Well, I'm not so sure how printers work, but I've read conflicting things that claim you cannot simply divide resolution by DPI, and the rational being that DPI and PPI are not the same. I just googled DPI vs PPI and got http://www.tildefrugal.net/photo/dpi.php which seems to agree that DPI doesn't really translate into physical size.
However, it does appear that setting the PPI is the magic parameter I'm searching for. So, all you printing gurus, am I safe to conclude the PPI is my magic ratio for figuring out what size my image will be? If so, the following command appears to do the trick: *lp -o ppi=value filename* On Tue, Mar 4, 2008 at 4:47 PM, Charles Cranston <[EMAIL PROTECTED]> wrote: > In what sense is a dot not a pixel? Are you considering the various > ways of representing pixel densities other than 0 or 1 by using > patterns of multiple dots to represent a single pixel (e.g., dithering)? > > On Mar 4, 2008, at 4:29 PM, Christopher Conroy wrote: > > > Problem is DPI is Dots Per Inch, not Pixels Per Inch. DOTS represent > > an amount of ink, not an amount of pixels AFAIK. I do imagine that > > there exists a linear or otherwise simple function that maps the DPI > > of any printer to PPI, but I don't think it is universal. Figuring > > out this relation involves inaccurate hand measurements and trial > > and error. It's not that bad however, and I'll probably end up > > having to do just that. > > > > But, if a nicer solution exists, I'd be happy. > > > > On Tue, Mar 4, 2008 at 4:01 PM, Rob Sherwood <[EMAIL PROTECTED]> > > wrote: > > On Tue, Mar 04, 2008 at 03:50:30PM -0500, Christopher Conroy wrote: > > > Hey guys. This is a bit of an off the wall question, and my google- > > fu hasn't > > > really produced any answers to my liking. > > > > > > I would like to be able to calculate what size of an image to > > produce in > > > order to print to a target size. Specifically, I'm going to be > > working on > > > some CAD like stuff which requires a fair amount of computation. > > Thus, I > > > need to do all the computational geometry outside of any program > > that > > > supports such a feature (e.g. AutoCAD). I haven't started coding > > any of > > > this, though I will probably use Ruby or Java if that makes a > > difference. > > > Either way, I'll be creating images from the calculations. > > > > > > So, really this is just a matter of figuring out the best way to > > map image > > > resolution in pixels to print dimensions in inches. > > > > > > Anyone know of an easy way to do this aside from trial and error > > testing? > > > I'd like to be as exact as possible, and ideally not have to do > > any image > > > scaling at print time. (Ideally, the solution will work across > > multiple > > > printers. Bonus points for anything OS agnostic.) > > > > Sounds like you need to know the dots per inch (DPI) of what your > > printing technique is. A lot of printers use 300 DPI, so if you > > want a > > 12" image, use 12*300 == 3600 pixels. Maybe you're asking something > > more > > complicated and I'm not understanding, but... > > > > - Rob > > . > > > > > > > > -- > > Christopher Conroy > > -- Christopher Conroy
