done, and did so in a way that commanded attention from the audience. A little competition sharpens the wits.

Rick "just some ideas to consider" Denney

[Noah A]

I don't know many people who are thinking of moving to digital who would be in a position to make a test like this. For instance, they would have to upgrade their current film system to 'state of the art' because this would be the alternative to buying the digital system.

No professional photographer I know would spend upwards of $50k on a system WITHOUT testing it out in person.


They certainly wouldn't be allowed to walk out with a full linhof and alpa system and P45+ and IQ180 systems, never mind the set of lenses. They would inevitably end up making a compromised test under the tutelage of the supplier.

I strongly suggest that anyone considering such a system contact a reputable shop. I've had several in-depth conversations with the folks at Capture Integration, and even though I've never spent a dime there, they were very free with their time and advice. They also offered a demo (yes, at their location, but I'm not local to them). They and other shops also do rentals and in many cases if you buy a system they'll credit the money back. However, doing a test with the help of a really knowledgeable shop can help you learn how to use the digital gear and get the most out of it.


I know where you are coming from and I know a small percentage of people will have a problem with it. Having worked with online communities for a long time, I know the 'public' reaction I will get (I've dealt with worse, I assure you!). However I also know the worth the results will have for a very large section of the photographic public.

I don't have a problem with your test, in fact I admire your courage and commitment. I think the results will be interesting.

Why not shoot some 4x5 as well, if you really want to be complete? Out of the fine art and professional/editorial shooters I know personally, more of them shoot 4x5 than 8x10. And this is the same audience of people who may be considering a switch. The 4x5 cameras are usually lower profile and less affected by the wind, the film may (?) be flatter in the smaller holders and the films are so good these days that grain is really minimal even at large print sizes.

If you do decide to make prints, 10x for the film is a minimum I'd say. I rather like the look of 4x5 Portra 160 even at 48x60...

[Nathan Potter]

Good discussion here on the very great difficulty in executing an experimental design. The comments by R. Denney are particularly relevant - that is clarity in why such a test is being done. Equally important is predetermining the metrics for the final evaluations.

Subjective evaluation is always interesting and even useful but it is not data that can be submitted to statistical analysis. Resolution data in the form of MTF measurement is data, hence my opinion that it should also be obtained. But is there other physically measurable data that could be obtained that would differentiate the attributes of digital capture vs 8X10 film? One needs to be careful that clarity in evaluation of the final result will be not be fuzzy.

A cardinal rule for experimental success is to know precisely what you want to find out before doing any experimental work. That usually means measuring something that will yield an unambiguous result.

Nate Potter, Austin TX.

[Asher Kelman]

Subjective evaluation is always interesting and even useful but it is not data that can be submitted to statistical analysis. Resolution data in the form of MTF measurement is data, hence my opinion that it should also be obtained.
Nathan, I love science and measuring and would gladly devour any data that is shared. However, the task, as I see it is to discover whether or not the IQi80 would do what the 8x10 does in making large prints and so well that one would want to switch.


A cardinal rule for experimental success is to know precisely what you want to find out before doing any experimental work. That usually means measuring something that will yield an unambiguous result.


When one chooses a wife, does on go by statistics or by whether or not one is moved? Here we are trying to produce prints that move the viewer. So subjective impressions are needed too! Can't escape that!

So, besides the data we all like, we also have to see whether or not the Phase One IQ180 at whatever the largest sizes the 8x10 film yields can deliver equal presence, feeling and persuasion to the 8x10 of the same size.

Asher

[Frank Petronio]

My wife is named Phase Two IQ720. She kicks 8x10s down the street for sport.

[timparkin]

Tim, I have a Leica Stereomicroscope, I'd like to hook up a camera like you have, what do I need?

Hi Brian,

You just need a camera converter which fits in the place of the eye peice magifier. Most microscopes are 30mm or 23mm and the converter I got has an adapter for both.

Mine was an olympus mount but there are converters from olympus to EF and various other makes.

Here is a list on ebay. The first lot of 23/30 adapters are the type I bought.

[timparkin]

If you want to test your film plane v GG there's a simple way. Take a holder the same brand as you usually use, hopefully ALWAYS use, cut a hole in the center of the holder, say 2x2.5" On the side that faces the lens glue a GG, clear side toward lens, mat side toward film, use 4x5 film to act as precision spacers, so the mat side sits exactly on the film. You now have a GG that sits exactly on the film plane, without ANY doubt. But if you use a variety of holders you'll need to make damn sure that they are all exactly the same thickness.

Thanks Brian - I'll have a go at this..

Tim

[timparkin]

What I gave them was a formal debate, with teams, times, rebuttals, and judges. In the end, I told the judges not to award a prize--let the attendees decide for themselves--but the judging was there to enforce the debate rules.

This is exactly what I am after, hence having proponents/experts of both systems to argue the case.

As far as the 'you can't judge colour texture, etc on a screen', if I supply the digital files, people are free to print them out for themselves using whatever service they desire. If they want to produce darkroom prints, I would be happy to loan the final film out to them on the condition that we get to 'hold' the final result once the print has been used for assessment (hence the final data will always sit in one place for people to compare).


My expectation (i.e., bias) is that my resolution [criteria] as I proposed will not be able to be defended

My personal bias agrees. Hopefully the Phase representative and owner of the P45+ will try to argue otherwise and the results will allow people to make their own mind up.

Really appreciate the input

Tim

[Joerg Krusche]

Hi,

with resolution being one of the issues .. you might consider to simplify your test. Run a test according Norman Koren .. shoot his test chart with both systems .. analogue and digital .. at 50x focal length. Then you can just read the number of linepairs/mm resolved at comparable contrast. Then multiply obtained linepairs/mm times film/sensor dimensions. You obtain a more quantitative number of linepairs resolved by each system .. and just see who is the winner when resolution is asked for. Well aligned camera/noshake/etc. assumed.

The advantage of this test is that it avoids subjective perceptions etc. .. you may be surprised how well LF performs.

good luck

joerg

[timparkin]

Hi,

with resolution being one of the issues .. you might consider to simplify your test. Run a test according Norman Koren .. shoot his test chart with both systems .. analogue and digital .. at 50x focal length. Then you can just read the number of linepairs/mm resolved at comparable contrast. Then multiply obtained linepairs/mm times film/sensor dimensions. You obtain a more quantitative number of linepairs resolved by each system .. and just see who is the winner when resolution is asked for. Well aligned camera/noshake/etc. assumed.

The advantage of this test is that it avoids subjective perceptions etc. .. you may be surprised how well LF performs.

good luck

joerg

I am loathe to start making this a purely resolution based test as this won't address the main reasons many people shoot 4x5/8x10 but I will try to include a test chart in the overall view if possible. (maybe a few either side of the focal point)

Tim

[TJV]

Correct me if I'm wrong, Tim, but isn't what we're seeing in your microscope vs. scan crops more to do with grain aliasing and the limitations of the analogue to digital workflow? I've found that when scanning film the grain structure is significantly enhanced due to pixels being square r, b and b values and not multi-layered organic and random structures, if that makes sense? From my understanding, to get a real world improvement in scan quality you'd have to scan at 8000dpi. This doesn't even begin to take into account the noise introduced within the digital chain, etc.


Agreed, I'm looking at 80x through my microscope and seeing detail way beyond 4000dpi.. e.g. take a look here (http://static.timparkin.co.uk/static/tmp/4000dp-vs-microscope.jpg) left hand side microscope ~30,000dpi , right hand side 4000dpi


Again, I know where you are coming from. I've found my spikes the most useful addition yet. Embedded an inch or so into the ground, it gives pretty good stability. Additional weight can't hurt at all!


OK - in which case my Ebony is still stock using a maxwell screen and I'm happy with the resolution. The Toyo 810MII is pretty mint and I've been told is unlikely to have a problem. With focus bracketing we'll find out a bit more though. I'll also choose a subect that isn't a flat planar wall so even if the focus is out a little, we will still show a sharp area on the focal plane.

Tim

[timparkin]

Correct me if I'm wrong, Tim, but isn't what we're seeing in your microscope vs. scan crops more to do with grain aliasing and the limitations of the analogue to digital workflow? I've found that when scanning film the grain structure is significantly enhanced due to pixels being square r, b and b values and not multi-layered organic and random structures, if that makes sense? From my understanding, to get a real world improvement in scan quality you'd have to scan at 8000dpi. This doesn't even begin to take into account the noise introduced within the digital chain, etc.

Yes noise does get introduced with the 4000dpi scans. I think for medium format, there would definitely be a benefit in 8000 or even 12000dpi in order to oversample. The returns are diminishing though. However it's great to know the extra resolution is available if needed.

Tim

[Joerg Krusche]

Tim,

I did not mean that you should run just that resolution test .. you are free to add whatever .. but if the resolution test tells you that 8x10 resolves with good contrast 30 lps/mm .. all you need is a 16x lupe and a light table .. if you then multiply 200mm (8 inch) by 30 then you obtain 6000 lps on the short side .. (and 7500 lps 0n the 10 inches) ..and if your digital system does give you around 3500 lps on the short side.. and is still claimed to be superior to 8x10..then I assume that there was something suboptimal with the 8x10 system. BTW it is quite interesting to run that simple test on a digital system .. resolution does break down suddenly to mush .. as you would expect .. film does show resolution slowly disppear due to loss of contrast .. just as we would expect. Multiplying actual linepair numbers is in my view much more meaningful than quoting pixel numbers . .two pixels do not represent a linepair resolved .. therefore I think that pixel numbers alone do not adequately describe system performance ..

joerg

[Jim Michael]

I'm also going to scan on an Epson V750 just to see what your average consumer can manage.

Tim

It may have been in the scanner comparison thread, but someone posted some results from a well tuned V7xx series scanner that were pretty outstanding. The calibration methodology for these scans should be published with the results since it has a bearing on the quality of the output.

Count me in as a contributor.

[timparkin]

It may have been in the scanner comparison thread, but someone posted some results from a well tuned V7xx series scanner that were pretty outstanding. The calibration methodology for these scans should be published with the results since it has a bearing on the quality of the output.

Count me in as a contributor.

Hi Jim,

I've made some test with the Epson V750 and have been quite surprised to see it resolving up to 2500dpi (possibly slightly more but that may have been sharpening etc). My technique is ... height adjustment in 0.25mm increments until sharpest. Wet mount on anti-newton glass. Dry mount loses a tiny amount but I've found is better in the shadows (strangely - I think newton rings are still evident when wet mounting because lack of tension gives variation in mylar to film gap).

What would you like to contribute?

Tim

[timparkin]

Tim,

I did not mean that you should run just that resolution test .. you are free to add whatever .. but if the resolution test tells you that 8x10 resolves with good contrast 30 lps/mm .. all you need is a 16x lupe and a light table .. if you then multiply 200mm (8 inch) by 30 then you obtain 6000 lps on the short side .. (and 7500 lps 0n the 10 inches) ..and if your digital system does give you around 3500 lps on the short side.. and is still claimed to be superior to 8x10..then I assume that there was something suboptimal with the 8x10 system. BTW it is quite interesting to run that simple test on a digital system .. resolution does break down suddenly to mush .. as you would expect .. film does show resolution slowly disppear due to loss of contrast .. just as we would expect. Multiplying actual linepair numbers is in my view much more meaningful than quoting pixel numbers . .two pixels do not represent a linepair resolved .. therefore I think that pixel numbers alone do not adequately describe system performance ..

joerg

Ah! I get you :-) I can compare the resolution of my microscope results because they are taken on a digital camera which makes a great measurement tool (fixed pixels per view width) need to calibrate this though. Agree that comparing any MTF style results or using pixels for resolution value is frought with danger because bayer array distorts these results (and is inconsistent with regard to direction and colour).

I'll concentrate on getting the best raw results for now.

Tim

Tim

[Joerg Krusche]

Tim,

I was pretty much disappointed when I compared nominal dpi numbers of a scanner with actual line pairs resolved ..(and these translated into dpi's) .. high dpi numbers mean nothing unless one has run something trivial as a USAF 1951 chart test .. quoted 6400 dpi may turn out to be just 1800 dpi .. if at all.. I learned that file size is nothing .. a decent true 1800 dpi scan may be better than a quoted 6400 dpi scan with inflated file size on a different scanner... running simple lp tests like on 8x10 allow to determine system performance .. and to optimize .. or live with it.

joerg

[Jim Michael]

What would you like to contribute?

Tim

I'll contribute a portion of the expenses for the tests.

[timparkin]

Tim,

I was pretty much disappointed when I compared nominal dpi numbers of a scanner with actual line pairs resolved ..(and these translated into dpi's) .. high dpi numbers mean nothing unless one has run something trivial as a USAF 1951 chart test .. quoted 6400 dpi may turn out to be just 1800 dpi .. if at all.. I learned that file size is nothing .. a decent true 1800 dpi scan may be better than a quoted 6400 dpi scan with inflated file size on a different scanner... running simple lp tests like on 8x10 allow to determine system performance .. and to optimize .. or live with it.

joerg

Hi Joerg - I don't think anyone is saying that scanning at 12,000dpi means 12,000 dpi of content (look at Epson scanning resolutions for confirmation of that). I also don't think a USAF 1951 resolution target is trivial to set up or read either (and how do you read resolution figures for digital? at 90 degrees or 45 degrees, with monochromatic light, green light or red light? How do you interpret the sharpening of the Bayer array and it's effect on contrast?

[Jim Michael]

The instructions for the USAF test chart are pretty clear and have you set an array of targets in various orientations (90 and 45 degree orientations included) at a set distance dependent on focal length. You then locate the least resolvable lines and look up the resolution. Nothing difficult at all, unless you start changing the distance and size of the targets in which case some math is involved. The benefit of using the test charts is being able to quantify the results and compare to the theoretical numbers.

[Jim Michael]

Re the Bayer related issues, I think the approach there is to shoot a zone plate (sinusoidal) to determine the true resolution of the sensor.

[Brian Ellis]

Hi Joerg - I don't think anyone is saying that scanning at 12,000dpi means 12,000 dpi of content (look at Epson scanning resolutions for confirmation of that). I also don't think a USAF 1951 resolution target is trivial to set up or read either (and how do you read resolution figures for digital? at 90 degrees or 45 degrees, with monochromatic light, green light or red light? How do you interpret the sharpening of the Bayer array and it's effect on contrast?

A USAF 1951 resolution target also isn't inexpensive. Prices start around $150 and can go up to more than $1,000. The ones you can download or buy for a lot less are just copies of the target.

[Jim Michael]

A USAF 1951 resolution target also isn't inexpensive. Prices start around $150 and can go up to more than $1,000. The ones you can download or buy for a lot less are just copies of the target.

Link to a USAF 1951 chart PDF on this page (http://www.normankoren.com/Tutorials/MTF5.html). No reason the USAF chart hasn't been done in PS, it's a pretty simple construct.

Oh, and I should have an SVG around somewhere, if there's any interest.

[Nathan Potter]

It may have been in the scanner comparison thread, but someone posted some results from a well tuned V7xx series scanner that were pretty outstanding. The calibration methodology for these scans should be published with the results since it has a bearing on the quality of the output.

Count me in as a contributor.

http://farm7.static.flickr.com/6171/6154033421_f5c72e4169_z.jpg (http://www.flickr.com/photos/argiolus/6154033421/)
EPSONcont-web-1 (http://www.flickr.com/photos/argiolus/6154033421/) by hypolimnas (http://www.flickr.com/people/argiolus/), on Flickr

Jim, I think you may be referring to a plot I made recently for the V750 Pro. Plot is above from Flickr. The plot was made from the resolution target after importation into PS and with the full density range from 0 to 256. In other words my interest was in finding the MTF curve for a scan where the linearity and full 0 to 256 density range was preserved during and after the scan.

Since the resolution mask has no intermediate densities (it is either clear or opaque)
it is possible to simply clip the white and black points to provide a high contrast image that would indicate a high resolution (I could get to 2600 SPI at above 50% contrast) but that sacrifices all middle density values and is not useful for many of us; and me.

I will stress again that the LL article is almost entirely about resolution between the IQ180 with snappy modern lenses and 8X10 format with high quality LF lenses. Zubers addenda recently, in answer to criticism about the 8X10 limitations, only describes a "richness of detail" benefit in using the IQ180. Dunno what that means precisely. One mans richness is another mans fault. Still wondering how to extract quantitative data out of "richness of detail".

I understand completely why a subjective judgement is useful at the end of the comparison. One could even tabulate and rate certain attributes of prints such as clarity, contrast, color saturation, etc. Then I think there would be no consensus as to which is best - if best is what we would be after. IMHO all the examiners will bring a different perspective and persuasion to bear on the images and yield no consensus, leaving us where we are now - in argument with endless discussion. :(

Nate Potter, Austin TX.

[Ken Lee]

Thanks for making that curve about the Epson. It makes me want to get a drum scanner :rolleyes:

It also shows why different people find different rates of enlargement to be acceptable: 3x, 5x, etc. They're merely stating their tolerance for low contrast, so to speak.

[Jim Michael]

No I hadn't seen that curve Nate. Thanks for do that. I'll have to go search the archives but what I was referring to may have been a scan of the reference target that was being used for the scan comparisons. There was a description of the methodology used in setting the scanner up IIRC.

[Jim Michael]

I think this (http://www.largeformatphotography.info/forum/showthread.php?t=46503&highlight=mylar) is the post to which I was referring. See #32.

[John NYC]

I think this (http://www.largeformatphotography.info/forum/showthread.php?t=46503&highlight=mylar) is the post to which I was referring. See #32.

Yes. Not to start this all over again (there have been many threads where we all disagree on the degree of the problem), but the advertised dpi of scanners is -- how should I say this? -- optimistic. Cough.

[ic-racer]

I like Nathan's idea of a real world comparison between the two. Another way to do it would be to get a friend or volunteer who does film/darkroom 8x10 to choose his/her best 8x10 equipment and shoot your target and process and print it.

In fact I'd set the two cameras up side-by side an shoot some pictoral scenes rather than test charts. Then let your audience decide which pictures look best.

[John NYC]

http://farm7.static.flickr.com/6171/6154033421_f5c72e4169_z.jpg (http://www.flickr.com/photos/argiolus/6154033421/)
EPSONcont-web-1 (http://www.flickr.com/photos/argiolus/6154033421/) by hypolimnas (http://www.flickr.com/people/argiolus/), on Flickr

Jim, I think you may be referring to a plot I made recently for the V750 Pro. Plot is above from Flickr. The plot was made from the resolution target after importation into PS and with the full density range from 0 to 256. In other words my interest was in finding the MTF curve for a scan where the linearity and full 0 to 256 density range was preserved during and after the scan.

Nate Potter, Austin TX.

This is interesting. And it affirms in a different way my own personal testing results of what I think for me is the practical max resolution to scan at on the V750.

[timparkin]

OK - looks like we're on for testing next week (14th October) and will have the best lens for the 10x8 too (Mr CAD (http://www.mrcad.co.uk/shop/home.php) are lending us a Sironar S 360 and possibly a 210 - great people who I've dealt with before.

The scene will be one interior in a small studio with window lights and one scene from a large garden. If we get time we may nip up the road for a proper 'scenic' shot.

I've yet to do the maths for appropriate distances but will have it all planned in advance.

I can include a resolution target and was wondering which one is recommended? (something I can print out preferably as the timescales are quite small).

I was thinking of including an 8x10 transparency on a light box as a resolution test as well.. should make for a good known target with very high resolution.

Tim

[mcfactor]

that's great. I cant wait to see the results.

-Noah

[Joerg Krusche]

Tim,

perhaps take a print-out of Norman Korens testchart ..download from his website .. as a print it is 25cm/10 inches long .. fits nicely in a 8x10 holder fo setting up .. at 50x focal length the chart will be 5mm on the film .. easy to read around the 50 lps/mm section with a 10-16x magnifying glass..

best regards,

joerg

[QT Luong]

If you include a resolution chart as part of the set up, I would suggest a slanted-edge pattern. It is very easy to print your own, and those can be used to automatically evaluate MTF values with the appropriate software, two things which cannot be said of the USAF chart. See: http://www.imatest.com/docs/sfr_instructions.html

I have an Imatest license, so if you send me the relevant portion of image files, I could run them through the software.

[rdenney]

Tim,

perhaps take a print-out of Norman Korens testchart ..download from his website .. as a print it is 25cm/10 inches long .. fits nicely in a 8x10 holder fo setting up .. at 50x focal length the chart will be 5mm on the film .. easy to read around the 50 lps/mm section with a 10-16x magnifying glass..

best regards,

joerg

I agree on the use of Norman Koren's charts. They will help you evaluate the critical resolution at 10% and 50% MTF, not just minimum discernible resolution. You can print as many as you need and arrange them however you want (and I agree to put one at a 45-degree angle--very important).

Koren's article on lens testing, but it applies here (http://www.normankoren.com/Tutorials/MTF5.html)

His chart can be downloaded and printed. I would use matte-surface paper. Include a big black square on the same print. Also, include a known-good gray card somewhere in the image for calibrating any needed color corrections.

Make sure your film exposures are perfect. That means confirming the shutter, or bracketing if you aren't sure.

Rick "who has done this with medium-format, and used a microscope for evaluation" Denney

[Brian C. Miller]

I would use matte-surface paper.

From experience with my Epson 2200, semi-gloss or gloss will give better results for printing a target. (Yeah, I ran some tests on my printer. Sniff! :( )

[timparkin]

I agree on the use of Norman Koren's charts. They will help you evaluate the critical resolution at 10% and 50% MTF, not just minimum discernible resolution. You can print as many as you need and arrange them however you want (and I agree to put one at a 45-degree angle--very important).

Koren's article on lens testing, but it applies here (http://www.normankoren.com/Tutorials/MTF5.html)

His chart can be downloaded and printed. I would use matte-surface paper. Include a big black square on the same print. Also, include a known-good gray card somewhere in the image for calibrating any needed color corrections.

Make sure your film exposures are perfect. That means confirming the shutter, or bracketing if you aren't sure.

Rick "who has done this with medium-format, and used a microscope for evaluation" Denney

The guys at Imatest have done me a nice deal on Imatest Master but I couldn't afford to buy a big target from them. Norman Koren has been in touch also and is advising.

I'm currently working out how to print out a very big target but also how to include a still life in the scene (it's expensive and dull to take just pictures of targets and there are lots of other non-resolution things we'd like to assess).

Can we mix parts of the target with parts of still life? can we place them side by side? etc. Can I print test target in 17" wide roll sections and then mount to matt board 60x40 mount board?

Also, Markus and I have been working on the old results and have found that we can compare the 55mm result (which had better focus and even though it had extra resolution from it's f/8 and 10% longer focal length was enough to make a simple comparison). This is the result. The IQ180 has about a 20% advantage here in terms of focal length and depth of field.

http://static.timparkin.co.uk/static/scans/55f11-240f32-numberplate.jpg

This is backed up by the diffraction and depth of field maths.

[Jim Michael]

Can we mix parts of the target with parts of still life? can we place them side by side?

How about an assortment: flowers, fabrics with patterns, some finely detailed items such as wood carvings, perhaps a wine bottle with intricate label?

The megapixel equivalent number varies with f-stop (higher res as you open up) so increased depth of field can be a disadvantage, making it preferable to keep things in the same plane.

[Joerg Krusche]

One should consider that scanning the test chart image on the film may not be a good idea for determining resolution in lps/mm .. if done according Norman Koren with 50x focal length distance of the chart .. then for the evaluation of the 50lps/mm region on the film you would need a scanner with at least true 3000dpi resolution .. few scanners are suitable .. visual determination is much easier.

best regards,

joerg

[timparkin]

One should consider that scanning the test chart image on the film may not be a good idea for determining resolution in lps/mm .. if done according Norman Koren with 50x focal length distance of the chart .. then for the evaluation of the 50lps/mm region on the film you would need a scanner with at least true 3000dpi resolution .. few scanners are suitable .. visual determination is much easier.

best regards,

joerg

Will an 11,000dpi Hiedleberg do? I've also got a 100x stereo microscope to which I can attach my 5Dmk2

[QT Luong]

There are a bunch of different "Norman Koren" targets, so it helps to know which one we are talking about. Since you were going to mix up scene and targets, I was thinking of a single slanted edges.

However from your post I assume you meant the SFR-type targets, which are essentially a bunch of slanted squares ? The SFR target was meant to fill up the entire image, however when you run SFR set up there is a way to specify a crop, so I think you may be able to mix it with other stuff. If you use such a target, keep in mind that focal plane/target alignment is more critical than with a single slanted edge.

RC papers are sharper than matte papers, however, I've found reflections to be a problem (in particular with wide lenses), so I use matte paper, which is OK with large targets. I think if you print the SFR target in stripes so that the squares do not overlap stripes, you should be OK.

[Ari]

RC papers are sharper than matte papers...

I think you meant glossy papers, didn't you?

[Joerg Krusche]

Tim,

just keep it simple .. that will do..to reply to your seemingly overkills .. I assume you are less familiar with the 11000 dpi tango/primescan .. these quoted 11000 dpi will eventually not do .. optically it is at max ca. 5500 dpi .. and these only if you are using the latest soft .. NEWSCAN .. with Linocolor you may end up in the range of 2500dpi.. issue is variable aperture or not .. a 16x Peak loupe and a light table is good enough to read up to 80-90 lps/mm region just in case you get there..,

best,

joerg

[Brian C. Miller]

How about something in the test designed to create Moiré patterns with digital sensors? Since the IQ180 doesn't have a filter, I think it would be interesting to see how it renders patterns like that.

[Brian K]

One should consider that scanning the test chart image on the film may not be a good idea for determining resolution in lps/mm .. if done according Norman Koren with 50x focal length distance of the chart .. then for the evaluation of the 50lps/mm region on the film you would need a scanner with at least true 3000dpi resolution .. few scanners are suitable .. visual determination is much easier.

best regards,

joerg

3000dpi is not that hard to achieve, I have an IQSmart 3, it's optical is 5500x10,000

[timparkin]

Tim,

just keep it simple .. that will do..to reply to your seemingly overkills .. I assume you are less familiar with the 11000 dpi tango/primescan .. these quoted 11000 dpi will eventually not do .. optically it is at max ca. 5500 dpi .. and these only if you are using the latest soft .. NEWSCAN ..

joerg

Sorry, glib comment. I didn't realise it was quite that bad I must admit, I presumed it would make 7 or 8k. Is the ICG380's 6,000 and 12,000 dpi and Aztek's 8,000 similar?

Tim

[timparkin]

How about something in the test designed to create Moiré patterns with digital sensors? Since the IQ180 doesn't have a filter, I think it would be interesting to see how it renders patterns like that.

I'll print out this one if possible..

http://www.imatest.com/docs/images/small_chart_contrast_20.svg

[rdenney]

Will an 11,000dpi Hiedleberg do? I've also got a 100x stereo microscope to which I can attach my 5Dmk2

I've done this several ways, and the microscope is by far the best option. You have a very nice setup for recording the microscope image--I wish I'd had that.

Rick "who had a cheap 30x hand scope" Denney

[Joerg Krusche]

Tim,

for the other drum scanners I do not know .. please do not forget the b/w Fineart community .. TMX100 might be a good candidate .. let's hope the Kodak films still be available in the future !

had a look at your website .. beautiful indeed !

joerg

[wentbackward]

Tim, I didn't have time to read everything here. Long an admirer of your work though. Just a throwaway comment, but this seems to be a test of resolving power of various systems, not a study of what looks nice or preferable between the different systems/results. I think you can get some lens testing equipment to generate MTF charts and do the comparison completely scientifically with actual measurements.

- Using an Alpa with the Rodenstock 90 HR-W (similar to the 70), it is optimised for about f5.6 - f8. You may be doing this but I think you should compare the sweet spot, not a relatively similar aperture. This would imply that only a hyperfocal distance that covers the sweet spot of the lenses could really produce comparable results, else there will be too much variance between what is sharp and what is not.

- I'm new to MFDB's but stitching images from the 90 HR W on an Alpa Max gave me some results that popped out the way a 4x5 image does. The MFDB crop from the centre of the lens may be unfair from a resolution perspective, whilst using more of the image circle at lower resolution on an 8x10 is surely 'easier' on the system and would certainly ooze character, thus unfair from a feeling point of view.

If I can be selfish for a moment, what I would love to see is both systems taken to some gorgeous spot at the right time of day, with some lovely scenery beyond the hyperfocal distance of either system, then fired off at the same time. You could then proceed to do another test with numerous shifted MFDB shots later stitched.

That produces three comparable LF prints that can be compared for tonality, depth, feeling, emotion.

Forget all the test charts and microscopes. After taking the shots, send the lenses off to a lab, get the MTF charts, let the resolutionaries multiply them up by the size of the image circle

Now we have three prints and scientific data so we can actually see if there is any correlation between resolution and preference.

Just some thoughts!
Paul

[timparkin]

Tim, I didn't have time to read everything here. Long an admirer of your work though. Just a throwaway comment, but this seems to be a test of resolving power of various systems, not a study of what looks nice or preferable between the different systems/results. I think you can get some lens testing equipment to generate MTF charts and do the comparison completely scientifically with actual measurements.

- Using an Alpa with the Rodenstock 90 HR-W (similar to the 70), it is optimised for about f5.6 - f8. You may be doing this but I think you should compare the sweet spot, not a relatively similar aperture. This would imply that only a hyperfocal distance that covers the sweet spot of the lenses could really produce comparable results, else there will be too much variance between what is sharp and what is not.

- I'm new to MFDB's but stitching images from the 90 HR W on an Alpa Max gave me some results that popped out the way a 4x5 image does. The MFDB crop from the centre of the lens may be unfair from a resolution perspective, whilst using more of the image circle at lower resolution on an 8x10 is surely 'easier' on the system and would certainly ooze character, thus unfair from a feeling point of view.

If I can be selfish for a moment, what I would love to see is both systems taken to some gorgeous spot at the right time of day, with some lovely scenery beyond the hyperfocal distance of either system, then fired off at the same time. You could then proceed to do another test with numerous shifted MFDB shots later stitched.

That produces three comparable LF prints that can be compared for tonality, depth, feeling, emotion.

Forget all the test charts and microscopes. After taking the shots, send the lenses off to a lab, get the MTF charts, let the resolutionaries multiply them up by the size of the image circle

Now we have three prints and scientific data so we can actually see if there is any correlation between resolution and preference.

Just some thoughts!
Paul


Hi Paul,

You've probably missed some of the discussion but I have made it quite clear that I don't want this to be just a resolution test. The whole idea is to compare the full package. Hence why we'll be taking shots with some other cameras as well. However, whilst we are taking the pictures and have access to multiple cameras it would be remiss not to include some form of objective measurements and so I've been chatting with Norman Koren and Imatest about performing these (they have both been very helpful).

The final result will be shots from each system, once in a studio under controlled conditions with resolution targets and a still life scene, once in the garden at Joe's house which will be more like a landscape scene and finally, if we get time, we will pop around the corner to Roseberry Topping to take a 'scenic'.

We won't be stitching pictures as this is possible with both LF and MFDB systems (I've stitched 5x4 on multiple occasions using a sliding back to produce very, very sharp 4x10 images and for longer distance shots where parralex isn't an issue, stitching two or three 10x8's on a rotational tripod would not be too difficult (although i think you would get better results stitching 4x5's using a graphmatic).

There is little point getting the lenses tested as the system will be film limited more than likely and I hate to think how much lab lens testing would cost. Also, most lenses have mtf charts with them and Chris Perez has performed a mighty fine resolution comparison of a huge range of lenses (http://www.hevanet.com/cperez/testing.html).

The final results will be a set of images that people can download and compare and I want people to make subjective assessments on all sorts of criteria and report back.

Tim

[timparkin]

Tim,

for the other drum scanners I do not know .. please do not forget the b/w Fineart community .. TMX100 might be a good candidate .. let's hope the Kodak films still be available in the future !

had a look at your website .. beautiful indeed !

joerg

Thanks Joerg - we'll be shooting Delta 100 and also, if we can get it in time, a colleague has bought some Rollei ATO in 10x8 and 4x5 for us to use.

I have a feeling Kodak will be around for a while (if only because the cinematography industry has a lot of very wealthy people around it - Google have also expressed an interest, that would be nice!)

[rdenney]

Thanks Joerg - we'll be shooting Delta 100 and also, if we can get it in time, a colleague has bought some Rollei ATO in 10x8 and 4x5 for us to use.

Please do not limit the test to black and white. Much of what might be different between the two technologies, both quantitative and qualitative, will only be visible in color.

Rick "suspecting the digital defenders would cry foul at a black-and-white-only test" Denney

[timparkin]

Please do not limit the test to black and white. Much of what might be different between the two technologies, both quantitative and qualitative, will only be visible in color.

Rick "suspecting the digital defenders would cry foul at a black-and-white-only test" Denney

Good lord no! Here is the film for the test

1) Velvia 50 for resolution - Provia 100 or Astia for backup shots
2) Fuji Pro160S - Dynamic range for outdoor shot
3) Portra 400 (only one or two sheets) - even more dynamic range
4) Delta 100 - clean fine grain
5) Rollei ATO - ridiculousness..

Tim

p.s. Having checked out the MTF's - Astia looks interesting again. Of course my graph extrapolation may be crap...

http://static.timparkin.co.uk/static/tmp/Velvia-Astia-Provia-MTFs.jpg

[rdenney]

p.s. Having checked out the MTF's - Astia looks interesting again. Of course my graph extrapolation may be crap...

http://static.timparkin.co.uk/static/tmp/Velvia-Astia-Provia-MTFs.jpg

I doubt the 8x10 will get very close to the parts you extrapolated.

My only concern with Velvia is that it is so contrasty that your exposure of a test chart must be dead-on perfect not to blow out the highlights, which will render the chart useless. And all your exposures must be exactly the same. The interpretation of the charts will change if the exposure changes. I have made this mistake before, when the lenses I was testing varied somewhat in transmission and weren't that accurate in their aperture markings. The shutters on the test cameras were even worse.

Transparency film has the huge advantage of being viewable on a light table with your microscope. It is very difficult to directly evaluate contrast on a color negative because of the color mask and because there is so much density compression. But transparencies are designed for direct viewing. I'd find the slide film with the widest latitude I could. I only shoot Velvia when I shoot transparency film so I have little to suggest other than pointing to my own scars.

Rick "who would want the Velvia exposure to be within a tenth of a stop for reading those MTF charts" Denney

[John NYC]

Good lord no! Here is the film for the test

1) Velvia 50 for resolution - Provia 100 or Astia for backup shots
2) Fuji Pro160S - Dynamic range for outdoor shot
3) Portra 400 (only one or two sheets) - even more dynamic range
4) Delta 100 - clean fine grain
5) Rollei ATO - ridiculousness..

Tim

p.s. Having checked out the MTF's - Astia looks interesting again. Of course my graph extrapolation may be crap...

http://static.timparkin.co.uk/static/tmp/Velvia-Astia-Provia-MTFs.jpg

I also share Rick's concern about the awful dynamic range of slide film and comparing that to digital.

I personally use E100G (and Portra 400). E100G is amazing for slide film, much more modern than Velvia in terms of dynamic range.

[georgl]

A few cents from me:

- I have a Sinar adhesive film-holder. It`s obviously just an ordinary Fidelity-holder without the guidances normally holding the film but with an adhesive foil applied. Ì have no idea how they compensated the height difference due to the foil or if this is the reason why they were so expensive!? When you find a way to compensate the height (like applying a identically thick foil onto the front of the holder?), you could just buy adhesive foil (e.g. 3M) and apply it to a normal holder, right!?

- orthochromatic films (Agfa Ortho, Rollei ATO...) can be used to evaluate the setup (lens, holder, focus, scanner). They are actually grain-free even at 4000ppi and have excellent MTF (I guess at least similar to a 5µm-MFDB). The image might be contrasty, but SHARP (yes, at 100%). If not, something went wrong...

- an excellent choice for the scanner might be an Eversmart (everything but the Jazz), they are really sharp up to the pixel-level, even at 5600 ppi - grain aliasing, contrast loss - all this shouldn`t matter too much with these.

- I`ve seen wonderful results from the Tango and the whole drum scanner business is surrounded by myths. I`m not sure how the 10µm aperture actually affects resolution, but from personal experience I can say that Howteks are far from being perfect, either. The build-quality is lightyears apart - I could imagine that it might affect overall IQ as well.

- choosing a large subject for high image ratio (in Berlin I like the Reichstag for example, a frontal photograph has to cover more than 150m - image ratio about 1:600 even with 8x10"). DoF will be sufficient even when testing for diffraction with wide apertures (f11, f16). Cold, clear (atmosphere) winter nights are perfect for that ;-)

@ Tim
Thank you very much for this assignment, I`m looking forward to the results! Will it be the "once and for all end of discussion"-test? Will it be feasible for everyday photography? Most likely not - but right now, a comprehensive comparison between these high-end approaches are hard to find.

[pdmoylan]

I would love to have Tim incorporate significant movements into the equation to evalute lens performance in the corners; though it appears that none of the digital lenses chosen has that function.

[timparkin]

I would love to have Tim incorporate significant movements into the equation to evalute lens performance in the corners; though it appears that none of the digital lenses chosen has that function.

I was thinking of doing just this. I thought the Digaron W's were the best lenses for movements on the digital platform? They seem to provide 90-125mm of movement (probably half a frame up or down). I can compare it directly with the 150 sironar S on the Ebony which has a similar amount of movement in relation to the sensor size

[pdmoylan]

Sorry Tim. I miscued on the digital lens movements. These are the best choice as you say.

An eager observer.

[John NYC]

Tim,

Any results?

John

[timparkin]

Tim,

Any results?

John

Yep! Just compiling stuff - two weeks to the full article

[Asher Kelman]

Yep! Just compiling stuff - two weeks to the full article

Any hints? Can LF stand it's own under the best of circumstances?

Asher

[timparkin]

Any hints? Can LF stand it's own under the best of circumstances?

Asher

OK - a little hint. Large format photographers won't be dissapointed ;-)

[Nathan Potter]

Hardly need to do the test. Most of us can guess the results. Under ideal conditions of absolute stability and focus with top lenses and the lenses not diffraction limited the 8X10 will exceed the IQ180 by some bit as far as resolution is concerned. For other factors such as color fidelity, color fringing and such the IQ180 may be superior. It's the practical issues of things like vibration, focusing ability, DOF and the like that will differentiate the two formats.

But actual tests are reassuring and illustrative, so Tim keep grinding away. Many of us appreciate your efforts.

Nate Potter, Austin TX.

[Ken Lee]

I recommend that this thread be closed, and that Tim create a new one when he's ready to share his findings. This one has gotten a bit... protracted.

[timparkin]

I recommend that this thread be closed, and that Tim create a new one when he's ready to share his findings. This one has gotten a bit... protracted.

A fine idea :-)