Velocity Reviews > Resolution and print size

# Resolution and print size

Marc Wossner
Guest
Posts: n/a

 01-23-2007
Hi ng,

Iīve a problem understanding digital camera resolution and max print
sizes for files derived by those cameras. I take the Canon EOS-D60 as
an example. The D60 has a pixel size of 7.4 microns, so there are 136
pixels/mm. However, that doesn't mean that the camera will actually
resolve 136 lines per mm. The real useful resolution will be 70-80% of
that, something like 105 lines per mm. Also, this is lines, while film
and lens resolution is always given in line pairs, and it takes two
lines to make a line pair. So the D-60 will actually resolve about 53
line pairs/mm at best. - I know that it makes no sense to compare
resolution in lines/mm as the sensor and 35 mm film are different in
size. Lines/picture height are better. For the D60 this means that
itīs measured resolution limit of 1600 lines/picture height is
equivalent to a 35 mm camera resolution of 1600 / 2 / 24 = 33 line
pairs per mm. - Anyway, if I take those 53 lp/mm into the formula to
calculate system resolution like 1/T = 1/l + 1/f = 1/400+1/53 =
1/0,0214 = 46,8 lp/mm and divide this value by the resolution limit of
the eye for 10 inches (6,88 lp/mm), I reach a max magnification of
x6,8. - I know that this formula gives only a rough approximation. -
The sensor is 22.7 x 15.1 mm, so max print size should be 6 inches
wide. Now I found a value based on experience that says that D60 images
printed at 200 dpi (thatīs a full 10x15 inch print) look very good and
sharp. Thatīs consistent with other statements I found for other
cameras that all give values for max print size that are higher than
can be expected from theoretical resolution alone. Itīs quite clear to
me that our perception of sharpness is based on more than resolution
alone, that grain/noise plays an important role as well and that a
camera like the D60 produces images that are virtually noise free. But
as resolution still matters can someone explain to me how those figures
correlate or where I made a mistake in calculating them?

Marc Wossner

SimonLW
Guest
Posts: n/a

 01-23-2007
"Marc Wossner" <(E-Mail Removed)> wrote in message
news:(E-Mail Removed) ups.com...
Hi ng,

Iīve a problem understanding digital camera resolution and max print
sizes for files derived by those cameras. I take the Canon EOS-D60 as
an example. The D60 has a pixel size of 7.4 microns, so there are 136
pixels/mm. However, that doesn't mean that the camera will actually
resolve 136 lines per mm. The real useful resolution will be 70-80% of
that, something like 105 lines per mm. Also, this is lines, while film
and lens resolution is always given in line pairs, and it takes two
lines to make a line pair. So the D-60 will actually resolve about 53
line pairs/mm at best. - I know that it makes no sense to compare
resolution in lines/mm as the sensor and 35 mm film are different in
size. Lines/picture height are better. For the D60 this means that
itīs measured resolution limit of 1600 lines/picture height is
equivalent to a 35 mm camera resolution of 1600 / 2 / 24 = 33 line
pairs per mm. - Anyway, if I take those 53 lp/mm into the formula to
calculate system resolution like 1/T = 1/l + 1/f = 1/400+1/53 =
1/0,0214 = 46,8 lp/mm and divide this value by the resolution limit of
the eye for 10 inches (6,88 lp/mm), I reach a max magnification of
x6,8. - I know that this formula gives only a rough approximation. -
The sensor is 22.7 x 15.1 mm, so max print size should be 6 inches
wide. Now I found a value based on experience that says that D60 images
printed at 200 dpi (thatīs a full 10x15 inch print) look very good and
sharp. Thatīs consistent with other statements I found for other
cameras that all give values for max print size that are higher than
can be expected from theoretical resolution alone. Itīs quite clear to
me that our perception of sharpness is based on more than resolution
alone, that grain/noise plays an important role as well and that a
camera like the D60 produces images that are virtually noise free. But
as resolution still matters can someone explain to me how those figures
correlate or where I made a mistake in calculating them?

Marc Wossner

Just take the # of pixels on the long side of the image, and divide by 300
dpi for high quality prints, 250dpi for good quality or 200dpi for fair
quality, ect. to get the recommended max print size. Very basic and it has
always worked for me.

When you start splitting hairs with numbers, you have to consider lens
sharpness and contrast, actual resolution of the printer, illumination
levels of print to be viewed and so on.
-S

Dr. Joel M. Hoffman
Guest
Posts: n/a

 01-23-2007
>pairs per mm. - Anyway, if I take those 53 lp/mm into the formula to
>calculate system resolution like 1/T = 1/l + 1/f = 1/400+1/53 =
>1/0,0214 = 46,8 lp/mm and divide this value by the resolution limit of
>the eye for 10 inches (6,88 lp/mm), I reach a max magnification of
>x6,8. - I know that this formula gives only a rough approximation. -
>The sensor is 22.7 x 15.1 mm, so max print size should be 6 inches
>wide. Now I found a value based on experience that says that D60 images
>printed at 200 dpi (thatīs a full 10x15 inch print) look very good and
>sharp.

I think a better way of looking at your numbers is this:

Up to 6" wide (or so), you get (almost) no increase in quality from
more pixels, because the increase in resolution is beyond the
resolving power of the human eye.

However, equipment is not perfect, and if any stage of your workflow
introduces fuzziness (which it surely does), you'll find the 6" figure
growing. Suppose, for example (and this is probably true), that you
use a printer that only prints at 200dpi. Then up to 10x15 or so you
get (almost) no increase in quality from more pixels.

It is also true, as you note, that we perceive sharpness based not
only on resolution. This is why sharpening a low resolution image can
make it look sharp, even at the expense of detail. Similarly, I have
seen 3'x4' images printed on canvas from very low resolution images,
and they look great. I suspect this has something to do with the
coarse texture of the canvas, but I don't know for sure.

-Joel

----------------------------------------------------------------------------
Free Tanach and Mishna printouts in Hebrew: http://liturgy.lashon.net/
----------------------------------------------------------------------------

Tzortzakakis Dimitrios
Guest
Posts: n/a

 01-23-2007

? "Marc Wossner" <(E-Mail Removed)> ?????? ??? ??????
news:(E-Mail Removed) ups.com...
Hi ng,

Iīve a problem understanding digital camera resolution and max print
sizes for files derived by those cameras. I take the Canon EOS-D60 as
an example. The D60 has a pixel size of 7.4 microns, so there are 136
pixels/mm. However, that doesn't mean that the camera will actually
resolve 136 lines per mm. The real useful resolution will be 70-80% of
that, something like 105 lines per mm. Also, this is lines, while film
and lens resolution is always given in line pairs, and it takes two
lines to make a line pair. So the D-60 will actually resolve about 53
line pairs/mm at best. - I know that it makes no sense to compare
resolution in lines/mm as the sensor and 35 mm film are different in
size. Lines/picture height are better. For the D60 this means that
itīs measured resolution limit of 1600 lines/picture height is
equivalent to a 35 mm camera resolution of 1600 / 2 / 24 = 33 line
pairs per mm. - Anyway, if I take those 53 lp/mm into the formula to
calculate system resolution like 1/T = 1/l + 1/f = 1/400+1/53 =
1/0,0214 = 46,8 lp/mm and divide this value by the resolution limit of
the eye for 10 inches (6,88 lp/mm), I reach a max magnification of
x6,8. - I know that this formula gives only a rough approximation. -
The sensor is 22.7 x 15.1 mm, so max print size should be 6 inches
wide. Now I found a value based on experience that says that D60 images
printed at 200 dpi (thatīs a full 10x15 inch print) look very good and
sharp. Thatīs consistent with other statements I found for other
cameras that all give values for max print size that are higher than
can be expected from theoretical resolution alone. Itīs quite clear to
me that our perception of sharpness is based on more than resolution
alone, that grain/noise plays an important role as well and that a
camera like the D60 produces images that are virtually noise free. But
as resolution still matters can someone explain to me how those figures
correlate or where I made a mistake in calculating them?

Hi,
I printed a full-size A4 (larger than 8X10")with a cheapo Kodak CX 7300 and
a Canon Pixma IP 4300 and looks perfect, much better than any B&W prints I
did in my film era, save colour.(With some german glossy paper, 25 euros for
50 A4 sheets, compu color by Felix Schoeler).Surely you have a better camera
than mine?
Hope this helps,

--
Tzortzakakis Dimitrios
major in electrical engineering
mechanized infantry reservist
dimtzort AT otenet DOT gr

Paul D. Sullivan
Guest
Posts: n/a

 01-23-2007
> Iīve a problem understanding digital camera resolution and max
> print sizes for files derived by those cameras.

I have historically used the 300 dpi guide, but find that with
digital cameras, I can get pretty good results with a minimum of
about 240 dpi - 250 dpi of resolveable detail, perhaps because
with digital there is no grain or other such distorting factors
to deal with.

This depends on a number of factors, such as the quality of the
camera / ccd / lens, the ISO level of the shot, lighting levels
and type of object being shot.

For outdoor nature shots, I can get away with as low as 200 dpi
and still be somewhat satisfied, but for shots of people,
anything below the 240-250 dpi range and I can start to notice
things.

A good 5mp camera like the Oly C5050 can give you great looking,
300+ dpi prints in 4x6 and 5x7 and still accommodate some
cropping. But for 8x10's, I think 5mp can be stretching it for
non-nature type shots, like portraits, etc.

That's just my personal feeling though. I'm not trying to
indicate that it's right for everybody - it is just what I have
found to be my preferences over the years.

Marc Wossner
Guest
Posts: n/a

 01-25-2007

As you all state the common dpi values is the traditional resolution
measure of line pairs per mm or lines per picture with or picture
height not relevant when calculating the resolution of a digital
imaging system? - I still use silver film and try to figure out the
theoretical foundations before I switch to digital.

Marc Wossner

Gary Eickmeier
Guest
Posts: n/a

 01-25-2007

Marc Wossner wrote:

> As you all state the common dpi values is the traditional resolution
> measure of line pairs per mm or lines per picture with or picture
> height not relevant when calculating the resolution of a digital
> imaging system? - I still use silver film and try to figure out the
> theoretical foundations before I switch to digital.

An interesting aside, Marc - my video projector has a basic 1440 x 1080
pixels resolution, but it displays video at fantastic resolution at a 10
foot wide screen size. To my eye, I can't imagine having any more detail
in some of the images I'm seeing on a good broadcast. My thought is that
all that matters is how big the resultant image is in your eye. If I sat
5 feet away from my screen, it wouldn't fare as well. So I sit about 20
feet back, and it is impressive and amazing.

I have also seen very large paper prints from various digital (and film)
cameras at exhibitions and shows. I guess the proof is in the pudding,
so try various outrageous sizes and see at what point it breaks down.
Probably doable anywhere above 100 DPI.

Gary Eickmeier

David J. Littleboy
Guest
Posts: n/a

 01-25-2007

"Marc Wossner" <(E-Mail Removed)> wrote:
>
> As you all state the common dpi values is the traditional resolution
> measure of line pairs per mm or lines per picture with or picture
> height not relevant when calculating the resolution of a digital
> imaging system? - I still use silver film and try to figure out the
> theoretical foundations before I switch to digital.

Forget the theory: look at the images.

This guy complains about moiré and fine text in street signs getting messed
up, but it sure looks to me that the 5D and 645 (actually, 6x6 cropped to
slightly smaller than 645) are pretty much equivalent for prints 17x26" and
smaller.

http://www.shortwork.net/equip/review-1Ds-SQ-scantech/

David J. Littleboy
Tokyo, Japan

Marc Wossner
Guest
Posts: n/a

 01-25-2007
> On 25 Jan., 15:29, "David J. Littleboy" <(E-Mail Removed)> wrote:
> > "Marc Wossner" <(E-Mail Removed)> wrote:

>
> > As you all state the common dpi values is the traditional resolution
> > measure of line pairs per mm or lines per picture with or picture
> > height not relevant when calculating the resolution of a digital
> > imaging system? - I still use silver film and try to figure out the
> > theoretical foundations before I switch to digital.

> Forget the theory: look at the images.

Not so easy if something keeps nagging your mind. But I agree with you:
Image quality of the 5D is absolutely stunning.

>
> This guy complains about moiré and fine text in street signs getting messed
> up, but it sure looks to me that the 5D and 645 (actually, 6x6 cropped to
> slightly smaller than 645) are pretty much equivalent for prints 17x26" and
> smaller. http://www.shortwork.net/equip/review-1Ds-SQ-scantech/

So thereīs the question again: How can you print that large from the
resolution the 5D gives?
When you calculate 4368 pixels * 0,7 = 3057,6 lines / 2 = 1529 line
pairs / 35,8 mm you reach 42,7 lp/mm.
I donīt know the true resolution of the lens, but as itīs a prime
lens letīs assume 100 lp/mm.
Taking it into account like this 1/T = 1/43 + 1/100 makes it a system
resolution of 30 lp/mm worst case.
The images presented are printed at 17x26" so thatīs a magnification
of x18.
Divide the 30 lp/mm by the 2 lp/mm that are necessary for that diagonal
you get a max magnification of x15.
But the guy shows clippings that appear sharp from 10" and for that
distance you need 6,88 lp/mm to have a sharp impression.
Note that those figures are all based on "normal" 20/20 vision or max
resolvable detail of 1 minute of an arc (see Norman Korenīs site for
the calculation: http://www.normankoren.com/Tutorials/MTF.html).
As I know that you are quite knowledgeable from various other threads
David, can you explain me how this is possible, or better, where my
possible misconception is?

Marc

Paul D. Sullivan
Guest
Posts: n/a

 01-25-2007
I hate math... lol...

> So thereīs the question again: How can you print that large
> from the resolution the 5D gives?
> When you calculate 4368 pixels * 0,7 = 3057,6 lines / 2 = 1529
> line pairs / 35,8 mm you reach 42,7 lp/mm.
> I donīt know the true resolution of the lens, but as itīs a
> prime
> lens letīs assume 100 lp/mm.
> Taking it into account like this 1/T = 1/43 + 1/100 makes it a
> system resolution of 30 lp/mm worst case.
> The images presented are printed at 17x26" so thatīs a
> magnification of x18.
> Divide the 30 lp/mm by the 2 lp/mm that are necessary for that
> diagonal you get a max magnification of x15.
> But the guy shows clippings that appear sharp from 10" and for
> that distance you need 6,88 lp/mm to have a sharp impression.
> Note that those figures are all based on "normal" 20/20 vision
> or max resolvable detail of 1 minute of an arc (see Norman
> Korenīs site for the calculation:
> http://www.normankoren.com/Tutorials/MTF.html).
> As I know that you are quite knowledgeable from various other
> threads David, can you explain me how this is possible, or
> better, where my possible misconception is?
>
> Marc

 Posting Rules You may not post new threads You may not post replies You may not post attachments You may not edit your posts BB code is On Smilies are On [IMG] code is On HTML code is OffTrackbacks are On Pingbacks are On Refbacks are Off Forum Rules

 Similar Threads Thread Thread Starter Forum Replies Last Post Frank ess Digital Photography 0 11-14-2006 05:08 PM Deep Thought Digital Photography 2 08-22-2005 10:31 PM jersie0 Digital Photography 3 10-22-2004 02:32 PM Steven M. Scharf Digital Photography 32 05-16-2004 10:41 PM Cyrus Chvala Digital Photography 2 07-10-2003 08:59 PM