Another attempt. I'd like to know what Lightroom 4 does when printing.
Assume that I have an image of 1000x1000 pixels. There's no dpi, no ppi, no any other resolution defined, just the pixel size of the image.
Then, let's assume I have a Canon ip4950 printer. Maybe a bad example because you just know that the printer does 9600dpi, but you have no clue how many ppi's it has (i.e. the native "real" print resolution). Let's assume it does 600ppi (maybe that's even the truth?).
My understanding is that the term "ppi" refers to the number of "pixels" that the printer can place on paper within one inch. Each single "pixel" is a combination of one or more ink dots of different colors: The printer cannot mix ink, so it needs to place tiny dots close to others (it changes the pattern using "dithering") to form one "color pixel". Such a pixel is then what counts for the ppi resolution. So, the printer might be capable of doing 9600dpi, but since several dots are needed to form a single "pixel", we end up - in this example - with only 600 ppi. Ppi is not dpi :-)
Now, I want to print my image with all the resolution that my printer can do. I don't want to make my image a specific size, all I want is to get my image out of the printer with 1 pixel of my image = 1 printer "pixel" (knowing that 1 printer pixel is made of some dithered color dots). So, I don't care whether my printout would be poster or postage stamp size in the end.
Now, I call up Lightroom's print tab. Besides of the paper size - let's assume A4 - I can define the size of a singular rectangular area there which contains my picture. I have the respective info box activated, so in the print preview, I see an overlay of the size of the rectangular area as well as the resolution of the image contained within. Let's assume I see "272,217 x 204,4 mm @ 428 ppi" there: I have my Image centered on the A4 paper in landscape mode and sized so it almost fills the paper from top to bottom. Nevertheless the're white borders left and right of the image, and small white borders at the top and the bottom.
From what I see I would conclude that if I would print this out, the resolution capabilities of my printer would exceed the resolution of my image: I see 428ppi in Lightroom, and I know my printer can do 600ppi.
So, as a consequence, I would make the rectangular area in Lightroom smaller, in other words, I would shrink my image until the info box reads "...@600ppi". That's what I'd assume as being the "right" resolution because it matches the printer ppi exactly. Sure, I'd have much more white borders around my image, but I don't care.
That's my theory. On the other hand, I have read things such as that printer drivers do their own thing regardless of the resolution of the image sent to the printer etc. etc.
And, there's that "print resolution" check box and input value in Lightroom. My explanation for this:
As I have described above, Lightroom seems to be incapable of just taking an image file of size x,y and put it on the paper as it is. Instead, it puts that paper into a rectangular area of which I myself need to define the size. Since the size of the box is measured in length units and not in pixels, there's no connection between my image size x and y and the width and heigth of that box. In other words, I cannot size the box exactly the size of my image (in pixels). I need to work around this more bad than good by taking the image ppi and kind of estimate the printer ppi if I don't know it - the process I described above.
If I now activate the "print resolution" check box, I add another level of complexity to the print process: I force the printer to use a specific ppi instead of the one that Lightroom already calculated. So, in my example, if I set the "print resolution" to 500 ppi, this would be more than the 428 ppi Lightroom is giving me. What Lightroom or the printer do in this case I don't know, but I suspect a process called "upscaling" would be invoked: Since there's no more image information that 428ppi, Lightroom needs to interpolate to fill the gap to 500ppi.
Question: Am I correct with the perception that I have, or where does my model need corrections, and does someone have links to resources with explanations?