Pages: 1 2 [3]   Go Down

Author Topic: Digiatl vs. film another comparison  (Read 15963 times)

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Digiatl vs. film another comparison
« Reply #40 on: August 20, 2009, 06:19:44 pm »

Hi!

A few comments:

1) Leica has a design target of 50-50-50. I know that 50-50 stands for 50% MTF at 50 lp/mm, but not what the last figure stands for.
2) Here is the best info I have on MF lenses: http://old.photodo.com/nav/prodindex.html
They don't have Pentax 80/2.8 but the Pentax 645 75/2.8 is like this:
http://old.photodo.com/prod/lens/detail/Pe...75_28-631.shtml
It achieves about 70% at 40 lp/mm on axis but falls rapidly off axis, this is fairly typical, even if some lenses perform better.
3) 50 lp/mm is significantly more than 40 lp/mm
4) Norman Koren shows an MTF curve for Velvia here, it's about 50% at 50 lp/mm here: http://www.normankoren.com/Tutorials/MTF1A.html

I have shot some new images and did different comparisons: http://83.177.178.241/ekr/index.php/photoa...-sony-alpha-900

In the above article I also measured MTF with Imatest and arrived at about 2300 LW/PH at 50% MTF. I estimated about 2700 LW/PH at 30% MTF from MTF data for film and lens, on axis, and the measured figure is quite close.

Images made with my Pentax 75/2.8, Dimage Scan Multi Pro, Vuescan and tools produce 70x100 cm (27 x 39") prints that impress experts with experince from working at the best professional labs of Sweden, and yes they look awful when displayed actual pixels. I did not make any big prints from my test images.

Best regards
Erik



Quote from: georgl
High contrast up to high frequencies is one of the biggest advantages of digital and therefore sharpness perception. But these 6x7 don't seem to contain much information at all, not even at low contrast!

Very problematic for those comparisons is the fact that the negative/slide itself isn't usable, you need another optical/chemical process to create a print or a "digital camera" like a scanner. It's MTF is multiplied with the source-MTF in results in a lower MTF in the end.

But when properly done and processed, a scan from modern film is capable of much more than 6MP @ 24x36mm or 24MP @ 6x7, which is well shown in several scans also available on the net.

Here is one of those (but done with a tiny Minolta-lens): http://www.boeringa.demon.nl/menu_technic_..._resolution.htm
40lp/mm easily resolved!

Modern Leica-lenses resolve about 80% contrast at 40lp/mm, high-quality medium-format lenses a little bit less (60-70%) but as I said, I have no idea what this Pentax is capable of!? But we should be careful making general assumptions "film vs. digital", it's sad to see that powerful tools like medium-format-cameras aren't appreciated because of that.
« Last Edit: August 20, 2009, 06:21:54 pm by ErikKaffehr »
Logged
Erik Kaffehr
 

Anders_HK

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1010
    • andersloof.com
Digiatl vs. film another comparison
« Reply #41 on: August 21, 2009, 12:01:36 am »

Quote from: georgl
Never had a Pentax 67, but these 6x7-scans look awful and I don't think they do the mdium film justice! A 6x7-slide/negative has to resolve only about 40lp/mm to match a 24MP-DSLR!
This one looks better: http://www.rockgarden.net/download/60MP_from_6x7/

... and that was on an Imacon, not a high quality drum scanner...  

@ Erik, I am Swedish. Lets face it, our country is a small one with 9 million people.... Back in 2003 (when more people shot film in Sweden) I had several 35mm Velvia 50 slides scanned in one of the top labs in Goteborg - results were not good, including they had problem with the colors...

Although I no more live in Sweden, how then do the labs in Sweden nowadays stack up to professional labs in big metropolitan areas in the world where there is still some market left for film? Unfortunate, per my own experience there are not that many professional and capable labs for film even in Hong Kong, Shanghai and Seoul nowadays - this is the unfortunate downside of film, but some labs do exist.

The positive side of film is that the images do not cease to impress people. Now that the world is digital that is perhaps even more true..., because different but at very high quality.  

With a Leica M8 and Leaf Aptus 65 I still shoot film... perhaps I should have stayed with it? The reason to switching digital in first place was that a "professional lab" in Busan South Korea in 2005 handed my three Velvia 50 slides back with fine scratches running across them...

Depending on subject, light etc, quality scanned 6x7, 617 and 4x5 beats my Aptus... as for which one is best... film and quality digital are different. Yes, at times I might like a shot from my Aptus better than from my 6x7, but far from always. The argument of 24MP dslrs beating 6x7 seem very lame... I cannot even say my Aptus does, and the 24MP DSLRs are of lesser quality...

Or to be more fair in comparison... instead of basing on substandard scanning, how about projecting the digital image from a dslr on a wall to compare with a quality shot and processed 6x7 slide with a pro 6x7 projector? I bet you $$$ on the slide will win.  

If Epson update their V750 I will buy one.  

Regards
Anders
« Last Edit: August 21, 2009, 12:15:45 am by Anders_HK »
Logged

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Digiatl vs. film another comparison
« Reply #42 on: August 21, 2009, 12:41:24 am »

Hi Anders,

I'm not making an argument that 24 MP is better than 67. I just published two images shot significant time apart. That said, I had done some more tests and also done some more "math" and may now have an opinion of my own.

Regarding what you write I don't have any issues with the point you make. Except perhaps mentioning that I have a professional 67 projector, a Götschman 67 with a Schneider 150/2.8. I see more detail (and aberrations) on the scan than on the projected image. Image flatness is an issue in projection even using all glass mounting in GP-slide mounts. Doing a projection comparison is not possible because to my best knowledge there are no LCD projectors having 6000x4000 resolution. So it leaves us with comparing on screen and print.

Resolution/sharpness is just one issue. Visual impact is another. You have experience of high end digital and high end scanning and are in the position to make judgments most of us can't.

I have some more testing here: http://83.177.178.241/ekr/index.php/photoa...-sony-alpha-900

I'm pretty sure that there are some issues. The lens I have on the Pentax is for sure not one of the best ever built. The Pentax is old and focusing has not been calibrated. Film flatness is always an issue on 120.

Regarding my own experience with scanned 67 film it's essentially is that:

- It enlarges well
- But is noisy. My scanner is known to enhance grain patterns.
- My scanner cannot handle the desity ov Velvia


Best regards
Erik



Quote from: Anders_HK
... and that was on an Imacon, not a high quality drum scanner...  

@ Erik, I am Swedish. Lets face it, our country is a small one with 9 million people.... Back in 2003 (when more people shot film in Sweden) I had several 35mm Velvia 50 slides scanned in one of the top labs in Goteborg - results were not good, including they had problem with the colors...

How then do the labs in Sweden nowadays stack up to professional labs in big metropolitan areas in the world where there is still some market left for film? Unfortunate, per my own experience there are not that many professional and capable labs for film even in Hong Kong, Shanghai and Seoul nowadays - this is the unfortunate downside of film, but some labs do exist.

The positive side of film is that the images do not cease to impress people. Now that the world is digital that is perhaps even more true..., because different at very high quality.  

With a Leica M8 and Leaf Aptus 65 I still shoot film... perhaps I should have stayed with it? The reason to switching digital in first place was that a "professional lab" in Busan South Korea in 2005 handed my three Velvia 50 slides back with fine scratches running across them...

Depending on subject, light etc, quality scanned 6x7, 617 and 4x5 I shot beats my Aptus... as for which one is best... film and quality digital are different. The argument of 24MP dslrs beating 6x7 seem very lame... I cannot even say my Aptus does, and the 24MP DSLRs are of lesser quality...

Or to be more fair in comparison... instead of basing on substandard scanning, how about projecting the digital image from a dslr on a wall to compare with a quality shot and processed 6x7 slide with a pro 6x7 projector? I bet you $$$ on the slide will win.  

If Epson update their V750 I will buy one.  

Regards
Anders
« Last Edit: August 21, 2009, 01:44:03 am by ErikKaffehr »
Logged
Erik Kaffehr
 

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Digiatl vs. film another comparison
« Reply #43 on: August 21, 2009, 01:38:44 am »

Hi,

This excellent article by Reichmann, Atkinson and Cramer is still relevant:

http://www.luminous-landscape.com/essays/back-testing.shtml

Best regards
Erik
Logged
Erik Kaffehr
 

ashley

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 94
    • http://www.ashleykaryl.com
Digiatl vs. film another comparison
« Reply #44 on: August 21, 2009, 04:09:36 am »

I must have heard these debates about sharpness, resolution and means of measurement a thousand times but maths has never really been my strong point. You can't look at an image with a calculator and the idea of physically trying to measure these things rather bemuses me. Surely the only real measure is what it looks like when published in the chosen media? This may be subjective, but to me that is the only worthwhile comparison that exists.
Logged
Ashley Karyl Photographer
[url=http://w

Anders_HK

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1010
    • andersloof.com
Digiatl vs. film another comparison
« Reply #45 on: August 21, 2009, 04:34:54 am »

Quote from: ErikKaffehr
The lens I have on the Pentax is for sure not one of the best ever built. The Pentax is old and focusing has not been calibrated. Film flatness is always an issue on 120.

Regarding my own experience with scanned 67 film it's essentially is that:

- It enlarges well
- But is noisy. My scanner is known to enhance grain patterns.
- My scanner cannot handle the desity ov Velvia


Best regards
Erik

Hej Erik,

In all politeness, in order to compare 6x7 to your Sony, then you should show each to their best. Above does not speak for the best of the 6x7... thus it is not a comparison on equal terms and does not tell correctly of digital vs. film.

F.w.i.w. I have never experienced issue with film flatness on 6x7 but it might be a camera issue. If I recall correct the Pentax 67 is a very reputable camera, but when a camera is old or after much use it may be require a check up and adjustments?

Regards
Anders
Logged

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Digiatl vs. film another comparison
« Reply #46 on: August 21, 2009, 05:20:02 am »

Hej Anders,

In all politeness, I agree on doing the best. But..

The Pentax has a good reputation but if you check out the old photodo site it seems that Pentax lenses do not belong to the best, there are no tests on Pentax 67 lenses but I looked at 645 lenses. I have also seen some tests in Color Photo long ago which indicated that the Pentax lenses were OK but not as good as the best MF lenses. There is a lot of development in MF, like the new HC-lenses for Hasselblad and "digital" lenses from Schneider and Rodenstock. I presume that many of the new lenses are better than my than the Pentax 90/2.8 lens which probably was designed in the seventies.

All my Pentax lenses have significant CA except the one I'm using for the test. I also spent significant effort on looking for errors and also did some MTF measurements using Imatest, which indicates that my results are about what could be expected from Velvia (ignoring scanner MTF and sharpening).

Velvia 50 has an MTF curve like this: http://www.normankoren.com/Tutorials/MTF_Velvia.gif it drops rapidly at around 50 lp/mm. So in all probability Velvia itself limits sharpness. Resolution may be high but tonal separation gets poor. So putting a Rodenstock Digitar in front of Velvia may not improve resolution a lot. If the "Digitar" has better MTF at low frequencies it may make the picture much sharper.

Here is an interesting comparison between Leica M7 and M8:
http://www.imx.nl/photo/leica/camera/M8/M8/page22.html
One observation may be that choice of film matters. The author of the above article is Erwin Puts, he seems to do a lot of lens evaluation work for Leica. He is always a good but not necessarily easy read.

Regarding my test and comments:

- I point out possible problem areas, that is not to say that I'm sloppy.

- That film flatness is an issue has been known for long time. Contax even had a vacuum plate to keep the film flat.

- I try pretty much not to "tilt" the comparison.

A question:

Why don't you yourself post some comparison images between film and digital? You are using both.


Best regards
Erik

Quote from: Anders_HK
Hej Erik,

In all politeness, in order to compare 6x7 to your Sony, then you should show each to their best. Above does not speak for the best of the 6x7... thus it is not a comparison on equal terms and does not tell correctly of digital vs. film.

F.w.i.w. I have never experienced issue with film flatness on 6x7 but it might be a camera issue. If I recall correct the Pentax 67 is a very reputable camera, but when a camera is old or after much use it may be require a check up and adjustments?

Regards
Anders
« Last Edit: August 22, 2009, 04:37:01 am by ErikKaffehr »
Logged
Erik Kaffehr
 

georgl

  • Full Member
  • ***
  • Offline Offline
  • Posts: 140
Digiatl vs. film another comparison
« Reply #47 on: August 21, 2009, 06:31:53 am »

Never heard of the "Leica design-target 50-50-50", propably it means 50% image height because all modern Leica lenses achieve significantly higher contrast?
The Rodenstock Digaron-W-lenses have similar MTFs as Leica-lenses but they also show the response for 80lp/mm and have a image circle large enough for 6x7! But I don't know how they perform on film, Schneider offers an additional glass to "simulate" the cover-glass of a sensor with film.

They achieve 70% over 6x7-image-height at 40lp/mm and nearly 50% at 80lp/mm!

As I see it, it's a test of a Pentax 67 + lens + consumer-scanner, not film & digital per se.

I had a 35mm-Pentax-system, the lenses were mush in comparison to my Leica-lenses, how much does vibration of the 67 affect IQ (does it even have mirror-lock?). I have many bad scans of slides/negatives, only a few show the potencial of this medium.

Norman Koren has an interesting site and even he claims an MTF50 of Velvia with over 56lp/mm!



Logged

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Digiatl vs. film another comparison
« Reply #48 on: August 21, 2009, 11:29:33 am »

Hi,

There are some sample images in the first article. Sorry about the math. If you look at the images, please keep in mind that you see very small crops. The images would be about 1.5-2 m wide, that 4.5-6 feet. So we are pixel peeping. The best way to compare is probably to make prints, about 70x100 cm, but prints are not easily shared.

One reason for the math is to check if the equipment is close to specifications or not.

Best regards
Erik

Quote from: ashley
I must have heard these debates about sharpness, resolution and means of measurement a thousand times but maths has never really been my strong point. You can't look at an image with a calculator and the idea of physically trying to measure these things rather bemuses me. Surely the only real measure is what it looks like when published in the chosen media? This may be subjective, but to me that is the only worthwhile comparison that exists.
Logged
Erik Kaffehr
 

ashley

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 94
    • http://www.ashleykaryl.com
Digiatl vs. film another comparison
« Reply #49 on: August 21, 2009, 11:35:43 am »

Quote from: ErikKaffehr
Hi,

There are some sample images in the first article. Sorry about the math. If you look at the images, please keep in mind that you see very small crops. The images would be about 1.5-2 m wide, that 4.5-6 feet. So we are pixel peeping. The best way to compare is probably to make prints, about 70x100 cm, but prints are not easily shared.

One reason for the math is to check if the equipment is close to specifications or not.

Best regards
Erik

Hi Erik,
I hope you didn't think I was criticising your efforts. That was not my intention at all 8-) I was merely trying to say what you have just pointed out, that this really is just pixel peeping and that the real test comes from prints but of course that is not really practical on the internet. Many of the perceived differences seen on screen between one digital camera or another or even film V digital will often be totally invisible once the image is printed.

Ashley
Logged
Ashley Karyl Photographer
[url=http://w

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Digiatl vs. film another comparison
« Reply #50 on: August 21, 2009, 12:02:41 pm »

Hi,

I read about the 50-50-50 target in one of Erwin Puts articles, but I cannot find it. I may use it in the wrong context. Now to your comments.
[Update]
This is the statement, I believe. Leica themselves have designed the whole system for a 50 lp/mm at 50% contrast, the benchmark value for excellent imagery.
It is about Leica M8, so it is out of context. Still it indicates that 50% MTF at 50 lp/mm is a very good value:
The article is here: http://www.imx.nl/photo/leica/camera/M8/M8/page20.html


The figure for Velvia having an MTF is taken from the Norman Koren page, http://www.normankoren.com/Tutorials/MTF_Velvia.gif , it's quite obviously 50% at 50 lp/mm.
MTF values can be multiplied. So if you have 70% MTF at 50 lp/mm for lens and 50% MTF for film you end up with 35% MTF. If you enlarge this with a very good enlarging lens having 70% MTF at 50 lp/mm you and up with 0.7 * 0.5 * 0. * 100 = 24.5% MTF at 50 lp/mm. This all assuming that lens and focusing device is in perfect alignment.

With digital workflow you loose MTF in scanner but can regain some with sharpening.

The scanner I use is not really a low end device, here is a comaprison with Imacon:

http://web.tiscali.it/saphoto/scancomp.html

and comparison images here:

http://web.tiscali.it/saphoto/scanimages.html

You are of course right on the issue that it's a Pentax 67 + lens + scanner combination compared to a Sony + lens combination, but that applies to pretty every comparison. I may also suggest that probably more people have Pentax 67 and CCD film scanners than those have Rodenstock Digarons and drum scanners.

There are a lot of QA issues in medium format, you may check this article:

http://www.josephholmes.com/news-medformatprecision.html

The same applies to DSLRs, for sure.

By the way, you would not consider posting some comparison images yourself? Sure it would be appreciated!

Best regards
Erik

Quote from: georgl
Never heard of the "Leica design-target 50-50-50", propably it means 50% image height because all modern Leica lenses achieve significantly higher contrast?
The Rodenstock Digaron-W-lenses have similar MTFs as Leica-lenses but they also show the response for 80lp/mm and have a image circle large enough for 6x7! But I don't know how they perform on film, Schneider offers an additional glass to "simulate" the cover-glass of a sensor with film.

They achieve 70% over 6x7-image-height at 40lp/mm and nearly 50% at 80lp/mm!

As I see it, it's a test of a Pentax 67 + lens + consumer-scanner, not film & digital per se.

I had a 35mm-Pentax-system, the lenses were mush in comparison to my Leica-lenses, how much does vibration of the 67 affect IQ (does it even have mirror-lock?). I have many bad scans of slides/negatives, only a few show the potencial of this medium.

Norman Koren has an interesting site and even he claims an MTF50 of Velvia with over 56lp/mm!
« Last Edit: August 22, 2009, 01:21:12 am by ErikKaffehr »
Logged
Erik Kaffehr
 

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Digiatl vs. film another comparison
« Reply #51 on: August 21, 2009, 12:16:22 pm »

Ashley,

Nope, no issue with me. On the contrary I agree with your points. I actually feel bad about math, in this context. As I said, there is some need to verify that the tests are not fooled up. Lot of things can go wrong. Comparing measured data with expected may indicate if there are gross errors. I would love to send around my large prints, but it's unfortunately not easily doable.

I'm in the process of publishing a longer article: http://83.177.178.241/ekr/index.php/photoa...-sony-alpha-900

My own findings are mixed. On the page mentioned above I look at some very small detail and there the Pentax is better. The scanned image is quite noisy, but some noise/grain/structure is a characteristic of film. That article is ongoing. I'll probably also add some more images in the future.

One of the interesting things is that testing does not always come out as expected. When I got my new Alpha 900 I did some comparison shooting with my Alpha 700. When pixel peeping the Alpha 900 images just blew away the Alpha 700. When I printed in A2 from Lightroom I could not tell them apart! That comparison is published here:

Actual pixels of image:
http://www.pbase.com/ekr/image/107619976/original
Actual pixels 300 PPI scan of print:
http://www.pbase.com/ekr/image/107823207/original

Scans were also made at 600 PPI but those files were just to impractical. The prints hade smaller visible differences than the scans.

I hope that my response was not rude, it was not my intention. But I may have been in some hurry.

Best regards
Erik

Quote from: ashley
Hi Erik,
I hope you didn't think I was criticising your efforts. That was not my intention at all 8-) I was merely trying to say what you have just pointed out, that this really is just pixel peeping and that the real test comes from prints but of course that is not really practical on the internet. Many of the perceived differences seen on screen between one digital camera or another or even film V digital will often be totally invisible once the image is printed.

Ashley
« Last Edit: August 21, 2009, 01:00:21 pm by ErikKaffehr »
Logged
Erik Kaffehr
 

georgl

  • Full Member
  • ***
  • Offline Offline
  • Posts: 140
Digiatl vs. film another comparison
« Reply #52 on: August 21, 2009, 01:03:49 pm »

I have some 6x7-velvias, but they're completely unsharp because they were made with a rented RB67, what a lousy camera - no mirror-lock -> work of two days ruined!
I tried a real half-way scientic comparisons but all I got from the guy who scanned them with his imacon were unsharp results (I have no idea what he did to this ppor machine!) - I'll try to make a new one within the next months.

Anyway, here's a Imacon-scan from a Velvia-slide (made with 35mm Summicron Asph) which I printed and was very happy with the result:
[attachment=16158:duisburg3.jpg][attachment=16159:duisburgcrop.jpg]

 It looked different than my M8-files, but no way less sharp! Older Ilfochromes had a great look, but they looked terribly soft!

Don't you think a decent 6x7-velvia with 4.5times this size should look much better?

And here's a 3200ppi-Imacon-scan from the same series shot with my belove Hassi 501CM + 80mm Planar (which is not as sharp as the Summicron) :[attachment=16160:duisburg2.jpg][attachment=16161:duisburg2detail.jpg]

I simply think it's strange that so many different "scientific comparisons" floating around the net, Mr. Reichmann even claimed his 11MP 1ds would be as good as drum-scanned 6x7, that would mean a 24x36mm-slide is about as good as a 3MP-DSLR...! New DSLRs are better than Velvia, no better than 645, better than 6x7, better than 4x5inch!
« Last Edit: August 21, 2009, 01:10:20 pm by georgl »
Logged

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Digiatl vs. film another comparison
« Reply #53 on: August 21, 2009, 01:22:51 pm »

Hi,

No, because we are looking at actual pixels. The Leica Lens is probably much sharper than my Pentax 67 90/2.8. The pixels are probably the same size as both you and I scan at 3200 PPI. I just have 4.5 times as many of them. One issue that I sharpened the slide image using Photokit Sharpener which has setting for medium format film.

The idea was to use the same sharpening I'd use for that kind of subject and with film it would be Photokit Sharpener and with digital "Landscape".

Thanks for publishing samples!

Best regards
Erik

Quote from: georgl
Don't you think a decent 6x7-velvia with 4.5times this size should look much better?
Logged
Erik Kaffehr
 

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Digiatl vs. film another comparison
« Reply #54 on: August 22, 2009, 01:25:20 am »

I found the the reference to 50-50:

This is the statement, I believe. Leica themselves have designed the whole system for a 50 lp/mm at 50% contrast, the benchmark value for excellent imagery.
It is about Leica M8, so it is out of context. Still it indicates that 50% MTF at 50 lp/mm is a very good value:
The article is here: http://www.imx.nl/photo/leica/camera/M8/M8/page20.html

There may be other references which I have not found.

You may find this article interesting: http://www.imx.nl/photo/leica/camera/M8/M8/page22.html , BTW.

Best regards
Erik


Quote from: georgl
I have some 6x7-velvias, but they're completely unsharp because they were made with a rented RB67, what a lousy camera - no mirror-lock -> work of two days ruined!
I tried a real half-way scientic comparisons but all I got from the guy who scanned them with his imacon were unsharp results (I have no idea what he did to this ppor machine!) - I'll try to make a new one within the next months.

Anyway, here's a Imacon-scan from a Velvia-slide (made with 35mm Summicron Asph) which I printed and was very happy with the result:
[attachment=16158:duisburg3.jpg][attachment=16159:duisburgcrop.jpg]

 It looked different than my M8-files, but no way less sharp! Older Ilfochromes had a great look, but they looked terribly soft!

Don't you think a decent 6x7-velvia with 4.5times this size should look much better?

And here's a 3200ppi-Imacon-scan from the same series shot with my belove Hassi 501CM + 80mm Planar (which is not as sharp as the Summicron) :[attachment=16160:duisburg2.jpg][attachment=16161:duisburg2detail.jpg]

I simply think it's strange that so many different "scientific comparisons" floating around the net, Mr. Reichmann even claimed his 11MP 1ds would be as good as drum-scanned 6x7, that would mean a 24x36mm-slide is about as good as a 3MP-DSLR...! New DSLRs are better than Velvia, no better than 645, better than 6x7, better than 4x5inch!
« Last Edit: August 22, 2009, 03:41:16 am by ErikKaffehr »
Logged
Erik Kaffehr
 

georgl

  • Full Member
  • ***
  • Offline Offline
  • Posts: 140
Digiatl vs. film another comparison
« Reply #55 on: August 22, 2009, 03:42:44 am »

I think we can compare two things:

MTF from sensor + processing vs. MTF from film + scanner-MTF - the lens doesn't matter here anymore, but we have to use the same lens in this comparison.

The 35mm-Velvia is scanned with 6300ppi, therefore you watch crops from a 50+MP-file!

Anyway, I think if we want to have serious conclusions, we'll have to change certain things ;-)

By the way, I just noticed that my Hassy-shot isn't ideal, I used hyperfocal distance...

Logged

ashley

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 94
    • http://www.ashleykaryl.com
Digiatl vs. film another comparison
« Reply #56 on: August 22, 2009, 04:47:25 am »

At the risk of taking this discussion in a different direction, in real world situations we don't often get to compare two different versions of an image side by side unless we are pixel peeping. As Erik pointed out earlier, he once compared his A700 to the A900 and on screen the A900 looked hugely better but in print he could not tell them a part. It is the medium of print which is the great leveller here so even if one medium has a huge theoretical advantage over another in practice it will be null and void in reality.

My take on this now is that if you can produce a good looking final print of your chosen subject it doesn't matter in the slightest if you shot film or digital. Overall I think 6x7 film is probably still capable of producing finer detailed big prints than a 35mm DSLR when everything is perfect and pushed to the limit but far too often the set up is not perfect and the differences will be so close that you might just as easily prefer the digital.

Reasons for shooting with the Pentax or the Sony would be better seen in the context of which was the more suitable tool for the job at hand. For example I generally don't much like the proportions of 35mm and much prefer medium format in that respect. I used to work with my Hasselblad 501CM and have two little red lines drawn on the viewfinder screen to indicate 645 proportions which were perfect for the magazines. There is no way that the Pentax could be used as quickly as the DSLR though and it can't show a client images within a couple of minutes either. There are so many variables involved in the processing and printing of an image that both film and digital can either look wonderful or terrible.
« Last Edit: August 22, 2009, 04:49:33 am by ashley »
Logged
Ashley Karyl Photographer
[url=http://w

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Digiatl vs. film another comparison
« Reply #57 on: August 22, 2009, 09:11:51 am »

Hi,

Well said!

Erik

Quote from: ashley
At the risk of taking this discussion in a different direction, in real world situations we don't often get to compare two different versions of an image side by side unless we are pixel peeping. As Erik pointed out earlier, he once compared his A700 to the A900 and on screen the A900 looked hugely better but in print he could not tell them a part. It is the medium of print which is the great leveller here so even if one medium has a huge theoretical advantage over another in practice it will be null and void in reality.

My take on this now is that if you can produce a good looking final print of your chosen subject it doesn't matter in the slightest if you shot film or digital. Overall I think 6x7 film is probably still capable of producing finer detailed big prints than a 35mm DSLR when everything is perfect and pushed to the limit but far too often the set up is not perfect and the differences will be so close that you might just as easily prefer the digital.

Reasons for shooting with the Pentax or the Sony would be better seen in the context of which was the more suitable tool for the job at hand. For example I generally don't much like the proportions of 35mm and much prefer medium format in that respect. I used to work with my Hasselblad 501CM and have two little red lines drawn on the viewfinder screen to indicate 645 proportions which were perfect for the magazines. There is no way that the Pentax could be used as quickly as the DSLR though and it can't show a client images within a couple of minutes either. There are so many variables involved in the processing and printing of an image that both film and digital can either look wonderful or terrible.
Logged
Erik Kaffehr
 

Anders_HK

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1010
    • andersloof.com
Digiatl vs. film another comparison
« Reply #58 on: August 22, 2009, 11:36:41 am »

Quote from: ErikKaffehr
A question:

Why don't you yourself post some comparison images between film and digital? You are using both.

Erik,

... I did. Same as thread linked #11 above http://luminous-landscape.com/forum/index....&pid=153549

Mine was comparing each to its best - first the ZD images, my processing and raw's for posters to process, last drum scans of 6x7. Post #64 in that thread summarize by stating basically same observation as I made of test prints at high resoution; "The film scans have a real wonderful quality to them. While looking at these last 100% crops, I find that I'm really drawn to them and the ZD images that looked so great before look lifeless in comparison."

At high ISO I am sure the 20+MPs dslrs beat ZD, but not at low..., different sensor designs, different sensor sizes/formats.

 

Rgds
Anders
« Last Edit: August 22, 2009, 11:41:51 am by Anders_HK »
Logged

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Digiatl vs. film another comparison
« Reply #59 on: August 22, 2009, 01:47:29 pm »

Anders,

I have seen the images but never knew that you also shot on film. I guess that drum scanning really can extract lot of detail that a CCD scanner can't. Whatever the media very good pictures.

Best regards
Erik

Quote from: Anders_HK
Erik,

... I did. Same as thread linked #11 above http://luminous-landscape.com/forum/index....&pid=153549

Mine was comparing each to its best - first the ZD images, my processing and raw's for posters to process, last drum scans of 6x7. Post #64 in that thread summarize by stating basically same observation as I made of test prints at high resoution; "The film scans have a real wonderful quality to them. While looking at these last 100% crops, I find that I'm really drawn to them and the ZD images that looked so great before look lifeless in comparison."

At high ISO I am sure the 20+MPs dslrs beat ZD, but not at low..., different sensor designs, different sensor sizes/formats.

 

Rgds
Anders
Logged
Erik Kaffehr
 
Pages: 1 2 [3]   Go Up