[quote user="Phil1923"]However, I'm puzzled as to why the size in meters or pixels has any meaning. Neither do I understand why dpi should have any meaning.[/quote] Hi again, Phil. I will try to explain DPI as best and simple as possible. 1 - if you import an image (bitmap) to CorelDRAW, suppose 1800x1200 pixel, and enlarge or reduce, the image will be 1800x1200 pixel no matter the physical size. On this case, the "resolution" is relative the visual quality of the image: if you reduce the image, each pixel will be a small square, if you enlarge it each pixel will be a bigger square, then the image will looks "pixelated": But the image will be always 1800x1200 pixel, and the size on Mb will be exactly the same, no matter if you enlarge or reduce it. 1- DPI is a printing language, Now we send to print directly to plates (CTP=Computer To Plate) but some years ago, we send to film first, then copy the film to a plate using a photographic method. The resuit should be almost the same (of course, newest printers allows more options and quality). Each plate or film is a color separation (Cyan, Magemta, Yellow, Black, a Pantone color or any custom Spot color).and the result is a big monochrome bitmap for each color. In other words, each color separation is output as a black and white image. the printer meachine will apply the ink according the design, although you can replace it with other (for example, if you send a job using cyan and magenta, the orinter could replace it with Blue and Red or Violet and Orange). For generate this black-and-whithe image for each color separation, you need a RIP (Raster image processor) , a software that output each color separation as a monochrome image. The printer (imagesetter or CTP) only can produce a binary option: black or white, a back point or not, using a laser. The laser print the image on the film or plate line by line. You can choose how many lines will "print" the laser. Some years ago, the standard quality was 150 lines per inch. That means, one inch of advance, and the widht of the material (film or plate) Since each point of the laser is divided on two parts 8two dots), the max number of dots is twice the amount of lines per inch. If you print at 150 lines per inch, you can print at 300 dpi max. If you output at 175 lines per inch (lpi) you can reproduce up to 350 dpi. If you output at 200 lpi, you can reproduce up to 400 dpi. 3 - The quality of each will be relative to the amount of squares that printer use for create each dot. The more squares, more quality. Then, a laser printer that could output at 4800x4800 dpi But now we found the first problem: there're two different things (at least) that use the same value: pi (dots per inch) : the first is each point made with the laser (it could be a circle, ellipse, square, etc usually is used elliptical although in the past the most common shape was circle. The second thing is each of the squares that you need for create each point. The more aoumt of squares per inch will allows to create a better shape. Each square is also called a "dot", then you could have a printer that output at 4800x4800 dpi (each square of 1x1 inch is divided on 4800x4800 little squares) for produce "points" (the circle) at 300 "points" ("dots") per inch, 350 points per inch, etc Then, you can found both values simultaneously. For example, a desktop laser printer that output at 600 dpi (that means, 600x600 little squares) could print 90 lines per inch, then it could reproduce "dots"(the circles) at 180 dpi (90x2). Or a high quality CTP coul output at 9600x9600 dpi (ittle squares) for create rounded dots at 300 dpi (circles). Each little square is not a pixel, because "Pixel" is a virtual size. The squares are printed at real size, is one inch divided on 9600x19600 squares, 4800x4800, 2400x2400, 1200x1200, 600x600 or the quality of the printer. And the points (the circles) that you create with this grid will be 300 "dots" (circles) per inch 4 - Big size inkjet printers (plotters) use a similar concept. A software (RIP) create a grid of squares (for example, 4800x4800 dpi) and apply color to each square for reproduce the image that we send to print. Of course, the printing method and the result will be different than a laser On this case, on most plotters there's not "circles" (only on a few Postscript printer for high qaulity proof) Unfortunately, desktop inkjet printers claim to use the same technology, and said "2400 dpi printing quality" "4800 dpi" etc. 5 - Why each image is measured on dpi? it's an old mistake, and as far I know it was a mistake when Adobe develop the PostScript language. The idea was resample the image in order to match the number of dpi (the printed circles) with the value of pixels, because a high quality bitmap will produce bigger files, and that need more time for send to print. And, since the printer can't reproduce all this information, it's almost useless to send an high quality image inf the printer can't reproduce it (and it was specially important some years ago, when computers and conection were slower than now). The concept was believe that the initial image of the camera was equivakent to 72 dpi (because the standard value for view in monitor was 72 pìxels per inch). of course, it was not real and it has not sense, as we can see on the first point.
↧