What equivalent size print is 1:1 ?

[Mark F]

In checking sharp focus in LR2 I often look at an important section of the image at 1:1 (and sometimes at 2:1).  If I were to make a full sized print at that magnification what size print would I be getting? On my 24" iMac the image is about 9.5 x 14.   Also, please tell me how you make the computation so I can do it myself next time.

Thanks.

Mark

[Tim Gray]

Depends on the DPI setting you use to print.  360 dots per inch is a reasonable rule of thumb, so just divide the pixel dimensions by 360 and that will give you the size.  Depending on content you can stretch the 360 to as much as 180 without significant loss of detail -  so there can be quite a range of sizes for "1:1" printing.

[Mark F]

Thanks, but I am not sure I understand.  How do I figure out the pixel dimensions? Is this camera information or monitor information?  Or is it part of the LR2 configuration?

Depends on the DPI setting you use to print.  360 dots per inch is a reasonable rule of thumb, so just divide the pixel dimensions by 360 and that will give you the size.  Depending on content you can stretch the 360 to as much as 180 without significant loss of detail -  so there can be quite a range of sizes for "1:1" printing.

[DarkPenguin]

I do not understand the question.

Are you asking what size print you would get if you printed an image out at 1:1?  Divide the image size by the dots per inch that you are specifying when you print.

Are you asking what size print you would get if you managed to print it out such that it was as big as what you are seeing on your monitor?  Determine the percentage of the image size that is visible on screen.  Measure your screen.  Divide that by the percentage.

[Dale_Cotton2]

Not knowing what camera you're working from and what your monitor specs are, I can only answer by example. Let's say I'm looking at an image from a 6 mp dSLR in LR at 1:1. The file will be 3000 pixels by 2000 pixels in size. Those numbers can be found when in LR's Library module by looking near the bottom of the Metadata pane on the right. You'll see a line called Dimensions with two numbers like 3000 x 2000 (although I'm using LR1.4, not LR2, so you may have to hunt around in a different place).

That's half the battle, the next half is to calculate what your monitor is doing. Mine is set to 1600 pixels x 1200 pixels resolution and, taking a ruler to it, the screen size is 12" x 16", which makes things very simple. 1600 pixels divided by 16" = 100 pixels per inch.

Going back the LR, with the 3000 x 2000 pixel image loaded at 1:1 display, and with the monitor showing me 100 pixels for every inch, I know that the image would have to be printed on paper at 30" x 20" to match what I'm seeing on the monitor. Use the same recipe but plug in your own camera's image file output size and your own monitor numbers, and let us know what you come up with.

[Slobodan Blagojevic]

Your 24" Mac has 94 ppi (pixels per inch) screen. When you are viewing your image at 1:1, it is equivalent of looking at a the same size print printed at 94 dpi (dots per inch). Say your file is 4000 x3000 pixels (or 12 mega pixels). When you look at it on screen at 1:1, it is the same as looking at a 42" x 32" print. Of course, the printout at 94 dpi is not going to look as nice as a minimum 180 dpi one, but at least you get the idea of the size of print.

[Mark F]

Let's see...

The Dimension per LR2 is  3744 x 5616
My monitor is set at 1920 x 1200 resolution and the viewing area of my 24" iMac is actually only 20.5" (W)  x 12.625" (H)

So trying to follow your example, 1920 pixels divided by 20.5" = 93.66 pixels per inch.   And to match what I am seeing on the screen (set 1:1) I would have to make a print 40 x 60 (?)

In case anyone is wondering why I asked the question, upon looking at many of my images at 1:1, they were not as sharp as I expected.  Then it occurred to me that maybe the equivalent print at 1:1 is so large that it was not actually lack of sharpness but instead lack of resolution.  If I did the calculation correctly and the 1:1 view is really the equivalent of a 40" by 60" print, then I do not feel so worried.  For the record, my images were shot with the 1Ds3 using the 70-200 2.8L and 28-70 2.8L lenses. All images taken on a tripod with most on mirror lock.  So if there actually is a sharpness problem it would likely be in the autofocus area and I would have to send the equipment back to Canon for adjustment.

Thanks for the help and please let me know if I did the computation correctly. And thanks also to everyone else who answered.

Mark

Not knowing what camera you're working from and what your monitor specs are, I can only answer by example. Let's say I'm looking at an image from a 6 mp dSLR in LR at 1:1. The file will be 3000 pixels by 2000 pixels in size. Those numbers can be found when in LR's Library module by looking near the bottom of the Metadata pane on the right. You'll see a line called Dimensions with two numbers like 3000 x 2000 (although I'm using LR1.4, not LR2, so you may have to hunt around in a different place).

That's half the battle, the next half is to calculate what your monitor is doing. Mine is set to 1600 pixels x 1200 pixels resolution and, taking a ruler to it, the screen size is 12" x 16", which makes things very simple. 1600 pixels divided by 16" = 100 pixels per inch.

Going back the LR, with the 3000 x 2000 pixel image loaded at 1:1 display, and with the monitor showing me 100 pixels for every inch, I know that the image would have to be printed on paper at 30" x 20" to match what I'm seeing on the monitor. Use the same recipe but plug in your own camera's image file output size and your own monitor numbers, and let us know what you come up with.

[Slobodan Blagojevic]

... Then it occurred to me that maybe the equivalent print at 1:1 is so large that it was not actually lack of sharpness but instead lack of resolution.  If I did the calculation correctly and the 1:1 view is really the equivalent of a 40" by 60" print, then I do not feel so worried ...

Indeed. For the same reason a print at 94 dpi is never going to look as sharp/good as one printed at 180-360 dpi. Also, take into account that you are looking only at a portion of a 40x60 image from your usual screen-reading distance. In order to see the whole 40x60 image, you would have to get up and step away from the screen, in which case the sharpness would not look that bad either  

[Dale_Cotton2]

Mark: looks to me as though you're spot on, both in calculation and in reasoning.

And to add to what Slobodan56 has stated, I feel that with practice/experience one can learn to translate between the degree of sharpness/acutance (such as it is) you see at 1:1 on-screen and the results you get when you print at (presumably) a much smaller/tighter size.

So if there actually is a sharpness problem it would likely be in the autofocus area and I would have to send the equipment back to Canon for adjustment.

I'll probably be re-stating things you already know, but just in case:

All pictures are going to be a bit soft out of the camera because of the anti-aliasing filter and who knows what other inevitable losses. In my experience, Canon's recommendation of USM at Amount < 150, Radius = 0.3, and Threshold = 0 should snap the image back into focus.

If not, the fact that images from both lenses look equally soft tends to eliminate either lens as a culprit. You could do a test shot using manual focus set to the measured distance between camera and target, then compare the results at 1:1 with shots taken at the same time with AF. Perhaps you could even systematically mis-focus both ahead and behind the target to see if a certain correction factor consistently works for both lenses. That would be useful both in communicating with Canon, and in testing the results of their work when the camera comes back.

[bjanes]

Depends on the DPI setting you use to print.  360 dots per inch is a reasonable rule of thumb, so just divide the pixel dimensions by 360 and that will give you the size.  Depending on content you can stretch the 360 to as much as 180 without significant loss of detail -  so there can be quite a range of sizes for "1:1" printing.

Tim,

I think that it is helpful to distinguish pixels from dots. For example, halftone screening as used in newspapers print a pixel as multiple variably sized dots in a fixed raster pattern. Inkjet printers usually use error diffusion dither in which each pixel is represented by multiple fixed sized dots with a variable and random appearing placement. With error diffusion, the relationship between pixels and dots is rather indirect. The "native resolution" of Epson printers is said to be 360 pixels per inch, and the dots are placed with a resolution of 720 x 2880 or 1440 x 5760 dots per inch.

Experiments done by Rags Gardner indicate that the Epson 2200 printer lays down the ink dots at 288 LPI (density), not the 360 as usually quoted.

Bill

[Mark F]

Thanks again, Dale.  I think I will follow your advice and do a series of controlled tests with both lenses using both manual and auto-focus and at various focal lengths. That should put this to bed.

Mark

If not, the fact that images from both lenses look equally soft tends to eliminate either lens as a culprit. You could do a test shot using manual focus set to the measured distance between camera and target, then compare the results at 1:1 with shots taken at the same time with AF. Perhaps you could even systematically mis-focus both ahead and behind the target to see if a certain correction factor consistently works for both lenses. That would be useful both in communicating with Canon, and in testing the results of their work when the camera comes back.