From: W.P. v. P. <pe...@pa...> - 2002-06-19 20:42:23
|
Hi, Thanks for your answers. Concerning the striping, I want to give it a try to implement the 1/4 advances per print sweep. I have studied the code and docs and are ready to go but do you have some tips to bring me up to speed. What is in your opinion the cleanest way to implement this? thanks in advance WP Thanks for the report. Please now use pnm2ppa-1.10pre2 which fixes a bug that sometimes hangs the printer at the end of a page, especially with A4 paper!. Re: scanning at 300dpi. Are you sending the image as pbm output directly to pnm2ppa without ps->pbm conversion by ghostscript? pnm2ppa needs 600dpi input, even though it really reduces the resolution to 300dpi. It would be reasonably simple to add a "--r 300" option to pnm2ppa, that doesnt exist now. to allow pnm2pp to accept 300dpi input. But as of now, it only takes (or assumes it is getting) 600dx600pi, which is related to the physical arrangement of the printer head nozzles. (600x300, 300x600, 300x0300 could easily be accomodated). If 300x300 input is sent to pnm2ppa, it will treat it as a half size image of 600x600, and print at an effective 150x150 resolution. If anything weird happens with pnm2ppa-1.10pre2, can you check whether it also happens with pnm2ppa-1.04, so I know whether it's due to the changes or not. Re Striping: In the tiger.ps example, there still is a tiny bit of striping in the grey background , or rather the lines between print sweeps are still visible. I speculate that this is be to mechanical printer inprecision in feeding the exact amout of paper between sweeps, rather than a pnm2ppa bug. This is when the color prints in 4 passes of the print head. The improvement I made in v1.10 is that the 1st component of color is alway left-to-right, the second is right-to-left, etc even if the sequence in which they are printed is not in order, so the sweep directions alternate. But all four passes are at the same vertical position. The windows driver advances the print sweep 1/4 of a sweep on each of these passes, so each of the four sweeps has a different sweep boundary, and any difference between the nozzles (partial blocking etc) gets averaged out. In principle I could try to implement this, but it adds an extra layer of complexity. The changes I made in pnm2ppa-1.10 do make this easier to implement now, but.... life is short....! -- WP. van Paassen - www.paassen.tmfweb.nl - "First they ignore you, then they laugh at you, then they fight you, then you win." Mahatma Gandhi (1869-1948) |