Thanks Bob, that totally clears it up for me. -----Original Message----- From: owner-photoforum@xxxxxxxxxxxxxxxxxx [mailto:owner-photoforum@xxxxxxxxxxxxxxxxxx] On Behalf Of Bob Blakely Sent: Friday, August 17, 2007 12:44 PM To: List for Photo/Imaging Educators - Professionals - Students Subject: Re: DPI and perception question Let's say you print the entire 20x24 inches (I assumed inches here, as we are talking about DPI - Dots Per Inch. If you meant something else, cm par example, you would have said so, eh?) 20*300 * 24*300 = 43,200,000 dots Assume ink (dye, wax) mixing technology. Assume each dot is assembled from a 32 bit (4 byte color) word. This is then 43,200,000*4 = 172,800,000 bytes = 172.8 mb. The math changes with the technology used. ================================================== Practical resolution limit of the human eye is about 1 arcmin. s = r * theata, so this means that the eye can resolve objects about r * (1/60)*(pi/180). At one foot (12 in) then, the eye can resolve dots 12*(1/60)*(pi/180) = 0.0035" in size. this is 286 dots per inch. For a creamy smooth, unpixelated look, you DO NOT want to resolve the dots (pixels), so you should have greater than 286 dots per inch. Note: the further away you are from the image, the less DPI you need . Am I right about this? That took a lot out of me. I need a spiritual consultation with the Reverand Jack Daniels. Regards, Bob... -------------------------------------------------------- "Life isn't like a box of chocolates . . it's more like a jar of jalapenos. What you do today, might burn your butt tomorrow." ----- Original Message ----- From: "Paul Weyn" <paul@xxxxxxxxxxxxx> > Hi Andy, > Not sure I understand the process exactly either but when I have a 35mm > negative scanned to be printed at a maximum 20x24 size, the file size > created by my lab is about 80meg. In my mind, that's huge; however, the > results are extremely good - no pixilation at all. However, I do not > understand the math because if the printer can only go as high as 300 DPI > then why is so large a file needed...are the rest of the pixels just > discarded in the process, and if so then why wouldn't a smaller file work > just as well? > > As an aside, the inks and papers are so good these days that compared to a > traditionally made print of the same image I can hardly tell the > difference. > And the advantage to printing digitally is the repeatability factor. Plus > it's really convenient for my lab to keep the file on record and for me to > just call in my order whenever I need one printed. > > All considered, I still favor the older technology...there's an art and > science to it I really want to master. For the last year or so I've been > procuring darkroom equipment. As soon as I have enough to create my own > > -----Original Message----- > From: owner-photoforum@xxxxxxxxxxxxxxxxxx > > Hi, > > I think it is customary practice for printers to "demand" image files at > 300 > dpi (whatever > that is) at final printed size of a reproduction. I guess this is to > reproduce images so > they have a high quality and don't look pixelated or something. (I think I > have > oversimplified things). > > In any case, I was pondering whether one can get a fair idea of whether an > image file has > sufficient digital "resolution" so that when printed it will look "good" > by > looking at the > image at a larger size than what it will be reproduced at. So if I have a > 5x5 cm image > file at 300 dpi but I look at it on my CRT or LCD screen at 200% or 300% > or > 600% or more > magnification and at 300% the image on my screen looks OK ... but at 600% > it > starts to > fall apart ... is that an indication of anything? > > Hope I have not been to obfuscating in this question ... drinking a > Snapple > only.
