Linux-Misc Digest #867, Volume #19               Fri, 16 Apr 99 14:13:15 EDT

Contents:
  Re: Linux filesystem and powerdown ("Herwig Huener (Siemens)")
  Re: Linux filesystem and powerdown (Enkidu)
  Re: resolution in X (JS)
  EGCS compiler problem. - Redhat 5.9.7 (latest Starbuck release) ("mouse")
  Re: Linux is dead (Harry Lewis)
  Re: LILO problem: SOLVED ([EMAIL PROTECTED])
  Re: How to print an envelope ? (Edwin Johnson)
  Rockwell ADPCM to .wav conversion (Joshua Grauman)
  Re: 'Doze 98 vs. UNIX multitasking (Matthias Warkus)
  Re: S3 Trio 3D SVGA support????? (Gunther Schleinkofer)
  Re: When is RedHat 5.3 coming?? (Tapio Riikonen)
  Re: Linux is dead (Harry Lewis)
  Semaphore limits. (Jerald Jackson)
  Re: When is RedHat 5.3 coming?? ([EMAIL PROTECTED])
  Waiting for arbitrary process exit ([EMAIL PROTECTED])
  Windows NTDETECT ("Justyn Bridge")
  Re: VMWare speed (mist)
  Printer-problem (Sverre Andreas Larssen)
  Re: Color problem when Running Cadence on Linux (Martin Mandl)
  Re: Install ? again. (brian moore)
  Re: Anyone play Q2 from their linux box? (Stew Benedict)
  Re: need help selecting a college for C / Unix (Duncan Simpson)

----------------------------------------------------------------------------

From: "Herwig Huener (Siemens)" <[EMAIL PROTECTED]>
Subject: Re: Linux filesystem and powerdown
Date: Fri, 16 Apr 1999 09:17:33 +0200
Reply-To: [EMAIL PROTECTED]

This is a multi-part message in MIME format.
==============B831399276221CF5B4222DEF
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit

1999-04-16 09:02:00 MEST

Harddisks are so cheap nowadays - I do a frequent disk-to-disk backup
(The magic word is "cp -a ..."), and now and then a recursive diff. This
lowers the probability for a full loss of some or all data to once in a
few
hunderd years or so. Best of all: practically no extra effort for the
backup, such as inserting a backup-medium.

Now: the residual probability for data-loss is dominated by infrequent
events such as an axe driven into the computer, fire, malevolent
hard-ware
or an air-raid. Have another computer in another room, connect it with
NFS and do the "cp -a" trick now and then. Then, the probability for a
full data-loss by pure technical reasons goes down to once in a few
million years.

Other influnences (intentional sabotage, system-administrator's error,
etc)
will cause a much more frequent data-loss.

HH
==============B831399276221CF5B4222DEF
Content-Type: text/x-vcard; charset=us-ascii; name="vcard.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Herwig Huener
Content-Disposition: attachment; filename="vcard.vcf"

begin:          vcard
fn:             Herwig Huener
n:              Huener;Herwig
email;internet: [EMAIL PROTECTED]
x-mozilla-cpt:  ;0
x-mozilla-html: FALSE
version:        2.1
end:            vcard


==============B831399276221CF5B4222DEF==


------------------------------

From: Enkidu <[EMAIL PROTECTED]>
Subject: Re: Linux filesystem and powerdown
Date: Fri, 16 Apr 1999 19:13:43 +1200
Reply-To: [EMAIL PROTECTED]

Jahn Otto Andersen wrote:
> 
> My questions are:
> 
> - Is there any alternative filesystem for Linux that handles
> powerdown better ?
> - What can be done to minimize these problems ? (An UPS would
> of course be nice, but what else can be done ?).
> 
Do you mean normal shutdowns or power failures? If you mean normal
shutdowns, do you use the "shutdown" command? If you mean loss of
power, then with a UPS you could code some emergency shutdown
routines.

Cliff

-- 
Cliff Pratt, CAP Consulting
Web build, web design. HTML, Javascript, CGI, ASP, Web Consulting
Email: [EMAIL PROTECTED]  Phone: 025 246 7747

------------------------------

From: JS <[EMAIL PROTECTED]>
Crossposted-To: 
alt.uu.comp.os.linux.questions,alt.linux,comp.os.linux.setup,alt.os.linux
Subject: Re: resolution in X
Date: Fri, 16 Apr 1999 18:56:20 +0200

Use <ctrl><alt> and + or -

Regards

Johnny


k wrote:

> Hi All,
>
> how do i change the monitor mode i am using in X windows?
>
> i seem to remember beign able to pass a command line argument to xinit
> to select from the allowable modes but i can't figure ti out.
>
> thanks,
>
> nick


------------------------------

From: "mouse" <[EMAIL PROTECTED]>
Subject: EGCS compiler problem. - Redhat 5.9.7 (latest Starbuck release)
Crossposted-To: linux.redhat.misc,alt.linux,alt.os.linux
Date: Fri, 16 Apr 1999 07:52:58 GMT

I'm trying to compile SSLeay, but its giving me an error...

===============
gcc -I../include -DL_ENDIAN -DTERMIO -DBN_ASM -O3 -fomit-frame-pointer
-m486 -Wall -Wuninitialized -DSHA1_ASM -DMD5_ASM -DRMD160_ASM   -c
destest.c -o destest.o
destest.c: In function `cfb_test':
destest.c:803: internal error--unrecognizable insn:
(insn/i 98 97 99 (set (reg/v 43)
        (plus:SI (reg:SI 42)
            (symbol_ref:SI ("bufs.76")))) -1 (insn_list 97 (nil))
    (expr_list:REG_DEAD (reg:SI 42)
        (nil)))
../../gcc/toplev.c:1367: Internal compiler error in function fatal_insn
make[1]: *** [destest.o] Error 1
make[1]: Leaving directory `/tmp/SSLeay-0.9.0a/test'
making tools...
make[1]: Entering directory `/tmp/SSLeay-0.9.0a/tools'
make[1]: Nothing to be done for `all'.
make[1]: Leaving directory `/tmp/SSLeay-0.9.0a/tools'
[root@box SSLeay-0.9.0a]# gcc -v
Reading specs from /usr/lib/gcc-lib/i386-redhat-linux/egcs-2.91.66/specs
gcc version egcs-2.91.66 19990314/Linux (egcs-1.1.2 release)

------------------------------

From: Harry Lewis <[EMAIL PROTECTED]>
Subject: Re: Linux is dead
Date: Fri, 16 Apr 1999 08:54:02 +0100
Reply-To: [EMAIL PROTECTED]

>>> If Microsoft ever published a piece of software that was
>>> worth the retail list price, it would be a flaming miracle.
>>
>> Internet Explorer? List price $0,000.00 (including local taxes)?
>
>Yes? But is it WORTH it?

Is Internet Explorer worth nothing? Is it worthless? Is that the
question? Or are you alluding to TCO? By emphasizing WORTH you make the
question sound so profound, know what I *mean*?

;o)

Harry


------------------------------

From: [EMAIL PROTECTED]
Crossposted-To: vmware.for-linux.installation
Subject: Re: LILO problem: SOLVED
Date: Fri, 16 Apr 1999 13:12:00 GMT

In article <[EMAIL PROTECTED]>,
  "G. Pollack" <[EMAIL PROTECTED]> wrote:
> I'm trying to set up vmware for Win98 as a guest OS, using the rawdisk
> method. Win98 is on /dev/hda1, and the MBR is on /dev/hda0. When I
> attemp to Power On, lilo writes just 'L' to the screen, and then I get
> the following message:
>
> Attempt to read 1 sector(s) starting at sector 6044173. This is outside
> the allowed range for raw disk '/home/jerry/vmware/win98/win98.hda'.
>
> Can anyone tell me what's wrong, and how to correct it?
>
> Thanks,
>
> --
> Gerald Pollack
> Dept. of Biology, McGill University
>

The problem was that in setting up vmware I did not give it any access to my
linux partition (thus presumably lilo was unable to read /etc/lilo.conf). I
changed this to give vmware read-only access, and now lilo works fine under
it.

============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own    

------------------------------

From: [EMAIL PROTECTED] (Edwin Johnson)
Subject: Re: How to print an envelope ?
Reply-To: [EMAIL PROTECTED]
Date: 16 Apr 1999 13:13:48 GMT

I use Lyx, not Klyx, but I simply set the margins for left and top to
small amounts, such as 1/4 in for the return address which in my case is a
graphic. Then you can set the next paragraph as centered with a defined
amount of space above it (such as an inch). By issuing ctrl-[enter] for the
line feeds it single spaces the address and centers it. If it is offset too
much left or right for your envelope size, change the paper size. Then save
the whole thing and use it as a template. 

I have a second file of centered addresses and simply paste from it to the
envelope template for my envelope, otherwize just type in the address. 
Works like a charm! :)

...Edwin

On 15 Apr 1999 15:49:12 GMT, David Pando <[EMAIL PROTECTED]> wrote:
>Is there any little utilitie for printing envelopes in a HP printer ?
>I�m very happy with Klyx, so I don�t want to install WP or StarOffice just
>to make adresses look good.
>Any suggestion ?
>Thanks in advance


-- 
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
~   Edwin [EMAIL PROTECTED]   ~
~        http://www.prysm.net/~elj        ~
~                                         ~
~ "Once you have flown, you will walk the ~
~ earth with your eyes turned skyward,    ~
~ for there you have been, there you long ~
~ to return." -- da Vinci                 ~
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~


------------------------------

From: Joshua Grauman <[EMAIL PROTECTED]>
Subject: Rockwell ADPCM to .wav conversion
Date: Fri, 16 Apr 1999 00:53:34 -0700

I know I've seen other posts on this before, but I haven't seen an
answer. Does anyone know if there is any code out there to convert
from Rockwell ADPCM to .wav (or whatever) for use with a Rockwell
compatible voice modem? Thanks. Please also reply to my email.

Joshua Grauman
[EMAIL PROTECTED]

------------------------------

From: [EMAIL PROTECTED] (Matthias Warkus)
Crossposted-To: comp.unix.bsd.freebsd.misc,comp.unix.misc
Subject: Re: 'Doze 98 vs. UNIX multitasking
Date: Fri, 16 Apr 1999 14:54:07 +0200
Reply-To: [EMAIL PROTECTED]

It was the Thu, 15 Apr 1999 17:10:31 +0100...
..and Csaba Raduly <[EMAIL PROTECTED]> wrote:
> QFD = quod fuit (?) demonstrandum = what needs to be proven
> QED = quod erat demonstrandum = what had to be proven ( but it is proven
> now )
> Depends on the context but from what I see, I'll say it's a typo.
> 
> Would please the native latin speakers stand up and correct me ?

"Quod fuit demonstrandum" is perfect; "quod erat demonstrandum" is
imperfect. Means essentially the same.

mawa
-- 
ACHTUNG! ALLES LOOKENSPEEPERS!
Das Net ist nicht f�r Fingerclicken und Giffengrabben. Ist easy
droppenpacket der Routers; ist nicht f�r Gewerken bei das Dummkopfen.
Das mausklicken Sichtseeren relaxen und watchen das Cursorblinken. 

------------------------------

From: [EMAIL PROTECTED] (Gunther Schleinkofer)
Subject: Re: S3 Trio 3D SVGA support?????
Reply-To: [EMAIL PROTECTED] (Gunther Schleinkofer)
Date: Fri, 16 Apr 1999 07:33:27 GMT


In a previous article, [EMAIL PROTECTED] ("John Cardoso") says:

>I have installed support for S3 Trio 3D but only runs Generic VGA and that
>is terrible. I only have 16 colors.
>Does anyone know of SVGA support for S3 Trio3D?
>
>Thanks
>-John
>

Check the commercial X servers; I think  X inside advertises support for 
this card --- but for the price you have to pay, it makes more sense 
getting a good graphics card. I had the same experience and decided on a 
ATI Xpert@Play with 8Mb; it works fine and cost me less than the for-pay 
X server.

------------------------------

From: [EMAIL PROTECTED] (Tapio Riikonen)
Subject: Re: When is RedHat 5.3 coming??
Date: Fri, 16 Apr 1999 08:02:02 GMT

On Thu, 15 Apr 1999 13:25:50 GMT, "William T. Trotter"
<[EMAIL PROTECTED]> wrote:


>There are problems, but most that I have seen
>involve buggy software.  For example, Netscape 4.5.1
>crashes repeatedly, especially whenever
>a separate Messenger window is opened.

Well, a major bug I've managed to run into is i740 support, XBF_i740
simply doesn't seem to work. 

Has anyone anywhere anytime succeeded in having i740 support work in
Starbuck 5.9.7?

Tapio

------------------------------

From: Harry Lewis <[EMAIL PROTECTED]>
Subject: Re: Linux is dead
Date: Fri, 16 Apr 1999 09:09:17 +0100
Reply-To: [EMAIL PROTECTED]

> > > If Microsoft ever published a piece of software that was
> > > worth the retail list price, it would be a flaming miracle.
> >
> > Internet Explorer? List price $0,000.00 (including local taxes)?
>
> It's not even worth that much.

You know, you're right. A piece of software that allows you to surf the
web, view image files in a large number of formats, play audio files,
use plug-ins, render HTML including all industry standard tags and the
majority of non-standard ones including those of its competitors and
interpret embedded Java scrips and run Java applets really isn't worth
anything. My god - what was I thinking? Prejudice alone is worthy. I'm
converted. Never again will I miss an opportunity to snipe and bad-mouth
Microsoft - in fact from now on I'll take to calling them M$, 'cos
that'll make Bill Gates sorry for all those tens of billions of dollars
he's amassed forcing crap, unreliable, unusable, worthless software that
not worth bothering with even if it's free.

Raj - you cured me. There I was being conned by the World Conspiracy
into making a good living from M$ crap and now I'll spend all my days
making nothing tinkering with Linux!

Harry


(BTW - please don't take this reply too seriously and start getting
worked up - it's just a send-up!)


------------------------------

From: Jerald Jackson <[EMAIL PROTECTED]>
Subject: Semaphore limits.
Date: Fri, 16 Apr 1999 06:28:03 -0700

How would I go about raising the max arrays limit as reported below?
Does it require a kernel rebuild or is there another method of tuning
these limits?

ipcs -al

====== Shared Memory Limits ========
max number of segments = 128
max seg size (kbytes) = 16384
max total shared memory (kbytes) = 16777216
min seg size (bytes) = 1

====== Semaphore Limits ========
max number of arrays = 384
max semaphores per array = 32
max semaphores system wide = 12288
max ops per semop call = 32
semaphore max value = 32767

====== Messages: Limits ========
max queues system wide = 128
max size of message (bytes) = 4056
default max size of queue (bytes) = 16384

Regards,

Jerald Jackson


------------------------------

From: [EMAIL PROTECTED]
Subject: Re: When is RedHat 5.3 coming??
Date: Fri, 16 Apr 1999 13:32:50 GMT

In article <[EMAIL PROTECTED]>,
  Al Dev <[EMAIL PROTECTED]> wrote:
> I heard RedHat would be releasing RH 5.3 during 1st week of april.
> Linux with new kernel version 2.2 (SMP).
>
> It had been a long time since RedHat 5.2 is released. Anyone has idea
> when it will be released??
>
>

Umm, Redhat already shipped a 5.9 cd with the 2.2 kernel on it...  They will
have a 6.0 very soon.

Eric Harrison

============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own    

------------------------------

From: [EMAIL PROTECTED]
Crossposted-To: alt.os.linux,alt.linux
Subject: Waiting for arbitrary process exit
Date: Fri, 16 Apr 1999 16:24:55 GMT

Does anyone know of a way to emulate the Solaris /usr/proc/bin/pwait command
in Linux? (Or any other OS that doesn't have pwait)

(This command will wait for the exit of any process in the system, not
necessarily children.)

--
Brian Morris
mailto:[EMAIL PROTECTED]
Freshman, Computer Engineering, CalPoly, San Luis Obispo, CA

============= Posted via Deja News, The Discussion Network ============
http://www.dejanews.com/       Search, Read, Discuss, or Start Your Own    

------------------------------

From: "Justyn Bridge" <[EMAIL PROTECTED]>
Crossposted-To: microsoft.public.windowsnt.misc,microsoft.public.windowsnt.setup
Subject: Windows NTDETECT
Date: Fri, 16 Apr 1999 14:33:16 +0100

Have got myself in quite a sticky situation. This quite a long story but
feel if anyone can help i will have to put in as much info as i can so
please bare with me.

I have got Win95b and WinNT 4 WS station installed om my PC. A few weeks ago
my PC started booting straight into 95 with out giving me the WinNT NTLoader
menu. It didn't bother me much at the time as i didn't have any use for NT.
However i have now created a network between 2 PC's and would like to be
able to WinNT again. I have always wanted Linux on my PC and from what i had
read up about it, i could get 95, NT and Linux to boot up from the Linux
menu. So i installed Redhat 5.2 and tried to get Lilo ( Linux command
promt ) to see my NT boot sector. Uhh,uhh - no way. When trying to boot into
NT from Lilo i got tis error message ( A kernal file is missing from the
disk. Insert a system diskette and restart the system ) So that didn't work
although i could get to 95 from Lilo. Hokay, not so bad, more or less the
same situation i was in before installing Linux. So i decided to use the NT
repair disk option and go through the repair, During the repair i get this
huge message explaining that  setup has determined that one or more of your
hard disks has more than 1024 cylinders. ( Not to sure about my cylinders
side of things - however at the end of this message it states "This message
does not necessarily indicate and error condition. It is intended to alert
you to the fact that one or more of your hard disks may actually be larger
than the size for which it is currently configured.

Right, so i read up on this and find that a DOS utility called Bootpart can
repair my WinNT  boot sector - nice one! Try to repair it - all seems okay,
cross my fingers and then reboot the PC hoping that NT menu will appear or i
will beable to get to it from Lilo.l What happens, I get an error message
NTDETECT Failed and the PC hangs. So now no 95 no NT but linux is ok.
However when using the 95 startup disk, i can see al my files no prob and
using NTFSDOS  i can see all my NT files. I tried to boot 95 from DOS and
got this error message." Registry file was not found. Registry services may
be inoperative for this session. XMS cache problem. Registry service may be
inoperative. A device or resource required by VFAT is not present or is
unavailable. VFAT cannot continue loading. System halted".

Pretty screwed huh!!! I am going to try and replace NTDTECT.COM this weekend
and NTLDR.  But it's looking pretty grim for any recovery. I appreciate
anyone reading this huge posting and if you have the slightest hint/tip or
any info - i will be most greatful.

Cheers






------------------------------

From: mist <[EMAIL PROTECTED]>
Subject: Re: VMWare speed
Date: Fri, 16 Apr 1999 17:15:33 +0100
Reply-To: mist <new$[EMAIL PROTECTED]>

[EMAIL PROTECTED] scribed to us that -
>Just wanted to know if anyone has any spex as to how fast vmware runs? For
>instance I have an AMD K62 300Mhz machine with 64 mb RAM and I just wanted to
>know if programs will run fairly equivalent to just running one OS.  I know
>that it will run a little slower, but how slow?
>

My Pee 2 350 with 64 Meg of ram runs the virtual Windows 95 at a
tolerably fast speed, not much more slowly than if I was running Windows
itself.  The only noticeable speed difference was with screen refreshes,
but I was not using the vmware x server.  

It works quite well, give it a try.  The only reason I stopped using it
was because I didn't want to become dependant on it and then find that
the license stops being free.

-- 
Mist.

------------------------------

From: Sverre Andreas Larssen <[EMAIL PROTECTED]>
Subject: Printer-problem
Date: Fri, 16 Apr 1999 19:57:06 +0200

I have a Canon bjc 620 which worked fine until I compiled the kernel
(RH5.1-2.0.35)
After the new kernel was installed linux won't talk to lp1 anymore :(

If I try to install a printer under the control-panel, the response is
lp0,lp1 and lp2
not detected. lp1 is the par.port to the printer.

Any suggestions on how to solve this ?

--
Email: [EMAIL PROTECTED] Voice: +47 37272154 (rarely)
Institutt for informatikk         Universitetet i Oslo




------------------------------

From: Martin Mandl <[EMAIL PROTECTED]>
Crossposted-To: comp.cad.cadence,comp.os.linux.setup
Subject: Re: Color problem when Running Cadence on Linux
Date: Fri, 16 Apr 1999 11:20:02 +0200

Hi Jacky, 

close netscape and others, make a color reset, start cadence, ... it
works (at least for me)

the problem is an adjustable color map, the first program started wins,
the others have to live with it ...

but i was told, new versions of cadence and allegro can handle now also
more colors (i've never tried ...)

cheers

        Martin

Jacky wrote:
> 
> Hi. I try to run cadence from Linux system. But I only can set the color
> to 8b. otherwise my application - Cadence will tell you it can't handle
> 16/32 planes. But if I run it w/ 8b colors, the colors are quite
> different from what they look like when I run it on a HP-Unix
> workstation. so I can't tell some of the colors at all.
> So, is there some way to change the color of linux? And also what's the
> problem here?
> 
> Thanks!
> 
> Jacky
> (Pls reply to me.)

------------------------------

From: [EMAIL PROTECTED] (brian moore)
Crossposted-To: comp.os.linux.setup,comp.os.linux,alt.os.linux
Subject: Re: Install ? again.
Date: 16 Apr 1999 17:47:18 GMT

On Fri, 16 Apr 1999 19:15:28 +0200, 
 Matthias Warkus <[EMAIL PROTECTED]> wrote:
> It was the Fri, 16 Apr 1999 10:14:40 -0600...
> ..and Bud <[EMAIL PROTECTED]> wrote:
> > >I read somewhere that your swap space should match your actual >RAM.
> > >Hence I have 64meg of ram, and a 64 meg swap file.
> > 
> > Hmmm, I thought it was twice your RAM.
> 
> That was a valid rule in the 1970s, where hard drive space was about
> half as expensive as RAM and having 1/3 live RAM and 2/3 swap was the
> cheapest option.

Actually, it was a valid rule in the '80's, too, where systems like
SunOS had to have at least as much swap as RAM -- if you had 32M of RAM
and 10M of swap, it would only use 10M of RAM.

So on SunOS4 systems, you'd need at 1xRAM as swap, and because the whole
point of virtual memory is that you can swap, 1.5xRAM was a reasonable
number.  (If you needed 2X, that was a sign that your machine was
swapping too much and you should up the RAM.)

On newer systems (like Linux and Solaris), virtual memory = RAM + swap,
so the computation is somewhat different, since you don't need to count
the size of the memory into the picture.

> Now, there aren't any rules any more, get the amount of RAM you can
> afford and add as much swap as you need.

And there's still the problem of 'too much' swap: just like the '10%'
free space that the file system tries to keep around for slop, the
swap size usually doesn't scale well: if you have a gig of RAM on a
machine, using a gig of swap would still mean your machine was
thrashing like hell.

For most people, 64M of swap is plenty.  If you're manipulating large
data sets (and hopefully your software is written well enough to
localize much of the data, so that unused data will page out and stay
paged out until its needed), that number can be raised much higher to
good effect.  The catch is that a lot of programs don't do the necessary
work to localize things, and end up thrashing as they page in and out.
All the swap space in the world won't fix that, but RAM will.

-- 
Brian Moore                       | "The Zen nature of a spammer resembles
      Sysadmin, C/Perl Hacker     |  a cockroach, except that the cockroach
      Usenet Vandal               |  is higher up on the evolutionary chain."
      Netscum, Bane of Elves.                 Peter Olson, Delphi Postmaster

------------------------------

From: [EMAIL PROTECTED] (Stew Benedict)
Crossposted-To: comp.os.linux.development.apps
Subject: Re: Anyone play Q2 from their linux box?
Date: 16 Apr 1999 16:13:52 GMT
Reply-To: stewb AT earthlink DOT net

Actually if you run X in 320x200, you can get full screen.  I run a
seperate X server just for Q2.

Stew

On 15 Apr 99 23:14:49 GMT, Daniel Robert Franklin <[EMAIL PROTECTED]> wrote:
>"Sabbat" <[EMAIL PROTECTED]> writes:
>
>>For what I tried its ok but you cant run it under X of course.
>>And I didnt find a tool like gamespy for linux yet...for console mode it can
>>be funny :)
>
>Of course Quake2 works in X (but full screen console mode is much nicer
>:). Read the accompanying documentation.
>
>I found Q2 to be extremely playable under Linux, at least as fast as
>Windows (I don't have a 3DFX card yet but I know they are supported too).
>
>- Daniel
>--
>******************************************************************************
>*       Daniel Franklin - Postgraduate student in Electrical Engineering
>*       [EMAIL PROTECTED]
>******************************************************************************


-- 


------------------------------

From: [EMAIL PROTECTED] (Duncan Simpson)
Crossposted-To: alt.os.linux
Subject: Re: need help selecting a college for C / Unix
Date: 16 Apr 1999 18:13:32 GMT

In <7esttl$ldc$[EMAIL PROTECTED]> [EMAIL PROTECTED] writes:

>In the sacred domain of comp.os.linux.misc didst Eric <[EMAIL PROTECTED]> 
>eloquently scribe:
>: Hell, screw a good school.  You got all your lowers yet?  If not, hit
>: any community college and take all of the CS stuff.  Unless you like
>: math though, stick with General studies, then you can take what you
>: want.  If you take a real CS degree you will be roped into taking crap
>: like discrete math (ten years working in IT, and I have yet to see the
>: point of discrete math), 

>LOL!
>Couldn't agree more.
>(although predecate logic did rear it's ugly head again during the Prolog
>and Compilers parts of the course here)...

>: calculus and other useless stuff.  

>CALCULUS? Nahhh... We touched VERY briefly on Fourier Transforms in Digital
>Image Processing this year, but if you avoid modules like that, you need
>never hear the words integration, differeciation and fourier.


Just recently A level, which means low level UG in the US AFAIK,
standard matrix algebra proved to be invaluable in a quantum
cryptography problem (research I did at Los Alamos National
Lab). Understanding linear indepence and basic column operations was
absolutely required. I can reasonably assume any CS person will
understand the results. (I have also actually used Lapace transforms
for real research).

>You just have to look closely at course outlines.`

And if you go to Oxford, and many other places, you will learn about
calculii that never talk about differentiantion or intergration. This
is stuff like CSP, CCS, and deriving programs from
specifications. Related stuff will also appear like Hoare logic. Use
of all these techniques without understandiung maths will be hard.

Maybe people who want to do CS and avoid maths should try reading a
few CS papers, Knuth and textbooks (such as Garey and Johnson's
NP-completeness bible). Throw in a tad of advanced statistics for
analysis of experimental results when you can not do the maths.

As for OO, JAVA and sutff like that many people with genuine CS
degrees do not swallow that stuff wholesale. Learning a new
intperative, or for that matter functional, programming language is
not a major challenge. If this is not the case for you then you
probably did the wrong degree. JAVA and C++ are just imperative
languages, enough said.

Of course if you want to be a real programmer then CS may be the wrong
subject. The desirability of real programmers is soemthing I decline
to make commments about.

Duncan (-:
Who has a 2 1 (Maths and CS, Oxford), MSc (Parallel Compuation,
Warwick) and hopefully a PhD (CS, Southampton) by January 2000
(theoretically September 1999, but I suspect changes will be
required).
--
Duncan (-:
"software industry, the: unique industry where selling substandard goods is
legal and you can charge extra for fixing the problems."

------------------------------


** 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: [EMAIL PROTECTED]

You can send mail to the entire list (and comp.os.linux.misc) via:

    Internet: [EMAIL PROTECTED]

Linux may be obtained via one of these FTP sites:
    ftp.funet.fi                                pub/Linux
    tsx-11.mit.edu                              pub/linux
    sunsite.unc.edu                             pub/Linux

End of Linux-Misc Digest
******************************

Reply via email to