Ad
Ad
Ad
Pages: [1] 2 3 »   Bottom of Page
Print
Author Topic: MF versus DSLR. Ending the argument.  (Read 11320 times)
JdeV
Full Member
***
Offline Offline

Posts: 120


WWW
« on: November 25, 2009, 07:48:25 AM »
ReplyReply

CBarret's test with his Arca and D3/P65+ showed a practical method of comparing sensors. I would like to see the same test done with a D3x but I suspect, based on his test done with the D3, my own experience and the DXO mark ratings, that at low ISO the best 35mm sensors are comparable with the best medium format sensors in terms of dynamic range, colour and sharpness/mm. (Though obviously there remain significant subtle differences, just as there are between different 35mm sensors and different medium format sensors).

Another extremely relevant test to do would be to hold the sensor constant and mount different lenses. I suggest that a practical test protocol would be to use a view camera with a P65+ and mount a couple of the best digital view camera lenses followed by a couple of the best 35mm lenses. This would require, say, an F-mount to Copal 3 adaptor. Anyone have any ideas? At a pinch one could do the relevant tests with flash and just get a female F-mount put on a lens panel. Of course most 35mm lenses would not cover the sensor area but one could determine comparative resolution per mm and also extrapolate resolution over a 35mm frame and compare it with the resolution of a full P65+ frame with lenses yielding the same field of view. For a more thorough comparison it would be great to also have Mamiya and Hasselblad to Copal 3 adaptors.

If CBarret's test was done more extensively along with this lens test we would have a much clearer picture of the relevant image quality issues across the different formats. Many other questions would also be illuminated. We would have a much better idea of how lens-limited sensors are, and consequently the useful theoretical ceiling in resolution for a particular sensor size, (given lens limits is there a genuine resolution gain going from a P65+ to whatever the resolution the next generation of sensors will be if the physical size of the sensor is not increased?). How much is lens resolution traded for increased coverage and as a consequence where is the sweet spot for resolution with sensor size/lens engineering limitations? This latter question is also very relevant for stitching. Which will yield more resolution, an x-part stitch with a lens with greater coverage or a single frame or stitch with lower frame count and sharper lens with less coverage? We would also get a sense of the outer parameters that are going to be possible: We aren't going to get sharper medium format or view camera lenses than the best current 35mm lenses.

None of this tells us anything about noise at higher ISOs or practical handling issues with different cameras but those matters are well known and understood already. Meandering discussions like the '3D look' one and arguments about MF versus DSLR remain very unsatisfactory with respect to basic image quality issues until good comparative sensor-to-sensor and lens-to-lens evaluations are made. CBarret's test provides a way of holding all variables constant except the sensor, I'm suggesting the same kind of thing should be done with lenses.

Logged
evgeny
Sr. Member
****
Offline Offline

Posts: 481



« Reply #1 on: November 25, 2009, 08:17:48 AM »
ReplyReply

If I understand correctly, this will compare MF lenses vs 35mm lenses. Who will use in practice 35mm lenses on a MF body?

I interested in a comparison of IMAGES shoot with one complete MF system vs another MF or 35mm system. I think the gear should be different, the subject and light should be the same.

I also think the images should NOT be processed. Processing images make a big difference, no matter how much processing were added. So, the tester(s) should agree to not change the original and show us the result stright out of the MFDB.
« Last Edit: November 25, 2009, 08:28:37 AM by evgeny » Logged
JdeV
Full Member
***
Offline Offline

Posts: 120


WWW
« Reply #2 on: November 25, 2009, 08:56:32 AM »
ReplyReply

Quote from: evgeny
If I understand correctly, this will compare MF lenses vs 35mm lenses. Who will use in practice 35mm lenses on a MF body?

I interested in a comparison of IMAGES shoot with one complete MF system vs another MF or 35mm system. I think the gear should be different, the subject and light should be the same.

I also think the images should NOT be processed. Processing images make a big difference, no matter how much processing were added. So, the tester(s) should agree to not change the original and show us the result stright out of the MFDB.

It isn't hard to take a DSLR and MF camera or view camera with MF back, shoot the same thing and compare images. What is difficult is determining the source of any differences observed. By isolating variables we can be a bit more scientific about it. This has practical consequences in terms of current usage and evaluation of realistic future possibilities. By testing lenses on their own whilst holding the sensor constant we can see lens limits and where the resolution/coverage trade-off falls. It could help understand how much room (if any) there is for increased resolution in the different formats given such limits and how we can use the lenses we have now. 35mm lenses are worth testing to see whether they are sharper than MF or view camera lenses and if so by how much.

To give one simple example: suppose we were to find that the best 35mm lenses could resolve 20% more than the best MF lenses which, in turn, could resolve 20% more than digital view camera lenses with 90mm image circles. We could then say that the lens limit on resolution for 35mm sensors was 20% higher than MF which, in turn, was 20% more than digital view cameras. Of course there are many complications to this picture, like resolution fall-off across a frame, aberrations etc. but I am sure there would still be rough generalisations that could be made.
« Last Edit: November 25, 2009, 12:12:48 PM by JdeV » Logged
Terence h
Full Member
***
Offline Offline

Posts: 105


WWW
« Reply #3 on: November 25, 2009, 09:11:21 AM »
ReplyReply

I have a Aptus 75 and a 5D MK11 and their is no comparison the Aptus files are way better.
It is something everybody has to find out themselves by actually using the cameras , hearing
what others have to say is useful but nobody should be buying until you have tried them
yourself.

Regards
Terence
Logged

Terence Hogben. Durban. South Africa. http://www.terencehogben.co.za
stevesanacore
Full Member
***
Offline Offline

Posts: 214


« Reply #4 on: November 25, 2009, 09:27:15 AM »
ReplyReply

Quote from: JdeV
It isn't hard to take a DSLR and MF camera or view camera with MF back, shoot the same thing and compare images. What is difficult is determining the source of any differences observed. By isolating variables we can be a bit more scientific about it. This has practical consequences in terms of current usage and evaluation of realistic future possibilities. By testing lenses on their own whilst holding the sensor constant we can see lens limits and where the resolution/coverage trade-off falls. It could help understand how much room (if any) there is for increased resolution in the different formats given such limits and how we can use the lenses we have now. 35mm lenses are worth testing to see whether they are sharper than MF or view camera lenses and if so by how much.

Next month I hope to take my 1DsMk3 and 5Dmk2 and compare with the new Leica S2 and  a Phase One P45 or 65.  I'm going to use one of my Leica lenses on my Canons and shoot a few sample landscapes.  That should really put and end to the debate in my mind. I will post the results. I have recently looked at a sample printed landscape shot from a H3D-50 and was blown away from the detail. I have no doubt the MF will trounce the Canon's as far as detail in large prints. I am very curious how the S2 will perform against the Phase P65. I think the bottom line will be the quality of the optics in the MF category.

I was recently at a demo for the new Phase camera and had the opportunity to talk directly with a representative about the optics. He said with the kind of work I do, I would not be happy with the Phase optics and would need to use the back on a tech camera to achieve the highest quality results. I was a bit shocked at his honesty. I would love to hear what others have to say about it.

Logged

We don't know what we don't know.
JdeV
Full Member
***
Offline Offline

Posts: 120


WWW
« Reply #5 on: November 25, 2009, 09:35:26 AM »
ReplyReply

Quote from: Terence h
I have a Aptus 75 and a 5D MK11 and their is no comparison the Aptus files are way better.
It is something everybody has to find out themselves by actually using the cameras , hearing
what others have to say is useful but nobody should be buying until you have tried them
yourself.

Regards
Terence

I've shot about 15,000 frames on a D3x and about the equivalent number with either a P65+, P45+ or an H3D39 on H-series backs and view cameras. I have a good general idea of the differences. What I couldn't honestly say is what the lens-limited maximum sensor resolutions are for the different formats nor how much you win by stitching a Rodenstock HR-W lens versus a single frame with an HR lens versus a Hasselblad H lens nor a host of other questions that are of practical working relevance and also significant for purchase decisions. Sometimes careful experiment is much more time efficient than rough try-outs.
« Last Edit: November 25, 2009, 09:37:56 AM by JdeV » Logged
Dick Roadnight
Sr. Member
****
Offline Offline

Posts: 1730


« Reply #6 on: November 25, 2009, 09:48:04 AM »
ReplyReply

Quote from: evgeny
I interested in a comparison of IMAGES shoot with one complete MF system vs another MF or 35mm system. I think the gear should be different, the subject and light should be the same.

I also think the images should NOT be processed. Processing images make a big difference, no matter how much processing were added. So, the tester(s) should agree to not change the original and show us the result stright out of the MFDB.
The complete system includes the lens correction software.
Logged

Hasselblad H4, Sinar P3 monorail view camera, Schneider Apo-digitar lenses
JoeKitchen
Sr. Member
****
Offline Offline

Posts: 640



« Reply #7 on: November 25, 2009, 11:28:06 AM »
ReplyReply

I doubt this is going to be the end all say all end of this argument.  And any way, its all about personal preference what type of camera you emotionally prefer to work with.  Yes I said emotionally, we are professional artist and need to be emotional about our work.
Logged

Joe Kitchen
www.josephmkitchen.com

"Photography is one percent inspiration and ninety-nine percent moving furniture."  Arnold Newman
"Try not to be just better than your rivals and contemporaries, try to be better than yourself."  William Faulkner
Juanito
Full Member
***
Offline Offline

Posts: 170


WWW
« Reply #8 on: November 25, 2009, 11:35:39 AM »
ReplyReply

Nadav Kander is fond of saying that when he's shooting, he always thinks about the end goal: ink on paper.

I've done my own informal tests between MFDB and small format. MFDB is clearly a better file. The problem is that by the time you get to ink on paper, the difference is lost. There's too much that goes on from capture to final output to show the difference. Even if you forego post-production to the file, you still have interpolation by printer drivers, the spread of ink on paper and even the tooth of the paper to contend with.

So, we can do a test here by looking at files on our computers, but that's like judging a game by looking at the scores at halftime. It's nice to know and it's interesting, but it tells us little about the final outcome. The true measure is prints sitting side by side. As professionals, our work is ultimately judged in print form - whether it's in our portfolios or in the final printed pages produced by the client. Unfortunately, that's kinda tough to compare online.

Equally unfortunate, the argument will never end.  

John
Logged

KLaban
Sr. Member
****
Offline Offline

Posts: 1611



WWW
« Reply #9 on: November 25, 2009, 12:07:38 PM »
ReplyReply

MF versus DSLR. Ending the argument.

That has to be the least likely line I've ever seen on this forum.
Logged

uaiomex
Sr. Member
****
Offline Offline

Posts: 933


WWW
« Reply #10 on: November 25, 2009, 12:13:39 PM »
ReplyReply

Very nice. This post is lucid and straightforward. I don't own a MF digital camera and I've only seen just 2 or 3 prints made from MF digital in my life. But from all posts I've read here and there for years, I've conclude the following:

1.  DMF files are better than D35
2.  Inkjet printing makes DMF lose most of the advantage it has over D35
3.  DMF files show its advantage only when printed really big.
4.  DMF files can be tweaked further

5. Is there an advantage when printed double page in standard glossy magazine?
    If the answer is: "Some, slightly, Barely or None", then I see no point in spending a patrimony over a camera. Carefully exposed pictures need less tweaking anyway.

This dispute will probably never get set. The only thing we can do is to listen to those photogs that use both systems extensively provided their are honest and unbiased.
Thanks all for some terrific posts
Eduardo


Quote from: Juanito
Nadav Kander is fond of saying that when he's shooting, he always thinks about the end goal: ink on paper.

I've done my own informal tests between MFDB and small format. MFDB is clearly a better file. The problem is that by the time you get to ink on paper, the difference is lost. There's too much that goes on from capture to final output to show the difference. Even if you forego post-production to the file, you still have interpolation by printer drivers, the spread of ink on paper and even the tooth of the paper to contend with.

So, we can do a test here by looking at files on our computers, but that's like judging a game by looking at the scores at halftime. It's nice to know and it's interesting, but it tells us little about the final outcome. The true measure is prints sitting side by side. As professionals, our work is ultimately judged in print form - whether it's in our portfolios or in the final printed pages produced by the client. Unfortunately, that's kinda tough to compare online.

Equally unfortunate, the argument will never end.  

John
Logged
Terence h
Full Member
***
Offline Offline

Posts: 105


WWW
« Reply #11 on: November 25, 2009, 02:15:08 PM »
ReplyReply

Quite often i hear the people say that you cannot see the difference on a DPS , but now if your work is printed
really large and at high resolution which is happening a lot now you really can see the difference , especially
as the 2x3 format of the 35mm based digital cameras has to be cropped to reach the more common poster
formats.
Please note i am not wanting to argue the point here just state what i can see and the fact that my customers
always say "please quote using the large camera " when the posters are going A3 and upwards  , recently did
a job where the images where going up in a retail warehouse and some images were 10 meters tall and needed
to be printed at a fairly high resolution , no 35mm camera would have cut it.

When i bought the 5D MK11 i thought to myself that maybe i could get rid of the Aptus because the file size is not
that far off and nobody would notice , well i can see a big difference and i would be short changing my customers
to give them anything but the best i can do.

And do not say the way D3X is way better :-) because this would soon digress into a DP Review type slanging match.

Back to lurking mode.

Regards
Terence
Logged

Terence Hogben. Durban. South Africa. http://www.terencehogben.co.za
rainer_v
Sr. Member
****
Offline Offline

Posts: 1120


WWW
« Reply #12 on: November 25, 2009, 02:35:26 PM »
ReplyReply

how often is apearing this discussion is really funny.
do anybdy seriously think or remember that in film days it appeard also similar often the question
if 35mm would be equal than 4x5" or the bigger mf formats as 6x7 ?  ( except from leica fans )  
and why didnt it be questioned in the same way ?

i think because the differences have been quite obvious and VERY big in film, in terms of resolution but also in terms of "look".
in comparation  this huge differences have been marginalized ( imo ) meanwhile the price differences between the systems have multiplied itself.


but this eternal mf-35 discussions wont stop, even they might increase cause the gap will be already smaller and smaller.
« Last Edit: November 25, 2009, 02:36:19 PM by rainer_v » Logged

rainer viertlböck
architecture photographer
munich / germany

www.tangential.de
Ed Jack
Full Member
***
Offline Offline

Posts: 225


« Reply #13 on: November 25, 2009, 02:58:15 PM »
ReplyReply


 Maybe I use MF system because I like using them, you know, the large view finders and leaf shutters etc. Maybe we focus too much on IQ these days, especially as any caerma or back released to pro's these days is usually pretty good (IQ) by default (high iso being a area where we have seen much recent improvement though).
Logged
cyberean
Full Member
***
Offline Offline

Posts: 161


WWW
« Reply #14 on: November 25, 2009, 03:16:09 PM »
ReplyReply



just use whatever floats your boat ...
why the incessant drive/desire to convince the next guy of your choice

... peace

Logged

check out the size of my sensor ...
bcooter
Guest
« Reply #15 on: November 25, 2009, 03:42:21 PM »
ReplyReply

Quote from: rainer_v
do anybdy seriously think or remember that in film days it appeard also similar often the question
if 35mm would be equal than 4x5" or the bigger mf formats as 6x7 ?..................... the gap will be already smaller and smaller.

This is just a money conversation, always has been.

In the film days a larger format system might costs five times as much but didn't costs $44,000 Vs. $3,500.

That's where this comes from, that and the fact the $44,000 camera doesn't have as many features as the $3,500 one.

If the top line medium format camera only costs $7,000 these conversations would be 1/100th of the volume we see now.

In the film days the difference between the film size of a hand camera vs a large format camera was huge, same with the image quality,  but now 645 vs. 35mm isn't that much of a difference.

That's why you hear this 3d look talk, or sharpness talk because in reality 22mpx is still 22mpx and even then it's a larger file than required for most printed pages.

Also in the film days every view camera had swings, tilts, movements that we're almost impossible to do with a smaller camera. Now with dslr tilt shift lenses that isn't true.

Then again in the film days, regardless of price you bought a camera for a decade not 18 months.

But basically, It's just a money thing,

IMO

BC
Logged
rainer_v
Sr. Member
****
Offline Offline

Posts: 1120


WWW
« Reply #16 on: November 25, 2009, 04:06:43 PM »
ReplyReply

Quote from: bcooter
This is just a money conversation, always has been.

In the film days a larger format system might costs five times as much but didn't costs $44,000 Vs. $3,500.

That's where this comes from, that and the fact the $44,000 camera doesn't have as many features as the $3,500 one.

If the top line medium format camera only costs $7,000 these conversations would be 1/100th of the volume we see now.

In the film days the difference between the film size of a hand camera vs a large format camera was huge, same with the image quality,  but now 645 vs. 35mm isn't that much of a difference.

That's why you hear this 3d look talk, or sharpness talk because in reality 22mpx is still 22mpx and even then it's a larger file than required for most printed pages.

Also in the film days every view camera had swings, tilts, movements that we're almost impossible to do with a smaller camera. Now with dslr tilt shift lenses that isn't true.

Then again in the film days, regardless of price you bought a camera for a decade not 18 months.

But basically, It's just a money thing,

IMO

BC

agreed.

i just bought a setup of these new shift lenses for the canon and shot the first job with it, to try it out. its soooo convenient for travelling, and its fantastic to have a real live view. workflow is very good on site and postpro is fast too.
just printed out a shot done with the 17mm tse at 60/90 cm and there is no millimeter of detail to desire.
in the few cases i will go bigger the prints are big enough that no one puts its nose on them, and even if, upinterpolated right and added some grain ( same is necessary with any mf file ) there will be  absolutely nothing to desire, at least this is what i think at this moment, i will see it soon.
i will see how my personal workflow will go on in future, i have the luxury to own two complete setups, so its up to me which i`ll find more convenient
( and better!! ). at this moment i am just more than surprised how these two new shift lenses from canon have changed the game in my working field.
« Last Edit: November 25, 2009, 04:07:56 PM by rainer_v » Logged

rainer viertlböck
architecture photographer
munich / germany

www.tangential.de
AlexM
Full Member
***
Offline Offline

Posts: 136


WWW
« Reply #17 on: November 25, 2009, 04:12:06 PM »
ReplyReply

I am afraid there is no end to this argument.
At the end of the day, everyone makes his/her own decision on the gear he/she wants and/or can afford to use.
Logged

Harold Clark
Full Member
***
Offline Offline

Posts: 184


« Reply #18 on: November 25, 2009, 05:19:12 PM »
ReplyReply

Quote from: Terence h
Quite often i hear the people say that you cannot see the difference on a DPS , but now if your work is printed
really large and at high resolution which is happening a lot now you really can see the difference , especially
as the 2x3 format of the 35mm based digital cameras has to be cropped to reach the more common poster
formats.
Please note i am not wanting to argue the point here just state what i can see and the fact that my customers
always say "please quote using the large camera " when the posters are going A3 and upwards  , recently did
a job where the images where going up in a retail warehouse and some images were 10 meters tall and needed
to be printed at a fairly high resolution , no 35mm camera would have cut it.

When i bought the 5D MK11 i thought to myself that maybe i could get rid of the Aptus because the file size is not
that far off and nobody would notice , well i can see a big difference and i would be short changing my customers
to give them anything but the best i can do.

And do not say the way D3X is way better :-) because this would soon digress into a DP Review type slanging match.

Back to lurking mode.

Regards
Terence

Terence,

How do you differentiate your pricing between MF & SLR, ie. what percentage extra do you factor in when quoting. This is an important consideration, as we have to quickly recover our "investment" in these disposable cameras.

Harold
Logged
JdeV
Full Member
***
Offline Offline

Posts: 120


WWW
« Reply #19 on: November 25, 2009, 05:37:03 PM »
ReplyReply

Quote from: bcooter
This is just a money conversation, always has been.

In the film days a larger format system might costs five times as much but didn't costs $44,000 Vs. $3,500.

That's where this comes from, that and the fact the $44,000 camera doesn't have as many features as the $3,500 one.

If the top line medium format camera only costs $7,000 these conversations would be 1/100th of the volume we see now.

In the film days the difference between the film size of a hand camera vs a large format camera was huge, same with the image quality,  but now 645 vs. 35mm isn't that much of a difference.

That's why you hear this 3d look talk, or sharpness talk because in reality 22mpx is still 22mpx and even then it's a larger file than required for most printed pages.

Also in the film days every view camera had swings, tilts, movements that we're almost impossible to do with a smaller camera. Now with dslr tilt shift lenses that isn't true.

Then again in the film days, regardless of price you bought a camera for a decade not 18 months.

But basically, It's just a money thing,

IMO

BC

It's a money thing combined with an ignorance thing. It's actually very hard to know in detail what is going on with these systems without laying out a lot of cash and shooting a lot. In film days most working pros and even people behind counters in shops new what shooting with a different equipment meant in terms of quality and look. I think there were two reasons for that: 1) Film was constant across formats. 2) Equipment and materials were around a long time and only changed incrementally.

Fortunately we now have forums though...right?
Logged
Pages: [1] 2 3 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad