Path: utzoo!censor!geac!torsqnt!news-server.csri.toronto.edu!cs.utexas.edu!yale!hsdndev!cmcl2!uupsi!sunic!dkuug!iesd!iesd.auc.dk!fischer From: fischer@iesd.auc.dk (Lars P. Fischer) Newsgroups: comp.text.tex Subject: Re: dvips (was: Re: tex dvi Previewer for Xwindows) Message-ID: Date: 7 Jan 91 18:27:56 GMT References: <139@dogmelb.dog.oz.au> Sender: news@iesd.auc.dk (UseNet News) Followup-To: comp.text.tex Organization: Mathematics and Computer Science, University of Aalborg Lines: 31 In-reply-to: hanche@imf.unit.no's message of 7 Jan 91 15:58:59 GMT >>>>> On 7 Jan 91 15:58:59 GMT, hanche@imf.unit.no (Harald Hanche-Olsen) said: Harald> Hmmm, yes, it has lots of good features, but when I last tried it the Harald> printer spent 2-3 times at much time per page with dvips output than Harald> with dvialw output. In my experience, dvips is a lot faster than just about any other dvi->ps program I have tried, *especially* for large output. dvips works a bit differently than most other dvi->ps programs: it tries to download all fonts and other postscript setup before it starts printing. This means that there will be a delay before printing begins, but after that it prints at nearly the maximum speed for the print engine (~7.8 ppm on QMS PS/810). For single page documents this setup won't help much, of course. Also, it is important to set the amount of RAM in the printer correctly in the setup.ps file. Harald> We have an Imagen 2308/S printer, and its PostScript Harald> processor can usually process TeX pages as fast as the print engine Harald> can crank them out. With dvips output, it's a different story Harald> altogether. I've never seen such a beast, so I can't say. With QMS and Apple printer, dvips flies. /Lars -- Lars Fischer, fischer@iesd.auc.dk | Q: How does a project get to be one CS Dept., Univ. of Aalborg, DENMARK. | year late? A: One day at a time.