A surprisingly good film/digital comparison at "OnLandscape"

[Noll]

I haven't seen this talked about yet, if it has then my apologies. Just gave a quick read to this lengthy comparison posted at:

http://www.landscapegb.com/2011/12/big-camera-comparison/

I haven't read all 3 articles yet, but there's a ton to digest here.
http://www.landscapegb.com/2011/12/big-camera-comparison-comments/
http://www.landscapegb.com/2011/12/camera-test-editors-commentary/

Most importantly, it vindicates how a lot of people feel about film, especially with respect to detail. Almost as important, it lets the Luminous Landscape folk eat a little crow on their prior 6x7/4x5/8x10 v Phase One comparisons.

It also vindicates my frustration over a lack of decent enlarging options (low quality of consumer scanners vs lack of darkroom access for optical enlargement.) I guess some things don't change!

 

[shadowfox]

Wow, that makes D3x looks horrible. 🙂

 

[Trius]

Very interesting.

 

[Sarcophilus Harrisii]

I haven't seen this talked about yet, if it has then my apologies. Just gave a quick read to this lengthy comparison posted at:

http://www.landscapegb.com/2011/12/big-camera-comparison/

I haven't read all 3 articles yet, but there's a ton to digest here.
http://www.landscapegb.com/2011/12/big-camera-comparison-comments/
http://www.landscapegb.com/2011/12/camera-test-editors-commentary/

Most importantly, it vindicates how a lot of people feel about film, especially with respect to detail. Almost as important, it lets the Luminous Landscape folk eat a little crow on their prior 6x7/4x5/8x10 v Phase One comparisons.

It also vindicates my frustration over a lack of decent enlarging options (low quality of consumer scanners vs lack of darkroom access for optical enlargement.) I guess some things don't change!

Interesting link. Thanks for posting it.

Speaking of the Ludicrous Landscape site; another anti-film diatribe surfaced recently by way of a review of some software titled "DxO Film Pack".

From the article:

My second impression is to once again confirm how truly poor film based imaging is / was compared to todays' digital capture. Using a variety of images I went through every available colour transparency and negative emulsion looking for one that appealed to me more than the original processed with my usual workflow. Not a single one even came close.
​

It would seem this opinion has been expressed after converting digital captures with the aforementioned software.

Surely, a more informed point of view might be reached by actually sampling images captured on film; not by making value judgements about film quality, on the basis of film emulation software applied to digital captures which seems poor methodology indeed. Am I missing something, here?

Regards,
Brett

 

[Aristophanes]

What L Landscapes really meant was he prefers the uniformly sterile, pixel-perfect images from digital, so he can spend x x 100 hours per image making them look "unique".

 

[HHPhoto]

I haven't seen this talked about yet, if it has then my apologies. Just gave a quick read to this lengthy comparison posted at:

http://www.landscapegb.com/2011/12/big-camera-comparison/

I haven't read all 3 articles yet, but there's a ton to digest here.
http://www.landscapegb.com/2011/12/big-camera-comparison-comments/
http://www.landscapegb.com/2011/12/camera-test-editors-commentary/

Most importantly, it vindicates how a lot of people feel about film, especially with respect to detail. Almost as important, it lets the Luminous Landscape folk eat a little crow on their prior 6x7/4x5/8x10 v Phase One comparisons.

It also vindicates my frustration over a lack of decent enlarging options (low quality of consumer scanners vs lack of darkroom access for optical enlargement.) I guess some things don't change!

Yes, that is indeed one of the best comparison tests made.
Very good work.

In the past the internet and the photomagazines were flooded with these nonsense tests "scanned at 4000 ppi with amateur scanner" vs. digital sensors.
And then they said it was a "film vs. digital" test. BS. It was not.
It was a digital (scanner) vs. digital test.
And 4000 ppi scanners can not exploit the full potential of film, not at all. The quality loss with 4000 ppi scanners is huge.

Zeiss did several real film tests and published them in their camera lens news. Their results:
Spur Orthopan UR: 400 linepairs per millimeter (lp/mm)
Agfaortho 25: 250 lp/mm
Agfa APX 25: 200 lp/mm
Kodak TMX: 180 lp/mm
Fuji Acros 100: 160 Lp/mm
Fuji Velvia: 170 lp/mm

With a 24 MP FF sensor the maximal resolution is 85 lp/mm. That is the Nyquist frequency, the absolut physical limit which cannot be surpassed.
In reality most 24 MP sensors have a real resolution of 65 - 75 lp/mm because of the AA filter which is reducing resolution.

Some photographers and me have tested the D3x in comparisons to different films:
We've got 70 - 75 lp/mm with the D3x.
We've got 120 - 140 lp/mm with ISO 100 slide films.
We've reached 115 - 150 lp/mm with ISO 100 T-grain BW films.
And we've got 160 - 250 lp/mm with high resolution BW films.

Our test results have been a bit lower compared to Zeiss because we've used a lower object contrast.

There have been other tests published by Antora; Seemann, Serger; Ventzke with similar or even better results for film.
And all these tests included scanning, optical printing and slide projection:
With optical printing and slide projection significantly better detail rendition is achieved compared to 8000 ppi drum scans and especially 4000 ppi amateur scanners.
Drum scanners are limited at 110 lp/mm with medium object contrast, 4000 ppi at 70 lp/mm.

But one important area lacks in Tim Parkins test: slide projection.
With slide projection the best quality at huge enlargemt sizes is achieved. Best resolution, best sharpness, best brillance.

Digital projection can't compete at all with slide projection, because of the extremely low resolution of the beamers (only 2 MP in horizontal direction, even 40% less in vertical direction) and the bad color rendition.
In digital projection you have the situation that you pay 7000€ for a 24 MP D3x, and then you have to pay another 7000€ for a beamer to get the 24 MP of the camera smashed down to 2 MP of the beamer.
What a ridicolous waste of money.
We've did the direct comparison of 35mm slide projection and the most expensive 2 MP 7000€ beamers.
The slide is a league of it's own, no chance at all for the beamer.

Cheers, Jan

 

[pvdhaar]

Am I missing something, here?

Yes.. you're missing that LL runs on money from tutorials and workshops on digital flows..

 

[Sarcophilus Harrisii]

Yes.. you're missing that LL runs on money from tutorials and workshops on digital flows..

Yes, well, that much has been clear to me for some time now...still, the less than rigorous approach they take to their film v digital comparisons continues to disappoint.
Regards,
Brett

 

[Aristophanes]

Yes, well, that much has been clear to me for some time now...still, the less than rigorous approach they take to their film v digital comparisons continues to disappoint.
Regards,
Brett

They are, for the most part, honest in their bias.

Even the comments from those doing the LF tests here:

http://www.onlandscape.co.uk/2011/12/big-camera-comparison-comments/

...readily admit that for commercial work, digital is the only cost-effective system that delivers enough quality (maybe not superlative) on time for clients. If someone can make a go of it using film, excellent. But the cold economic reality is most cannot. Everyone in the review and from LL basically agree on this point.

At a certain point, unless you are printing gallery size photos, comparisons like this are the equivalent of discussing how many angels can dance on the head of a pin.

 

[HHPhoto]

...readily admit that for commercial work, digital is the only cost-effective system that delivers enough quality (maybe not superlative) on time for clients. If someone can make a go of it using film, excellent. But the cold economic reality is most cannot. Everyone in the review and from LL basically agree on this point.

That is not true in general. There are lots of professionals using film as primarily medium.
I know some who shoot film not only because of quality reasons, but also because of economic reasons.
Lots of pros can't afford a digital medium format back and the necessary hardware.
They have less costs using film.

Even Imacon / Hasselblad has said this quite clearly: They said the total global market for medium format digibacks is only about 5000 units per year worldwide.
It is a very tiny "niche in the niche in the niche" market.
There is a reason why manufacturers of digital medium format backs have so much financial trouble (some even had to leave the market).
For example the insolvency of Franke&Heidecke (Rolleiflex) was caused by the collapsing demand for the Jenoptik digital backs (Hy6).
The successor company DHW is therefore now mainly focussing on film based cameras: Rolleiflex 2,8 FX, W and T, Rolleiflex 6008, Hy6 with 4,5x6 and 6x6 film backs, Rollei 35, Rolleivision Twin MSC and Dual 66P slide projectors.
At last Photokina their rep told me demand for their film based medium format cameras is quite strong in China, Russia, Korea, HongKong, Taiwan, Thailand. And they see growing interest in Brasil.

Cheers, Jan

 

[DrTebi]

But one important area lacks in Tim Parkins test: slide projection.
With slide projection the best quality at huge enlargemt sizes is achieved. Best resolution, best sharpness, best brillance.

Digital projection can't compete at all with slide projection, because of the extremely low resolution of the beamers (only 2 MP in horizontal direction, even 40% less in vertical direction) and the bad color rendition.
In digital projection you have the situation that you pay 7000€ for a 24 MP D3x, and then you have to pay another 7000€ for a beamer to get the 24 MP of the camera smashed down to 2 MP of the beamer.
What a ridicolous waste of money.
We've did the direct comparison of 35mm slide projection and the most expensive 2 MP 7000€ beamers.
The slide is a league of it's own, no chance at all for the beamer.

That's what I've been saying all along. Well, I should say that's what I have discovered in the recent two years when I started shooting film again... Slide projection is fantastic. No pixel counting. Sit as close as a meter to your screen, you still don't see no "pixels." Re-live your photography!

 

[scottsa]

So many threads, so little time.

I've been working with digital a long time, film a lot longer. I have great respect for the capabilities of both. So what I say next should _not_ be taken as an outright support of either.

Just from the surface, because I'm beyond out of time to properly comment, so I'll just say this:

ITS ALL ABOUT THE SENSOR SIZE VS RESOLVING POWER

Pixels and grain have a lot in common, crap glass, etc. and you know the results.

Give the D3 a similar sized sensor to what it is being compared against, and the results will be _much_ different.

In other words, the results of the LF VS small-format is an age-old discussion from long before any digital approached the market.

Tech-Pan, etc. anyone?

Sorry, gotta leave it at that.

 

[Spanik]

But one important area lacks in Tim Parkins test: slide projection.
With slide projection the best quality at huge enlargemt sizes is achieved. Best resolution, best sharpness, best brillance.

This is the reason why I still use slides. Projecting a slide is wonderful.

 

[Skiff]

Yes, that is indeed one of the best comparison tests made.
Very good work.

In the past the internet and the photomagazines were flooded with these nonsense tests "scanned at 4000 ppi with amateur scanner" vs. digital sensors.
And then they said it was a "film vs. digital" test. BS. It was not.
It was a digital (scanner) vs. digital test.
And 4000 ppi scanners can not exploit the full potential of film, not at all. The quality loss with 4000 ppi scanners is huge.

Zeiss did several real film tests and published them in their camera lens news. Their results:
Spur Orthopan UR: 400 linepairs per millimeter (lp/mm)
Agfaortho 25: 250 lp/mm
Agfa APX 25: 200 lp/mm
Kodak TMX: 180 lp/mm
Fuji Acros 100: 160 Lp/mm
Fuji Velvia: 170 lp/mm

With a 24 MP FF sensor the maximal resolution is 85 lp/mm. That is the Nyquist frequency, the absolut physical limit which cannot be surpassed.
In reality most 24 MP sensors have a real resolution of 65 - 75 lp/mm because of the AA filter which is reducing resolution.

Some photographers and me have tested the D3x in comparisons to different films:
We've got 70 - 75 lp/mm with the D3x.
We've got 120 - 140 lp/mm with ISO 100 slide films.
We've reached 115 - 150 lp/mm with ISO 100 T-grain BW films.
And we've got 160 - 250 lp/mm with high resolution BW films.

Our test results have been a bit lower compared to Zeiss because we've used a lower object contrast.

There have been other tests published by Antora; Seemann, Serger; Ventzke with similar or even better results for film.
And all these tests included scanning, optical printing and slide projection:
With optical printing and slide projection significantly better detail rendition is achieved compared to 8000 ppi drum scans and especially 4000 ppi amateur scanners.
Drum scanners are limited at 110 lp/mm with medium object contrast, 4000 ppi at 70 lp/mm.

As we have done quite similar tests in our photographers group, I can completely agree with these test results of Tim Parkin, Zeiss, you (HHPhoto), because our results are very similar.
Another excellent scientific test summary (four tests by different test labs):
http://www.aphog.de/index.php?option=com_content&task=view&id=401&Itemid=1

But one important area lacks in Tim Parkins test: slide projection.
With slide projection the best quality at huge enlargemt sizes is achieved. Best resolution, best sharpness, best brillance.

Digital projection can't compete at all with slide projection, because of the extremely low resolution of the beamers (only 2 MP in horizontal direction, even 40% less in vertical direction) and the bad color rendition.
In digital projection you have the situation that you pay 7000€ for a 24 MP D3x, and then you have to pay another 7000€ for a beamer to get the 24 MP of the camera smashed down to 2 MP of the beamer.
What a ridicolous waste of money.
We've did the direct comparison of 35mm slide projection and the most expensive 2 MP 7000€ beamers.
The slide is a league of it's own, no chance at all for the beamer.

Cheers, Jan

Exactly!
Slide projection is indeed a league of it's own, and significantly surpassing even the most expensive current beamers (we've also done side by side tests).
In projection digital is decades behind with its extremely low resolution, bad color reproduction and insane prices.

With slide projection we get the best quality for big enlargements, with unsurpassed detail rendition and colour brillance. And that at extremely low costs:
One of these 1m x 1,5m or even bigger pictures costs you less than a buck!
A high quality print of that sizes costs you much more than 100 bucks!

 

[Ranchu]

bump. :dance:

 

[ptpdprinter]

Can someone do the math and tell me with 300dpi inkjet printing (80 micron), what is the maximum resolution that can be taken advantage of? I suspect that drum scanning 8x10 film and printing 300 dpi you are not able to take advantage of all (most?) of its resolution capture advantage over other formats.