From: Digestifier To: Linux-Misc@senator-bedfellow.mit.edu Reply-To: Linux-Misc@senator-bedfellow.mit.edu Date: Mon, 10 Oct 94 23:13:09 EDT Subject: Linux-Misc Digest #915 Linux-Misc Digest #915, Volume #2 Mon, 10 Oct 94 23:13:09 EDT Contents: Re: 320x200 Xconfig line (Harm Hanemaaijer) Re: ez (was Re: Word (Text) processors for Linux?) (davis@pacific.mps.ohio-state.edu) PPP Providers? Dialup? ISDN? (Thomas E Zerucha) Re: c and me (Dan Pop) Re: Curious: Why is Linux DOOM so much slower than DOS doom (Harm Hanemaaijer) Re: 320x200 Xconfig line (Peter Mutsaers) Re: Linux Dialback software anywhere? (Rob Janssen) Re: Dialup problem (Rob Janssen) Re: Nailed down to 386bsd or linux, now which one? (J.J. Paijmans) Re: Word (Text) processors for Linux? (J.J. Paijmans) Re: Linus' visit to Perth (Kurt M. Hockenbury) Re: Word (Text) processors for Linux? (davis@pacific.mps.ohio-state.edu) Re: Q: own Bootdisk with Ramdick ? (J.J. Paijmans) [Q] capture echo & error ? (david her) Re: Fidonet s/w for linux? (Thomas Parmelan) ---------------------------------------------------------------------------- From: hhanemaa@cs.ruu.nl (Harm Hanemaaijer) Subject: Re: 320x200 Xconfig line Date: Mon, 10 Oct 1994 15:21:25 GMT In <377ecr$ifg@nntp1.u.washington.edu> tlines@u.washington.edu (Tim Lines) writes: >I know this question has probably been asked a thousand times, >so feel free to flame me for asking the thousand and first. >Here goes: How do I set up my Xconfig to support a "320x200" >resolution. It's a standard enough video mode that I suspect >the same line added to the Xconfig would work anywhere. So >flame away guys, but I will consider you unqualified to >flame me if you don't know the answer. You'll have to >excuse me now, I've got to hurry and pick up my asbestos >suit from the cleaners... There is no sane SVGA style 320x200 mode timing that produce the standard VGA horizontal and vertical scan rates. The problem is that the standard VGA 200-line modes use a certain 'doublescan' bit that XFree86 2.1/2.1.1/ 3.1 do not know about (3.1.1 will). If you attempt to define a 320x200 mode anyway with a 25 MHz dot clock you are likely causing a horizontal sync twice and a vertical sync four times as high as the standard VGA 320x200 mode needs. This is way above the spec of most monitors, even most high-end ones. If even if such a mode appears to work it may be operating at a sync rate 20% or so above spec and could cause damage. If your server/chipset driver provides a 12.5 MHz clock, you can define a sane 320x400 mode and work with that. Here's a little C program that will enable/disable the VGA doublescan bit. It should safe, but you never know so try it AT YOUR OWN RISK. Run it while in the server with a mode like 320x400. Read the description at the start. ===== /* * This program enables the VGA scanline doubling bit at * CRTC register 0x09. * It should work on all SVGA chipsets that honour the bit * when in SVGA mode. * * It requires double vertical timings to be programmed, i.e. * set up mode timings for 320x400, and use this program while running * the server in this mode to get an effective 320x200 resolution with * exactly the same monitor timings. This means it should be safe * given that the 320x400 mode is within specs. * * Sample mode: (70 Hz, 31.5 kHz hsync) * Modeline "320x400" 12.588 320 336 384 400 400 409 411 450 * * Note that the cursor position mapping will be incorrect after * enabling the doublescan bit. * * If the driver you are using doesn't provide a 12.5 MHz clock * you are out of luck. There are ways to obtain such a clock * but experimenting can be dangerous. The only safe way is to have * the chipset driver provide it in its set of clocks. * * Here are alternative low-res mode with doublescan with a 25 MHz * clock, but they require a high horizontal sync. THESE MODES CAN ONLY * WORK IF YOUR MONITOR IS SPECED FOR 40/48 kHz HORIZONTAL SYNC (which * usually means a monitor than can do 1024x768 non-interlaced at * at least 60 Hz). The first mode results in 480x300, with the same * aspect ratio as 320x200. The second is 400x300, derived from * 800x600 @ 72 Hz. * * # 480x600, 63 Hz, 39.6 kHz hsync * Modeline "480x600" 25 480 496 576 632 600 601 605 628 * # 400x600, 72 Hz, 48.0 kHz hsync * Modeline "400x600" 25 400 424 488 520 600 637 643 666 * * Here's an alternative mode with a dot clock of 18 MHz that * results in 400x300: * * # 400x600 @ 56 Hz, 35.2 kHz hsync * ModeLine "400x600" 18 400 416 448 512 600 601 603 625 * * Integrated doublescan support with a mode timing flag will be * in XFree86 3.1.1. */ #include #include static inline void outb( int port, int value ) { __asm__ volatile ("outb %0,%1" : : "a" ((unsigned char)value), "d" ((unsigned short)port)); } static inline int inb( int port ) { unsigned char value; __asm__ volatile ("inb %1,%0" : "=a" (value) : "d" ((unsigned short)port)); return value; } #define ENABLE 1 #define DISABLE 2 void main( int argc, char *argv[] ) { int crtcport, CR9; int mode; if (argc != 2) { printf("Run with a single argument '1' to enable doublescan "); printf("and '0' to disable.\n"); exit(0); } mode = 0; if (argv[1][0] == '0') mode = DISABLE; if (argv[1][0] == '1') mode = ENABLE; if (mode == 0) { printf("Invalid argument.\n"); exit(-1); } ioperm(0x3cc, 1, 1); crtcport = 0x3d4; if ((inb(0x3CC) & 0x01) == 0) crtcport = 0x3b4; ioperm(crtcport, 2, 1); outb(crtcport, 0x09); CR9 = inb(crtcport + 1); if (mode == ENABLE && (CR9 & 0x80)) { printf("CRTC Scan Double bit already enabled.\n"); exit(0); } if (mode == DISABLE && (CR9 & 0x80) == 0) { printf("CRTC Scan Double bit already disabled.\n"); exit(0); } outb(crtcport, 0x09); if (mode == ENABLE) { outb(crtcport + 1, CR9 | 0x80); printf("CRTC Scan Double bit enabled.\n"); } else { outb(crtcport + 1, CR9 & 0x7f); printf("CRTC Scan Double bit disabled.\n"); } exit(0); } ------------------------------ From: davis@pacific.mps.ohio-state.edu Crossposted-To: comp.unix.questions Subject: Re: ez (was Re: Word (Text) processors for Linux?) Date: 10 Oct 1994 15:34:29 GMT Reply-To: davis@amy.tch.harvard.edu In article <37bguj$3qv@canopus.cc.umanitoba.ca>, rahardj@cc.umanitoba.ca (Budi Rahardjo) writes: : byron@gemini.cc.gatech.edu (Byron A Jeff) writes: : : : Linux needs it's own Wordprocessor. Something simple, elegant, and moderately : : powerful. And we need it yesterday. : : The question is how to accomplish this? : : Who about Andrew's "ez" ? I think that's what you want. It is X only. I think Linux needs something more general for non-X terminals. I spend 90% of my time using MS-Kermit connected via a dialup line. I imagine that I am not alone. -- _____________ #___/John E. Davis\_________________________________________________________ # # internet: davis@amy.tch.harvard.edu # bitnet: davis@ohstpy # office: 617-735-6746 # ------------------------------ From: zerucha@shell.portal.com (Thomas E Zerucha) Subject: PPP Providers? Dialup? ISDN? Date: 8 Oct 1994 22:44:23 GMT Who has the best rates for dialup PPP access. I am in the Southeastrn Michigan (810 or 313 area code). I am currently using Merit, but my local dialins are 2400 baud. I have a local Compuserve indial and know of some services, but they have hourly surcharges (which aren't bad, but I want to shop around). Does anyone know of a good PPP Provider I can dialup with a local call at a reasonable speed? Also, can I get ISDN through a local provider at a reasonable cost? --- zerucha@shell.portal.com - main email address ------------------------------ From: danpop@cernapo.cern.ch (Dan Pop) Subject: Re: c and me Date: Wed, 5 Oct 1994 19:39:12 GMT In <36uj3bINNh31@mickey.eng.gulfaero.com> bmccarth@gulfaero.com (Bill McCarthy) writes: >reall basic intro to C for linux users-type book. To give you an >idea of what I've tried: borrowed a book on C from work. Tried the >newbie program that supposed to print "hello, world". Typed it in >as per instructions in the book. Ran cc hello.c and got a parse >error. Duh....huh? Took a look at LJ #5, and wouldn't you know it, Nobody can help you without seeing what you were doing, based only on a vague description of the problem, like yours. So, next time post something like: ues4:~/tmp 18> cat hello.c #include main() { printf("hello, world!\n"); return 0; } ues4:~/tmp 19> gcc hello.c ues4:~/tmp 20> ./a.out hello, world! As you can see, it worked for me :-) Dan -- Dan Pop CERN, CN Division Email: danpop@cernapo.cern.ch Mail: CERN - PPE, Bat. 31 R-004, CH-1211 Geneve 23, Switzerland ------------------------------ From: hhanemaa@cs.ruu.nl (Harm Hanemaaijer) Subject: Re: Curious: Why is Linux DOOM so much slower than DOS doom Date: Mon, 10 Oct 1994 15:34:06 GMT In kkto@ipc14.csd.hku.hk (To Kar Keung Isaac) writes: >First, DOOM in DOS have the permission to do anything on the machine, but Linux >one can't. The DOS one actually use DMA to transfer data from memory to DMA, >while the Linux one call X to display an image. What it means, with shared >memory, is to copy the data to an area provided by X, then wait X to find >whether any clipping is necessary (e.g. if another window obscure the DOOM >window that shouldn't be displayed), and finally the X server will copy that to >the video memory after a color mapping. That long process should be the >bottleneck of linxdoom. It is not on most systems. If you run ps -u after running Doom for a while, you'll find that time spent in the X server (i.e. shared memory image overhead, copy to video memory, X overhead) is very much less than the time spent in the Doom process itself (although some of the shared memory image overhead could be in the Doom process). >Second, DOOM in DOS is near to the sole memory user. In Linux, it must compete >with all other clients, like the Xserver, the 4 virtual console, the window >manager, all system daemons, etc., and must also compete CPU time with them. >This is another bottleneck of the linxdoom. I don't think this is a significant factor. Programs/processes that don't need CPU don't use CPU. That's the whole point of Linux. hhanemaa@cs.ruu.nl ------------------------------ From: plm@atcmp.nl (Peter Mutsaers) Subject: Re: 320x200 Xconfig line Date: Mon, 10 Oct 1994 16:10:26 GMT >> On Mon, 10 Oct 1994 15:21:25 GMT, hhanemaa@cs.ruu.nl (Harm Hanemaaijer) said: HH> There is no sane SVGA style 320x200 mode timing that produce the standard HH> VGA horizontal and vertical scan rates. The problem is that the standard HH> VGA 200-line modes use a certain 'doublescan' bit that XFree86 2.1/2.1.1/ HH> 3.1 do not know about (3.1.1 will). If you attempt to define a 320x200 HH> mode anyway with a 25 MHz dot clock you are likely causing a horizontal HH> sync twice and a vertical sync four times as high as the standard HH> VGA 320x200 mode needs. This is way above the spec of most monitors, even HH> most high-end ones. If even if such a mode appears to work it may be HH> operating at a sync rate 20% or so above spec and could cause damage. HH> If your server/chipset driver provides a 12.5 MHz clock, you can define HH> a sane 320x400 mode and work with that. However, the SVGA server from XFree 3.1 knows 320x200x256 without the need for a modeline; choose generic svga and it works. -- Peter Mutsaers | AT Computing bv, P.O. Box 1428, plm@atcmp.nl | 6501 BK Nijmegen, The Netherlands tel. work: +31 (0)80 527248 | tel. home: +31 (0)3405 71093 | "... En..., doet ie het al?" ------------------------------ From: rob@pe1chl.ampr.org (Rob Janssen) Subject: Re: Linux Dialback software anywhere? Reply-To: pe1chl@rabo.nl Date: Sat, 8 Oct 1994 18:41:39 GMT In <36jiai$f6s@kubds1.kub.nl> gevel@kub.nl (Rutger van de GeVEL) writes: >Hi all, >Well, the subject says is all. I'm looking for dialback software for >Linux. Does anyone know where to find is (or that it even exists)? My ZyXEL program (available on sunsite in /pub/linux/system/Serial) can do this. It is written for ZyXEL modems, but with some care and by passing the -Z flag you can use it on some generic fax+data modems as well... Rob -- ========================================================================= | Rob Janssen | AMPRnet: rob@pe1chl.ampr.org | | e-mail: pe1chl@rabo.nl | AX.25 BBS: PE1CHL@PI8UTR.#UTR.NLD.EU | ========================================================================= ------------------------------ From: rob@pe1chl.ampr.org (Rob Janssen) Subject: Re: Dialup problem Reply-To: pe1chl@rabo.nl Date: Sat, 8 Oct 1994 18:42:56 GMT In stinger@romeo.rlmk.no (Stinger) writes: >I'm having a small problem getting dialup to work 100%. That is; it works >most of the times, but every once in a while nothing happens after I >get the connect message. This means that I have to hangup and call again >to get the "login:" prompt. >The modem is a Zyxel 1496 Plus and I've got a 16550. The software I'm using >is Slackware 2.0 with the 1.1.45 kernel. I'm using the recommended line for >dialups in inittab with the exception that I've added RTS/CTS flow control. >Any suggestions? Use my special program for the ZyXEL, available from sunsite in directory /pub/linux/system/Serial. For me it works all the time. Plus you can use it as a FAX and an answering machine as well... Rob -- ========================================================================= | Rob Janssen | AMPRnet: rob@pe1chl.ampr.org | | e-mail: pe1chl@rabo.nl | AX.25 BBS: PE1CHL@PI8UTR.#UTR.NLD.EU | ========================================================================= ------------------------------ From: paai@kub.nl (J.J. Paijmans) Crossposted-To: comp.os.386bsd.misc Subject: Re: Nailed down to 386bsd or linux, now which one? Date: 5 Oct 1994 19:19:04 GMT In article tinguely@plains.NoDak.edu (Mark Tinguely) writes: ... > >3) there is a slight culture difference between all of the groups. IMO, Linux > is more DOS friendly, *BSD are more DOS-phobic. NetBSD is multi-platform, ^^^^^^^^^^^^ Ummm. If I see what the "DOS friendly" linux users here have to say about DOS and MS-Windows, it certainly makes me curious how the DOS-phobic BSD crowd talk about the Microsoft world... :-) Paai ------------------------------ From: paai@kub.nl (J.J. Paijmans) Crossposted-To: comp.unix.questions Subject: Re: Word (Text) processors for Linux? Date: 5 Oct 1994 19:25:44 GMT In article <36ugha$2p5@sashimi.wwa.com> blackbob@wwa.com (Terence S. Murphy) writes: >In article <36u4r9$bcp@kubds1.kub.nl>, J.J. Paijmans wrote: > >>Mind that emacs IS the other thing you are looking for. >>The combination of LaTeX and emacs can't be beaten. > >Does editing with emacs offer additional features for the LaTeX user that >aren't present in vi? I'm curious about what they are, since I really love >vi/LaTeX, and don't have problems with it; of course, I haven't done anything >very complicated vith LaTeX. (This is a question, not a flame :) >-- I think that emacs as a wordprocessor/editor is in a league of its own, regardless if you are using it as a programmers editor or as "front-end" for LaTeX. That is how I meant this remark. But yes, Emacs comes with pre-configured assistance for *.tex and *.bib files (e.g. templates for bibtex-entries and stuff like that). If Matt Welsh continues his Emacs-tutorial in Linux Journal, he eventually will get to this subject. (I hope). Paai. ------------------------------ From: kmh@linux.stevens-tech.edu (Kurt M. Hockenbury) Subject: Re: Linus' visit to Perth Date: Sat, 8 Oct 1994 22:53:59 GMT Rob Janssen (rob@pe1chl.ampr.org) wrote: : Ok, let's make a start: : 1. November 2, 1993. NLUUG, Ede, the Netherlands 2. May 12, 1994. DECUS, New Orleans, USA. 3. Summer 1994 (I don't know the date). USENIX, Boston, USA. ------------------------------ From: davis@pacific.mps.ohio-state.edu Crossposted-To: comp.unix.questions Subject: Re: Word (Text) processors for Linux? Date: 10 Oct 1994 05:13:42 GMT Reply-To: davis@amy.tch.harvard.edu In article <379kpg$3ji@solaria.cc.gatech.edu>, byron@gemini.cc.gatech.edu (Byron A Jeff) writes: : You and I are in agreement on this, Richard. Linux is in desparate need : of it's own wordprocessor, not a typesetter. : : The problem is what's be best/fastest way to accomlish this. Some observations : : 1) Must be native. Until we reach a point where emulation is standard and : stable in Linux distributions we need a unpack and go package. I do not think that portability is much of an issue here. The actual code that is specific to Linux would be small. If it runs under Linux, it should run under SunOS, Ultrix, etc... : 2) Simple. Unfortunately that means that Richard's pet peeve - multiligualness (See below) : : 3) Multi-layered interface. I know GUI's are cool. I know that WYSIWYG is hot. : But frankly a GUI/WYSIWYG interface isn't much good when I'm sitting I agree. In fact, I say that simply make it run on a character based terminal first and worry about X later. I simply would refuse to work on it if it is X only. The main problems are not going to be the display but the underlying representation of the data. If this part is not done correctly, it will never ``sell''. : The question is how to accomplish this? A while back a group was formed to write a word processor for Linux. However, I knew from the beginning that it would fail. If another group is formed to start this project again, please consider why the project failed and do not make the same mistakes again. The reason for its failure is simple: The initial goals and expectations were too high. The wordprocessor was to read/write LaTeX files. It was to be extensible like emacs. It was a pipe dream. Think about what features you really need and do not expect to compete with TeX/LaTeX. Instead, concentrate on simply getting something like PC-Write about 1984 or so. Do not even worry about reading/writing Word/Wordperfect files. These commercial wordprocessors contain features that the initial Linux one will not be able to handle any way. Just concentrate on the Wordprocessor itself and worry about compatablility with other WPs. Once you get that level you can start adding more features-- like thinking about an X version. If you set realistic goals I will be more than happy to participate. -- _____________ #___/John E. Davis\_________________________________________________________ # # internet: davis@amy.tch.harvard.edu # bitnet: davis@ohstpy # office: 617-735-6746 # ------------------------------ From: paai@kub.nl (J.J. Paijmans) Subject: Re: Q: own Bootdisk with Ramdick ? Date: 10 Oct 1994 16:54:16 GMT In article <1994Oct10.103745.28589@uklirb.informatik.uni-kl.de> duhl@informatik.uni-kl.de (Stephan Duhl) writes: >-- >Hi all, > >I want to create my own Bootdisk with a Ramdisk, like the Slakeware Boot/Root ... I don't know about the Ramdisk, but the subjectline of this message alarms me greatly... Paai. ------------------------------ From: davidher@netcom.com (david her) Subject: [Q] capture echo & error ? Date: Sat, 8 Oct 1994 17:19:47 GMT How can I capture both echo & error while a script or program is running ? Sometimes I have to send the complete session of a job , but all I know is 1) redirect echo ex. make > myfile 2) redirect error ex. make 2> error then manually combine myfile+error and arrange error correspond to event inside of myfile. Is any easy way just capture all echo as stdout what I saw on terminal ? ------------------------------ From: tom@darshiva.efrei.fr (Thomas Parmelan) Subject: Re: Fidonet s/w for linux? Date: Fri, 7 Oct 1994 18:23:06 GMT Ray Blaak écrivait: >The subject line says it all. I am interested in hooking up to Fidonet. Is the >necessary software available for Linux? You can try ifmail. I'am running my Fidonet point with it. Current version is 2.5. There is also FidoGate (see the recent post in c.o.l.announce for more info on this one). -- Thomas Parmelan - tom@darshiva.efrei.fr ------------------------------ ** FOR YOUR REFERENCE ** The service address, to which questions about the list itself and requests to be added to or deleted from it should be directed, is: Internet: Linux-Misc-Request@NEWS-DIGESTS.MIT.EDU You can send mail to the entire list (and comp.os.linux.misc) via: Internet: Linux-Misc@NEWS-DIGESTS.MIT.EDU Linux may be obtained via one of these FTP sites: nic.funet.fi pub/OS/Linux tsx-11.mit.edu pub/linux sunsite.unc.edu pub/Linux End of Linux-Misc Digest ******************************