The detector in a scanner is a linear CCD array. The array is like a row of buckets. A bright light shines on your document. When photons bounce off your document and hit the buckets, the buckets fill up with electrons. Read-out electronics essentially count the number of electrons in each bucket to know how bright that spot was. The CCD is a silicon integrated circuit (IC). The native resolution in the horizontal direction is determined by the bucket spacing, which can't be changed in software or firmware. My guess is that even though ICs are normally designed in micrometers, the bucket spacing is probably chosen to be a particular round number in dots per inch (DPI). The other dimension is filled in by scanning the linear array, so the resolution in this dimension is determined by optics and the stepper motor moving the array. The resolution in the second dimension is normally close to the resolution in the first, (after all, pixels on computers screens are nearly square.)
There are plenty mathematical techniques, called "deconvolution," for increasing resolution higher than the native resolution. However, they are extremely CPU intensive and generally don't gain much. So, practically speaking, you can't do better than the native resolution. On the other hand, it's easy to lower the resolution in software. John On Sunday 09 November 2003 10:47, Mighty Chimp wrote: > Could you explain further what you mean by "native resolution"? > > Euric > > > > -----Original Message----- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On > Behalf Of John S. Ward > Sent: Sunday, 2003-11-09 13:33 > To: U.S. Metric Association > Subject: [USMA:27521] RE: USENET list > > Euric, > > I'm neither Marcus nor a computer science genius. But I do know that > the > actual hardware resolution is referred to as the "native resolution." > All > the scanners I've seen have native resolutions in hard-inches. > > John > > On Sunday 09 November 2003 09:32, Mighty Chimp wrote: > > I made the assumption that the use of dpi is not part of a printers > > hardware, but of its software. Either through its drivers or the word > > programmes. Could you tell us, if this is actually true or not? > > --- > Incoming mail is certified Virus Free. > Checked by AVG anti-virus system (http://www.grisoft.com). > Version: 6.0.537 / Virus Database: 332 - Release Date: 2003-11-06 > > > --- > Outgoing mail is certified Virus Free. > Checked by AVG anti-virus system (http://www.grisoft.com). > Version: 6.0.537 / Virus Database: 332 - Release Date: 2003-11-06
