dwterry
Well-known member
Actually, as it print ... it was AWESOME! Sorry for posting such a large image, I didn't think about how it would appear on your monitor.
Reading mode:
Install the app
How to install the app on iOS
Follow along with the video below to see how to install our site as a web app on your home screen.
Note: This feature may not be available in some browsers.
-
Welcome to the new forums! Please read this first. For known issues we are working to resolve, click here.
You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.
You should upgrade or use an alternative browser.
Is 10.1 MegaPixels really enough?
- Thread starter JoostA
- Start date
SFJP
Senior Member
With my old Fuji 6900, which is a 3 mp camera interpolated to 6 mp with the effective resolution of a 4.5 mp camera, I systematically got good A3 (11x17 inches) from good shots.
Of course if you want to get huge prints with a lot of sharp details, like a landscape or a fashion poster, more mp's are needed. But DR, good colors, absence of noise, natural sharpness and accutance of the combination sensor/lens is the most important factor, follwed also by the camera ergonomy and propensity for good exposure and spot-on AF or MF. Then, only after you get that, you can look at the mp's count, and of course the more is better, if this does not imply more noise at the ISO sensitivity you use the most.
For instance, I am currently looking to upgrade my 1DS mark I to a 1 DS Mark II. Not for the greater mp's count but because I believe the Mark II is easier to get correct exposure with, has less noise and is faster. My only fear is that the Mark II could be less sharp and with less accutance than the Mark I, while its resolution is of course higher.
--
SFJP
http://www.pbase.com/sfjp
Of course if you want to get huge prints with a lot of sharp details, like a landscape or a fashion poster, more mp's are needed. But DR, good colors, absence of noise, natural sharpness and accutance of the combination sensor/lens is the most important factor, follwed also by the camera ergonomy and propensity for good exposure and spot-on AF or MF. Then, only after you get that, you can look at the mp's count, and of course the more is better, if this does not imply more noise at the ISO sensitivity you use the most.
For instance, I am currently looking to upgrade my 1DS mark I to a 1 DS Mark II. Not for the greater mp's count but because I believe the Mark II is easier to get correct exposure with, has less noise and is faster. My only fear is that the Mark II could be less sharp and with less accutance than the Mark I, while its resolution is of course higher.
--
SFJP
http://www.pbase.com/sfjp
No, that isn't the "whole matter" - only part of it. Having a smaller pixel certainly lets you get more pixels on the subject, but they have to be resolved pixels to be meaningful. To determine what is resolved, you need to take into account a whole host of issues including lens MTF, camera stability and subject motion to name a few....And the whole matter is a function of sensels-density on a givenA cropped sensor doesn't give you more "reach" or "magnification."
"unit" of surface
If the whole story of extra reach was just pixel density then you would be comparing 122pixel/mm on the 5D to 135pixels/mm on the 1D3 - an extra reach of 11%. However, those smaller pixels on the 1D3 require 10% more MTF from the lens than the larger pixels in the 5D, which eliminates a large part of the "extra reach" benefit since few lenses deliver that. Other issues pretty much wipe it out entirely.
You are right.A cropped sensor doesn't give you more "reach" or "magnification."
It's the question how many pixels you get into the area you are interested in.
And what's the qyuality of each of the pixels.
In the 1.3 cropped area 1D3 has more pixels that those 12MP FF sensors.
So if you subject is in that area, you get more information of that.
This is especially true in telephotrography, where you quite often want to get as much information of a small far-away subject as possible. In these cases the cropped sensors - having higher pixel density - usually helps.
But of course 1Ds3 has even denser sensor, so you get more information with no "area limitations" with that camera.
Greg Hawkins
Well-known member
I think its also important to the quality of the pixels not just the number of pixels.
Not so fast there, Boyo! You forgot about the 3rd sentence in your posting: quality of pixels - and that includes the practical quality, ie. what pixels the glass can resolve.You are right.A cropped sensor doesn't give you more "reach" or "magnification."
It's the question how many pixels you get into the area you are
interested in.
And what's the qyuality of each of the pixels.
In the 1.3 cropped area 1D3 has more pixels that those 12MP FF sensors.
So if you subject is in that area, you get more information of that.
But those are the very conditions where much glass, even L glass, just isn't up to the job of resolving the small sensor pixels (even diffraction is significant), so the benefit of the small densely packed pixels is much reduced in practice. Its all wrapped up in your 3rd sentence - don't underestimste its significance.This is especially true in telephotrography, where you quite often
want to get as much information of a small far-away subject as
possible. In these cases the cropped sensors - having higher pixel
density - usually helps.
Not so fast there, Boyo! You forgot about the 3rd sentence in your posting: quality of pixels - and that includes the practical quality, ie. what pixels the glass can resolve.You are right.A cropped sensor doesn't give you more "reach" or "magnification."
It's the question how many pixels you get into the area you are
interested in.
And what's the qyuality of each of the pixels.
In the 1.3 cropped area 1D3 has more pixels that those 12MP FF sensors.
So if you subject is in that area, you get more information of that.
But those are the very conditions where much glass, even L glass, just isn't up to the job of resolving the small sensor pixels (even diffraction is significant), so the benefit of the small densely packed pixels is much reduced in practice. Its all wrapped up in your 3rd sentence - don't underestimste its significance.This is especially true in telephotrography, where you quite often
want to get as much information of a small far-away subject as
possible. In these cases the cropped sensors - having higher pixel
density - usually helps.
Unless we are talking of the lowest quality lenses, I have not seen yet this to be a seriously limiting factor. We talk now about pixel densities a lot less than in those crop1.6 cams, and there according to lens test I have seen (in magazines like ColorFoto) we get noticable improvement resolving power when pixels are added (and I'm not talking about high-ISO images as those may have their own problems, which with the cams in question are though low)Not so fast there, Boyo! You forgot about the 3rd sentence in yourYou are right.A cropped sensor doesn't give you more "reach" or "magnification."
It's the question how many pixels you get into the area you are
interested in.
And what's the qyuality of each of the pixels.
In the 1.3 cropped area 1D3 has more pixels that those 12MP FF sensors.
So if you subject is in that area, you get more information of that.
posting: quality of pixels - and that includes the practical quality,
ie. what pixels the glass can resolve.
Diffraction may become significant if you want very high DOF. But if I remember right the diffraction limit in crop1.6 cams is around f11 and in these crop1.3 cams even higher around f16. Then of course atmosphere diffraction start to reduce achievable sharpness when you shoot very far away.But those are the very conditions where much glass, even L glass,This is especially true in telephotrography, where you quite often
want to get as much information of a small far-away subject as
possible. In these cases the cropped sensors - having higher pixel
density - usually helps.
just isn't up to the job of resolving the small sensor pixels (even
diffraction is significant), so the benefit of the small densely
packed pixels is much reduced in practice. Its all wrapped up in
your 3rd sentence - don't underestimste its significance.
No, it is significant with high quality lenses too. See this thread:Unless we are talking of the lowest quality lenses, I have not seenNot so fast there, Boyo! You forgot about the 3rd sentence in yourYou are right.
It's the question how many pixels you get into the area you are
interested in.
And what's the qyuality of each of the pixels.
In the 1.3 cropped area 1D3 has more pixels that those 12MP FF sensors.
So if you subject is in that area, you get more information of that.
posting: quality of pixels - and that includes the practical quality,
ie. what pixels the glass can resolve.
yet this to be a seriously limiting factor.
http://forums.dpreview.com/forums/read.asp?forum=1032&message=25838251
Not at all. The MTF of a diffraction limited lens is almost a straight line from 0cy/mm to approximately 50% of the cut-off resolution, at which point it tapers off. So even at apertures well below those at which diffraction limits the resolution, the MTF of even a perfect lens is proportionally lower for the smaller pixels than it is for the larger ones. For example, at f/8 and below, the MTF of a perfect lens (ie. the diffraction component alone) is 66% at the Nyquist limit of the5D, but only 60% at the Nyquist limit of the 1D-III. ie. even a perfect lens has 10% less resolution in each axis on the 1D-III than on the 5D, and the difference is more significant with practical, even high quality, lenses or higher f/#s. With low contrasts at long ranges, that is a very significant loss which makes the 10% gain in pixel density pale into insignificance.Diffraction may become significant if you want very high DOF. But ifBut those are the very conditions where much glass, even L glass,This is especially true in telephotrography, where you quite often
want to get as much information of a small far-away subject as
possible. In these cases the cropped sensors - having higher pixel
density - usually helps.
just isn't up to the job of resolving the small sensor pixels (even
diffraction is significant), so the benefit of the small densely
packed pixels is much reduced in practice. Its all wrapped up in
your 3rd sentence - don't underestimste its significance.
I remember right the diffraction limit in crop1.6 cams is around f11
and in these crop1.3 cams even higher around f16.
I think you mean "refraction" or "distortion". The atmosphere itself doesn't diffract.Then of course
atmosphere diffraction start to reduce achievable sharpness when you
shoot very far away.
Hi
Little bit OT
If You show Your pictures on the net will You benifit from more pixels?
I think so, if they are large enough or cropped.
Comparisson (not mine) between Canon 1Ds MarkIII (21MP) and Canon 1D MarkIII (10MP)
There are not 100% crops, just whole picture.
I see more detail in the 1Ds Mark III shoot
http://www.digital-imago.net/?p=42#more-42
Best regards from Sweden
Omar Brännström
--
http://sydnet.net/omar
Little bit OT
If You show Your pictures on the net will You benifit from more pixels?
I think so, if they are large enough or cropped.
Comparisson (not mine) between Canon 1Ds MarkIII (21MP) and Canon 1D MarkIII (10MP)
There are not 100% crops, just whole picture.
I see more detail in the 1Ds Mark III shoot
http://www.digital-imago.net/?p=42#more-42
Best regards from Sweden
Omar Brännström
--
http://sydnet.net/omar
Max Savin
Senior Member
i'm printing 20x30inches with an epson 7600. incredible detail that can be viewed from 6 inches. (leica M8). i get similar results with a 5d at 24x36 and see no reason you could not do the same with a 1dm3.
--
max
--
max
DavidSvensson
Senior Member
- Messages
- 1,374
- Reaction score
- 9
Sure it does scatter. Through several mechanisms. Starting with Rayleigh scattering.I think you mean "refraction" or "distortion". The atmosphere itself
doesn't diffract.
Then there is dust, fog, thermal gradients.
David
DavidSvensson
Senior Member
- Messages
- 1,374
- Reaction score
- 9
But then there is the joy of being able to crop more freely with a bigger MP sensor.i'm printing 20x30inches with an epson 7600. incredible detail that
can be viewed from 6 inches. (leica M8). i get similar results with
a 5d at 24x36 and see no reason you could not do the same with a 1dm3.
--
max
David
Max Savin
Senior Member
i don't like to waste any pixles and maintaining a perfect 2/3 ratio in a crop is a pain in the a$$.
--
max
--
max
Very good explanation - I do not have the knowledge to say tes or no.No, it is significant with high quality lenses too. See this thread:Unless we are talking of the lowest quality lenses, I have not seenNot so fast there, Boyo! You forgot about the 3rd sentence in yourYou are right.
It's the question how many pixels you get into the area you are
interested in.
And what's the qyuality of each of the pixels.
In the 1.3 cropped area 1D3 has more pixels that those 12MP FF sensors.
So if you subject is in that area, you get more information of that.
posting: quality of pixels - and that includes the practical quality,
ie. what pixels the glass can resolve.
yet this to be a seriously limiting factor.
http://forums.dpreview.com/forums/read.asp?forum=1032&message=25838251
Not at all. The MTF of a diffraction limited lens is almost aDiffraction may become significant if you want very high DOF. But ifBut those are the very conditions where much glass, even L glass,This is especially true in telephotrography, where you quite often
want to get as much information of a small far-away subject as
possible. In these cases the cropped sensors - having higher pixel
density - usually helps.
just isn't up to the job of resolving the small sensor pixels (even
diffraction is significant), so the benefit of the small densely
packed pixels is much reduced in practice. Its all wrapped up in
your 3rd sentence - don't underestimste its significance.
I remember right the diffraction limit in crop1.6 cams is around f11
and in these crop1.3 cams even higher around f16.
straight line from 0cy/mm to approximately 50% of the cut-off
resolution, at which point it tapers off. So even at apertures well
below those at which diffraction limits the resolution, the MTF of
even a perfect lens is proportionally lower for the smaller pixels
than it is for the larger ones. For example, at f/8 and below, the
MTF of a perfect lens (ie. the diffraction component alone) is 66% at
the Nyquist limit of the5D, but only 60% at the Nyquist limit of the
1D-III. ie. even a perfect lens has 10% less resolution in each axis
on the 1D-III than on the 5D, and the difference is more significant
with practical, even high quality, lenses or higher f/#s. With low
contrasts at long ranges, that is a very significant loss which makes
the 10% gain in pixel density pale into insignificance.
But still when looking lens tests in ColorFoto magazine and similar you notice that denser 10MP crop1.6 cameras resolve more than 8MP crop1.6 cams with the same lens, so the lens is not here limiting the resolving power. Actually the difference seems to be quite close 10:8 in most cases.
And the 10MP crop1.3 is in this respect an easier case.
So I do not know what the theory says, but practicval measurements show there is an advantage.
No I ment diffract - and perhas you should study the practical things more ;-)I think you mean "refraction" or "distortion". The atmosphere itselfThen of course
atmosphere diffraction start to reduce achievable sharpness when you
shoot very far away.
doesn't diffract.
Oh jes, its not the air itself, but the water in the air.
da_Feeniks
Leading Member
A 200mm lens on a 5D is a 200mm lens on a 40D. The optics have not changed.
You're talking about "pixel density," an entirely different matter unrelated to sensor size. The original 1D has precisely the same sensor size as they 1D/2 and 1D/3. The D30 has the same sensor size as the 40D. Using the "Focal Length Magnification" marketing smoke that so many apparently have been deluded by, they both "magnify" the same lens identically.
Sensor size has never changed the optics of a lens in any dSLR.
Brendan
=====
I am the last sane person on earth.
You're talking about "pixel density," an entirely different matter unrelated to sensor size. The original 1D has precisely the same sensor size as they 1D/2 and 1D/3. The D30 has the same sensor size as the 40D. Using the "Focal Length Magnification" marketing smoke that so many apparently have been deluded by, they both "magnify" the same lens identically.
Sensor size has never changed the optics of a lens in any dSLR.
Brendan
=====
I am the last sane person on earth.
Greg Henry
Senior Member
Yeah, I was going to say - no offense to the photographer, but at least on my calibrated monitor at work this image isn't that great. Lots of artifacts and what appears to be very strong noise reduction maybe (?). Looks a bit "watercolorish" to me.Don't know if I'm missing something here, but there are loads of
artifacts and NR on this print. I think this has gone too large.
Very true, but some parameters traditionally considered as "lens parameters" do change. I mean here the the FOV (fiedl of view) which traditionally have been considered to be function of focal lenght, but today is a function of focal lenght and sensor size.A 200mm lens on a 5D is a 200mm lens on a 40D. The optics have not
changed.
You're talking about "pixel density," an entirely different matter
unrelated to sensor size. The original 1D has precisely the same
sensor size as they 1D/2 and 1D/3. The D30 has the same sensor size
as the 40D. Using the "Focal Length Magnification" marketing smoke
that so many apparently have been deluded by, they both "magnify" the
same lens identically.
Sensor size has never changed the optics of a lens in any dSLR.
Of course in history the same "sensor size" or actually film size has mattered, but as 35mm lenses were ment to 35mm cameras only that was no problem. Medium format had their own lenses and I cannot remember if there was ever any APS size SLR body (if was then the same "focal lenght multiplier" have applied).
None of which are d i f f r a c t i o n.Sure it does scatter. Through several mechanisms. Starting withI think you mean "refraction" or "distortion". The atmosphere itself
doesn't diffract.
Rayleigh scattering.
Then there is dust, fog, thermal gradients.
altair8800
Senior Member
- Messages
- 2,000
- Reaction score
- 443
If you crop the 5D image for the same FOV as the 1D III, you end up with only 8 Megapixels. For telephoto, you are moving up in resolution.Coming from a 5D, the only thing I will be giving up when going to a
1D Mark III is the full frame 12.7 MegaPixel sensor. Now, the 1.3
crop factor has it's charms, because I usually shoot telephoto, so
it's like having an extender in your camera, but moving down in
resolution.... will I notice?
How large can you print from 10.1 Megapixels?
Anyone experience the same step?
Dan
Keyboard shortcuts
- f
- Forum
About
Editorial content
Cameras & Lenses
All content, design, and layout are Copyright © 1998–2025 Digital Photography Review All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.