film vs. digital comparison method

[Michael Heinrich]

I am working on a study about architectural photography at a German University and in a part I have to compare different analog and digital systems.
Its nearly impossible, if you have a look around the web you can find a lot of answers what digital equipment you can compare with a 4/5 inch dia.
From 8Mp to 250Mp !! The Problem of nearly all comparison methods is that they try to compare a "hammer with a srewdriver".
So I tried to find a way, and I would like to know what the experts on this forum think about it.

The Factor Q = 1 means that a photographic system has a sufficient image Quality for a print in the size DIN A 4 (European Standard 1/16 m2).
I compare prints, because in my case its the result of each processs.
But you can "translate" it.
For a print in a magazin in the size of 8x10 inch you also need a factor Q=1, in a good printed book factor 2 and so on.
On a 19 inch monitor you need Q=0,8 and so on....
An Example: A slr camera (XD-7) with a normal lens(Rokkor 1.7/50), a BW Film 400Asa( Tri X9, developped in fine grain developper(Emofin), enlarged on normal paper(Ilford Multigrade FB) gives you a good picture on DIN A 3 (16/12 inch), at the double size (DIN A 2)most of the viewers saw problems. So this only system has the factor Q= 2

The first results(only examples)
All the combinations are with a special lens/film, etc, to make it easier only some examples:

Film 135 from a point and shoot system/ plastic lens factor 0,3
to film 135 to High end factor 3,5

MF Systems Film:
simple MF System factor 2
LF with RF factor 4
hasselblad SWC/M factor 5
Plaubel proshift factor 6

LF systems:
4/5 old symmar factor 4
4/5 super symmar Velvia factor 9
138 apo sironar factor 15
810 Grandagon 22

Digital
casio 2MP factor 0,3
lumix 6MP factor 0,9
Canon 20D 8 MP factor 1,2
Canon 5D 12MP factor 2,4
Canon 5D 12MP 3 p. stitched factor 5
Eyelike 17MP Back on Alpa factor 4
22Mp Back factor 6
22Mp 2 pict. stitched factor 11
39Mp used as point and shoot factor 3
39Mp Back factor 11
39 MP Back Multishot factor 14

Different raw conversion and tuning are producing differnet factors.
Different persons have differnent opinions on quality, but I think the factor is the best way to compare.
So each person can try to improve his own way.
My favorite is a 13/18 technika with a super symmar 110, Kodak Film and a Ciba Print. I mada a print on 50x70 inch (factor 32) and it was great.
Annother picture made with the same camera, but a WA Lens, but problems of a long exposure, a Drumscan and a print on paper wasn`t`more than a 15...

I hope for responses

Michael

[ErikKaffehr]

Hi!

I think that the idea is good, but I would probably use A3 or even A2 as reference. Also sharpening is an important aspect of digital workflow, so this would need more consideration.

I´m somewhat surprised that you find that Tri-X and Emofin is Q = 2 and Canon 20D is 1.2, but I have no experience with Emofin. I would expect Canon 20D be in the same league as 100 ISO slide film like Provia.

Best regards
Erik

I am working on a study about architectural photography at a German University and in a part I have to compare different analog and digital systems.
Its nearly impossible, if you have a look around the web you can find a lot of answers what digital equipment you can compare with a 4/5 inch dia.
From 8Mp to 250Mp !! The Problem of nearly all comparison methods is that they try to compare a "hammer with a srewdriver".
So I tried to find a way, and I would like to know what the experts on this forum think about it.

The Factor Q = 1 means that a photographic system has a sufficient image Quality for a print in the size DIN A 4 (European Standard 1/16 m2).
I compare prints, because in my case its the result of each processs.
But you can "translate" it.
For a print in a magazin in the size of 8x10 inch you also need a factor Q=1, in a good printed book factor 2 and so on.
On a 19 inch monitor you need Q=0,8 and so on....
An Example: A slr camera (XD-7) with a normal lens(Rokkor 1.7/50), a BW Film 400Asa( Tri X9, developped in fine grain developper(Emofin), enlarged on normal paper(Ilford Multigrade FB) gives you a good picture on DIN A 3 (16/12 inch), at the double size (DIN A 2)most of the viewers saw problems. So this only system has the factor Q= 2

The first results(only examples)
All the combinations are with a special lens/film, etc, to make it easier only some examples:

Film 135 from a point and shoot system/ plastic lens factor 0,3
to film 135 to High end factor 3,5

MF Systems Film:
simple MF System factor 2
LF with RF factor 4
hasselblad SWC/M factor 5
Plaubel proshift factor 6

LF systems:
4/5 old symmar factor 4
4/5 super symmar Velvia factor 9
138 apo sironar factor 15
810 Grandagon 22

Digital
casio 2MP factor 0,3
lumix 6MP factor 0,9
Canon 20D 8 MP factor 1,2
Canon 5D 12MP factor 2,4
Canon 5D 12MP 3 p. stitched factor 5
Eyelike 17MP Back on Alpa factor 4
22Mp Back factor 6
22Mp 2 pict. stitched factor 11
39Mp used as point and shoot factor 3
39Mp Back factor 11
39 MP Back Multishot factor 14

Different raw conversion and tuning are producing differnet factors.
Different persons have differnent opinions on quality, but I think the factor is the best way to compare.
So each person can try to improve his own way.
My favorite is a 13/18 technika with a super symmar 110, Kodak Film and a Ciba Print. I mada a print on 50x70 inch (factor 32) and it was great.
Annother picture made with the same camera, but a WA Lens, but problems of a long exposure, a Drumscan and a print on paper wasn`t`more than a 15...

I hope for responses

Michael
[a href=\"index.php?act=findpost&pid=99609\"][{POST_SNAPBACK}][/a]

[free1000]

I am working on a study about architectural photography at a German University and in a part I have to compare different analog and digital systems.
Michael

Michael, this is an interesting study. One thing I like about it is the way that it shows clearly that there are a huge number of variables involved.

Clearly from your test I can see that the combination of the quality of the optic and size of the sensor area are critical and depending on these there can be considerably overlap.  Your favorite example using the Super-Symmar 110 uses what I understand is one of the best LF lenses available, so I guess this is not too surprising.

The main problem with the study is its subjectivity and its unrepeatability. I think given the same brief and a different approach with raw processing others might get similar results.

I have found that with an Aptus 75 (33 megapixel) I have been getting better results than with my 5x4 film. But its also made it clear to me that some of the LF lenses I was using are very poor.  

From my own tests an image stitched from 2 frames on this camera (giving around 58 megapixels) is adequate for art gallery prints up to 2 meters wide.  

I've concluded that a 58Mp digital camera will finally obsolete pretty much any 5x4 film results that I could economically achieve. Note that I am not saying that someone else couldn't do better, only that I can't given my financial and time resources.

[Ray]

I get the impression there's an 'apples and oranges' comparison taking place here. How do we compare a 'grain-free' result from a digital camera, with a 'grain-plagued' result from film?

My impression is, with the best technique and the finest grain films (B&W), it's possible to count more line pairs per mm comparing the same size piece of film with the same size digital sensor.

Unfortunately, the over-all image quality of the superior resolving film might be (certainly appears to be) worse.

This issue was dealt with in Michael's first (perhaps over-enthusiastic review) of Canon's first DSLR, the D30.

If it's pure resolution you are after, identifiable text or number of line counts, you could probably achieve higher results with a perfectly flat, fine grained B&W film.

However, grain whatever its artistic potential, is not what we see in reality. A digital camera tends to impart this freedom from grain which more than compensates for any 'ultimate' lack of resolution in huge enlargements.

[Michael Heinrich]

Yes I try to compare apples and oranges...
You can`t really compare digital to analog systems... you only can compare a single picture to an other single picture. So I think you have to work empiric. Thats what I try to do...
Hi Free 1000 It is not the same thing wether you take a part of a 2m picture and look it at A4 print or you are looking the hole print.
I am only comparing small prints, because you don`t need the same "quality/resolution" for a wall print than for a small print.

Michael