Comment 7 for bug 289852

Hi there,

I've got the same problem. I simply wanted to print this (http://www.fim.uni-linz.ac.at/lva/IT_Recht_Computerforensik/ws2008/Introduction_to_Computer_Forensics.pdf) with 4 pages on one. Printing from okular and acrobat failed due to too heavy memory/cpu load and eventually an error.

It seems as if cups does some heavy filtering (ps takes a lot of cpu cycles), which makes the whole experience not at all enjoyable.

I tried to print the pdf with 4 pages on one with acrobat to a .ps file, which takes only a couple of seconds. The finishing file has only about 10MB. Then I tried to turn the pages in the .ps file around (using ps2ps), but this produced only white pages in the output. Trying to rewrite the .ps file with ghostscript produced a huge (~70MB) output file, but it was at least readable.

I finally managed to print the files by manually pipeing it through my printer-filter and then to lpr -l. The filter is called "filterHL5040" for my Brother HL5040 and obviously produces the right results for my printer. This only takes a couple of seconds for the whole file and works nicely.

So finally I don't understand why cups does not do the same, namely pipeing the input postscript file through the filter and writing it to my printer.

BTW I am using Kubuntu 8.10 32 bit with latest updates with proposed enabled.