Linux-Hardware Digest #228, Volume #11 Sat, 11 Sep 99 01:13:23 EDT
Contents:
wierd bug: NS communicator crashes sending mail (Chris Abajian)
Re: AMD K6-3 + FX PA-2013 SIG 11 problems ([EMAIL PROTECTED])
Re: AMD K6-3 + FX PA-2013 SIG 11 problems (Mike Frisch)
Re: Scanner for Linux (David Cooley)
Re: Imation Superdisk ("Greg Lucas")
Re: lpr fails with HP Deskjet 612C ("Omar Bradleyhower")
Drivers for Diamond Stealth III s540? ("Wayne O. Jackson")
Re: Problem With TNT2 Video Card Please Help (Luckydaze1)
Re: UDMA problem? (David Cooley)
Re: UDMA problem? (David Cooley)
Re: Permission Denied: Please Help (David Cooley)
Re: AMD K6-3 + FX PA-2013 SIG 11 problems (Bryan)
Re: AMD K6-3 + FX PA-2013 SIG 11 problems (Bryan)
Re: 13GB Hard Drive problems -- appears much smaller... (Chris)
Re: Dual Celeron (Bryan)
Re: AMD K6-3 + FX PA-2013 SIG 11 problems (Rod Roark)
Re: Sound Blaster PCI 64 not good... ("Joon Nan")
Re: Miro PCTV Card (Bttv Card) (Peter T. Breuer)
----------------------------------------------------------------------------
From: Chris Abajian <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup,comp.os.linux.help
Subject: wierd bug: NS communicator crashes sending mail
Date: Fri, 10 Sep 1999 22:01:59 +0000
I am at a total loss. I've been struggling with this for weeks.
Ever since I got my DSL connection, netscape communciator mail is broken.
specifically, when I hit the send mail button the app hangs, memory grows
without bound till it brings my system to it's knees. No amount of messing
with the mail host configurations has fixed it to date.
I'm running RH linux with a 2.0.36 kernel, ns communicator 4.61
What I'd LIKE to do is use my localhost as smpt server.
I installed qmail, it works great. I can send messages with pine
or command line mail.
I'm running a cacheing-only named.
My domains are finally straightened out. My address resolves correctly.
I have several names for the system in etc hosts, and my ISP has those
in the ns records as well. nslookup works exactly as expected from here
and afar...
I reinstalled NS 4.61. Not that I expected it to do anything, it's just that
it's the first thing you do on NT ;-)
I renamed my .netscape directory & configured everything from scratch.
I can't stand it anymore. I'm teaching the wife and kid to use pine.
What is up with this? Any suggestions?
--
Chris Abajian
http://www.abajian.com
------------------------------
From: [EMAIL PROTECTED]
Subject: Re: AMD K6-3 + FX PA-2013 SIG 11 problems
Date: 10 Sep 1999 22:12:11 GMT
Rod Roark <[EMAIL PROTECTED]> wrote:
> Brady Montz <[EMAIL PROTECTED]> wrote:
>>...
>>So, I buy myself a AMD K6-3 400Mhz CPU, FIC PA-2013 MB (with 2MB cache - I
>>know many say the 1MB is just as good, but they didn't have it locally so oh
>>well), and a 64MB PC 100 DIMM. slap it all together, boot linux, and
>>everything is happy until I compile the kernel and boom, SIG 11....
> This is usually the CPU.
No, it's not, it's usually some broken RAM-chips or too high busspeeds,
or wrong cpu-voltage or too bad cooling or instability of the motherboard.
There are manymany reasons...
All AMD cpu's I had from DX4-133 to k6-iii 400 had even been overclockable
at least 25Mhz. So i can't agree that there should be a stabilityproblem
with those CPU's.
> Don't home-build a system with an AMD CPU.
AMD CPUs are still the best upgrade-option for many systems.
> About 10-20% of the AMD systems I build fail the overnight multiple
> kernel compilations test. This seems to be across the whole range of
> K6-2 and K6-3 CPUs.
Hmm...then you better don't sell any of them ;-)
Greetinx,
fIPS
------------------------------
From: [EMAIL PROTECTED] (Mike Frisch)
Subject: Re: AMD K6-3 + FX PA-2013 SIG 11 problems
Date: 11 Sep 1999 00:06:54 GMT
On 10 Sep 1999 22:42:10 GMT, Rod Roark <[EMAIL PROTECTED]> wrote:
>Define "rock solid stable". Just because it doesn't crash running
>whatever you happen to want to run doesn't mean it's stable. Do you
>really test every one? If so, how exactly? If not, you better start
>testing.
"Rock solid stable" is exactly how it reads. I'm no Johnny Comelately to
the PC world. Having been building PCs for more than 15 years now, I like
to think I know what I'm doing, and based on my clients' feedback and
reliability statistics, I think that says it all.
>333's at the price of the 300's; he refused, saying "that's what you
>should expect, if you want reliability go with Intel".
Sure...
>By the way I have nothing to gain from bashing AMD, and would much
>prefer that Intel have better competition. I'm just stating my
>observations as one who ought to know and who cares about quality.
I'll continue building reliable PCs based on AMD CPUs and whatever you do
is your choice. Just stop spreading FUD.
Mike.
--
======================================================================
Mike Frisch Email: [EMAIL PROTECTED]
Northstar Technologies WWW: http://saturn.tlug.org/~mfrisch
Newmarket, Ontario, CANADA
======================================================================
------------------------------
From: David Cooley <[EMAIL PROTECTED]>
Subject: Re: Scanner for Linux
Date: Fri, 10 Sep 1999 22:12:49 -0400
"Fabrice P. Laussy" wrote:
>
> Hi
>
> I'd like some advice from peoples who have experience with running a
> scanner for Linux. Which of them are supported, with average to good
> results, for low to average price.
>
> Or are there FAQs or something that could help?
There are FAQ's on it, and the drivers you need are from the SANE
project...
Do a search on the web for SANE project and LINUX and you'll be set.
------------------------------
From: "Greg Lucas" <[EMAIL PROTECTED]>
Subject: Re: Imation Superdisk
Date: Fri, 10 Sep 1999 21:13:12 -0500
I just installed 5.2 and have the same problem. My LS 120 works fine from
bios, but not available after Linux boots. LS 120 continues to function
normally under Windows 98.
Greg
------------------------------
From: "Omar Bradleyhower" <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup,comp.os.linux.help
Subject: Re: lpr fails with HP Deskjet 612C
Date: Fri, 10 Sep 1999 18:39:09 -0400
OK, this *may* not be corrrect, but I *BET* it is...
I've got an HP712C. With my Win98 computer, I thought I was getting a very
nice printer at a reasonable price.
WRONG.
My HP712 uses the PPA printer language/protocol. It is the printer
equivalent of a WINMODEM - in order to make the printer cheaper, they
offloaded much of the printer function to Windows & the CPU. So if you're
not using Windows, you are out of luck.
Just to make sure, you can visit www.hp.com, look through the website, find
the specs for your model. If PPA is listed as a "feature," your luck is as
bad as mine...
LOL
Wei-shi Tsai <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Hello. I have a HP Deskjet 612C connected to LPT1. However, I cannot
print from any program whatsoever.
>
> Everytime I attempt to print, I get some error message (and no printout).
When I attempted to print with Wordperfect 8, it complains of a spooling
error. When I attempt to print in Acrobat Reader 4, I get the message that
lpd cannot find the offical name of the local machine.
>
> I have the printer and port support compiled into the kernel. I am using
Slackware 4.0 with kernel version 2.2.10.
>
> Thanks for any help.
>
> --
> Wei-shi Tsai
> e-mail: [EMAIL PROTECTED]
> MoonieCode(1.8.11)
> SM:6-[8]m+ F:sMe++[+]>Mo+>:vNe+>Bl+<:aLu+[+]>Ry<:pR+>Cl< D:s:Ma<:vJa--
>
------------------------------
From: "Wayne O. Jackson" <[EMAIL PROTECTED]>
Subject: Drivers for Diamond Stealth III s540?
Date: Fri, 10 Sep 1999 21:55:41 -0400
Howdy,
I just got a Diamond Stealth III S540 video card, and it does a fine job
in Win98, but I really messed up because it's not yet supported in
Xfree86. I have checked the Internet in every nook and cranny and can't
find a driver that will work for this card. It has the S3 Savage4 Pro+
chipset. There is support for the Savage4 in the XF86_SVGA server, but
not Diamond's Savage4 Pro+. I've been able to find a lot of different
drivers for the Savage4 card, but they won't work with the Diamond card.
Does anybody have an idea when this card will be supported.
Thanks,
Wayne
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Luckydaze1)
Subject: Re: Problem With TNT2 Video Card Please Help
Date: 11 Sep 1999 02:01:00 GMT
>Subject: Re: Problem With TNT2 Video Card Please Help
>From: John Soltow [EMAIL PROTECTED]
>Date: Fri, 10 September 1999 09:18 PM EDT
>Message-id: <[EMAIL PROTECTED]>
>
>Luckydaze1 wrote:
>
>> I get a garbled screen when I choose RivaTNT as the chip from the list. The
>> screen goes screwy when Xconfigurator tries it out. I then bypassed
>> Xconfigurator and tried configuring it manually with XF86config but I get
>the
>> same results. Has anyone got their TNT2 card working and what setup did you
>> use?
>> Any suggestions please? I'm using mandrake 6.0
>
> I have RH 6.0 (Mandrake is supposedly RedHat with enhancements from what
>I hear) running with a Diamond TNT2/Ultra using the drivers from
>www.nvidia.com Just download the RIVA-X-GLX-1.0-glibc-i386-dyn.tar.gz file
>from their site, un-tar it and follow the instructions.
>
> John
>
I have the file but I don't know how to untar it. I wish someone would tell me
what diectory to place it in and all the steps to untar and install the file. I
tried but get "tar" [child} errors and permission denied errors. Can someone
help me out please cause this is driving me nuts.
------------------------------
From: David Cooley <[EMAIL PROTECTED]>
Subject: Re: UDMA problem?
Date: Fri, 10 Sep 1999 22:18:03 -0400
Dale Pontius wrote:
> No mention of UDMA, either in dmesg or from hdparm.
>
> Sounds like I've got some untapped performance to go after one
> of these days, when I get spare time. Goody.
What kernel are you using?
For the 2.0.x kernels there is a patch that enables UDMA.
for 2.2.X the patch is built in.
------------------------------
From: David Cooley <[EMAIL PROTECTED]>
Subject: Re: UDMA problem?
Date: Fri, 10 Sep 1999 22:18:53 -0400
Dale Pontius wrote:
>
> In article <[EMAIL PROTECTED]>,
> David Cooley <[EMAIL PROTECTED]> writes:
> >
> > Check DMESG. SHould see the drives listed the 2nd time with C/H/S and
> > UDMA at the end if it's activated... You can also use hdparm to see if
> > UDMA is on
>
> No mention of UDMA, either in dmesg or from hdparm.
Also, When you compile your kernel make sure the proper EIDE/PCI-Bridge
drivers are selected under menuconfig or xconfig.
------------------------------
From: David Cooley <[EMAIL PROTECTED]>
Subject: Re: Permission Denied: Please Help
Date: Fri, 10 Sep 1999 22:15:41 -0400
Luckydaze1 wrote:
>
> I'm trying to install a file. When I try to execute it I get a permission
> denied error.
> I'm new to Linux so can you please give me a step by step process to eliminate
> this problem? I'm using Mandrake 6.0 Thank you.
Make sure you are in the directory the file is in, and iff the file is
called foo.bar, type the following:
chmod 777 foo.bar
The 777 is permissions and they are cumulative... The 1st 7 is the root
permission... Read is 4, Write is 2, Execute is 1... give it Read +
write + execute and that's 7.
the middle number is Group permissions, and the last is World
permissions
------------------------------
From: Bryan <Bryan@[EMAIL PROTECTED]>
Subject: Re: AMD K6-3 + FX PA-2013 SIG 11 problems
Date: Sat, 11 Sep 1999 03:22:26 GMT
Rod Roark <[EMAIL PROTECTED]> wrote:
: Brady Montz <[EMAIL PROTECTED]> wrote:
: >...
: >So, I buy myself a AMD K6-3 400Mhz CPU, FIC PA-2013 MB (with 2MB cache - I
: >know many say the 1MB is just as good, but they didn't have it locally so oh
: >well), and a 64MB PC 100 DIMM. slap it all together, boot linux, and
: >everything is happy until I compile the kernel and boom, SIG 11....
: This is usually the CPU. Don't home-build a system with an AMD CPU.
: About 10-20% of the AMD systems I build fail the overnight multiple
: kernel compilations test. This seems to be across the whole range of
: K6-2 and K6-3 CPUs.
its your ram, I'm willing to bet. it always is ;-)
there are SO many grey gradients in ram quality and in what
assumptions the mobo makers make in terms of tolerances of timing.
sometimes I wonder why PCs work at all! ;-)
I usually reset the bios to factory defaults (NEVER assume they are
from the box) and go from there. and NEVER skimp on ram quality. I
buy mine locally from a reputable store. I pay more, but I've never
had a bad stick of ram. and I've built 3 AMD k6 systems and probably
about 10-15 intel systems.
--
Bryan, http://www.Grateful.Net - Linux/Web-based Network Management
->->-> to email me, you must hunt the WUMPUS and kill it.
------------------------------
From: Bryan <Bryan@[EMAIL PROTECTED]>
Subject: Re: AMD K6-3 + FX PA-2013 SIG 11 problems
Date: Sat, 11 Sep 1999 03:24:27 GMT
Mike Frisch <[EMAIL PROTECTED]> wrote:
: On 10 Sep 1999 20:14:19 GMT, Rod Roark <[EMAIL PROTECTED]> wrote:
: >This is usually the CPU. Don't home-build a system with an AMD CPU.
: >About 10-20% of the AMD systems I build fail the overnight multiple
: >kernel compilations test. This seems to be across the whole range of
: >K6-2 and K6-3 CPUs.
: It's definitely not the CPU! Did you ever think it was inferior cooling,
: a bad motherboard, or RAM? I've build a dozen AMD K6-2 machines and every
: one of them has been rock solid stable under Linux (and any other OS, for
: that matter).
: Cheap motherboards and RAM don't mix with AMD's CPUs, but who's stupid
: enough to cut corners on two of the most important parts of a stable PC?
we seem to share the same sentiments.
I don't go mailorder for mobo's cpus (unless boxed and sealed) or ram.
perhaps someone overclocked or overfried a mobo and returned it.
mailorder being the margin-sensitive thing it is, could possibly be
getting B stocks. good, reputable stores (there are a FEW still left)
will never sell returned or B stock as new.
also there was a rash of k6-2 chips that were 66mhz ONLY. if you ran
THOSE at 100, you'd certainly have trouble.
--
Bryan, http://www.Grateful.Net - Linux/Web-based Network Management
->->-> to email me, you must hunt the WUMPUS and kill it.
------------------------------
From: [EMAIL PROTECTED] (Chris)
Subject: Re: 13GB Hard Drive problems -- appears much smaller...
Date: Sat, 11 Sep 1999 00:08:35 GMT
On Fri, 10 Sep 1999 16:57:30 GMT, "William B. Cattell"
<[EMAIL PROTECTED]> wrote in comp.os.linux.hardware:
>> I purchased a 13GB Western Digital hard drive and partitioned it for Linux
>> I upgraded to Slackware 4.0 and then put in a 2.2.10 kernel. It took me a
>> while, but I finally noticed that I'm missing part of my hard drive.
>> Disk /dev/hda: 255 heads, 63 sectors, 1582 cylinders
255h * 63s * 1582c * 512 bytes = 13,012,392,960.
Manufacturers consider 13,000,000,000 bytes to be 13Gb because it allows
them to legally overstate the size of the drive for advertising purposes.
In computer terms, however, each G is 1024 Mb, each M is 1024 kb, and each
k is 1024 bytes. This means that a Gigabyte is really 1,073,741,824
bytes, and what a manufacturer calls a "13G" drive is really only
12.1Gbytes in size.
>> Units = cylinders of 16065 * 512 bytes
>> Device Boot Start End Blocks Id System
>> /dev/hda1 1 1000 8032468+ 83 Linux native
>> /dev/hda2 1001 1500 4016250 83 Linux native
>> /dev/hda3 1501 1582 658665 82 Linux swap
>>
>> Where's the rest of me? Or my disk?
(8,032,468 + 4,016,250 + 658,665) * 1024 = 13,012,392,960.
All bytes present and accounted for!
>> (And, YES, I know that's a stupid size for a swap partition. My brain is
>> the size of a turnip, but not as intelligent.)
I can't argue about the swap size. Different people have different needs.
I would suggest that if you really need a huge swap partition it should be
located near the middle of the drive for performance reasons (to minimize
head travel).
>Disk mfr's always tell you the total capacity of a drive if there were no
>'OS overhead'. Their documentation always includes a caveat to the effect
>that "based on the OS used you may have less diskspace than advertised."
That warning normally refers to "slack", which is the amount of space
wasted when a small file is written to a large allocation unit. It has
nothing to do with the fact that a "13G" drive is really only 12.1
gigabytes in size.
------------------------------
From: Bryan <Bryan@[EMAIL PROTECTED]>
Subject: Re: Dual Celeron
Date: Sat, 11 Sep 1999 03:19:31 GMT
Kevin <[EMAIL PROTECTED]> wrote:
: How do I go about testing the stability ?
seti@home is a grand test. a real torture test for cpus (heat, ram,
bios settings, etc).
: Kevin
: On 9 Sep 1999 23:32:16 GMT, [EMAIL PROTECTED] (David Ripton) wrote:
: >In article <[EMAIL PROTECTED]>,
: >Kevin <[EMAIL PROTECTED]> wrote:
: >>Bear in mind that if you are told that 80% of all celerons will
: >>overclock, you are fitting two of them, and so the likelihood of
: >>success will be 64%.
: >
: >That analysis makes two implicit assumptions.
: >
: >First, it assumes that each processor's chance of overclockability
: >is independent. But processors from the same batch tend to be
: >equally overclockable, and if you buy two together the odds are
: >that they will be from the same batch. If you can see the serial
: >numbers before you buy, you can guarantee this.
: >
: >Second, it assumes that SMP itself does not have an effect on
: >overclockability. This is a controversial statement.
: >
: >As an aside, I think that most overclockers do a lousy job of
: >stability testing, and that most of the overclocking success
: >rates we see posted are exaggerated. Not that people are
: >intentionally lying; they just don't test hard enough. Most
: >people run buggy software on an unstable OS, have come to
: >expect routine crashes, and just don't notice an occasional extra
: >crash due to slightly unstable hardware.
--
Bryan, http://www.Grateful.Net - Linux/Web-based Network Management
->->-> to email me, you must hunt the WUMPUS and kill it.
------------------------------
From: Rod Roark <[EMAIL PROTECTED]>
Subject: Re: AMD K6-3 + FX PA-2013 SIG 11 problems
Date: 11 Sep 1999 03:24:16 GMT
Mike Frisch <[EMAIL PROTECTED]> wrote:
>...
>I'll continue building reliable PCs based on AMD CPUs and whatever you do
>is your choice. Just stop spreading FUD.
I'll do the same, but continuing to test and send back the bad CPUs.
And just to bring this matter to a close, Brady (the original poster)
reports that he replaced the CPU and now everything works perfectly.
Oops sorry, I guess that's FUD.
Have a nice weekend.
-- Rod
======================================================================
Sunset Systems Preconfigured Linux Computers
http://www.sunsetsystems.com/ and Custom Software
======================================================================
------------------------------
From: "Joon Nan" <[EMAIL PROTECTED]>
Subject: Re: Sound Blaster PCI 64 not good...
Date: Sat, 11 Sep 1999 11:47:42 +0800
I'm using SB PCI 64 under RedHat 6.0.
There is no problem for installation.
Falk Henrich <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Am Thu, 09 Sep 1999 hat Gregory Stoll geschrieben:
> >Hi! I'm using a Sound Blaster PCI 64 under Linux, and am trying
to...make
> >it work!
>
> I am trying to get the SB PCI 64 to work under Debian 2.1. During boot up
the
> following messages appear:
>
> Sound initialization started.
> Sound initialization complete.
>
> I believe the kernel does not find the card. What kernel settings should I
use?
>
> Thanks in advance.
>
> -Falk
------------------------------
From: [EMAIL PROTECTED] (Peter T. Breuer)
Subject: Re: Miro PCTV Card (Bttv Card)
Date: 11 Sep 1999 04:26:37 GMT
Reply-To: [EMAIL PROTECTED]
Thingi ([EMAIL PROTECTED]) wrote:
: How do I get the thing actually woking with RedHat 5.2?
: I'm a newbie to Linux and can't seem to find any how-to's on the subject...
: can someone please point me in the right direction.
Just load the bttv driver, and run xawtv.
: I'd like to use KWinTV, I've got K working on my i740 but I don't know
: how to setup the bttv hardware... + can't find any helpful documentation.
Search for linux+video or linux+tv or linux+bttv on the net.
: help!!!!!!!!!!!!!!!!!!!!!!!
I have no idea about redhat. I would guess you need to get the
video4linux code, patch it into your kernel, compile a module or two.
Trivia.
You might also upgrade to RH 6, which might have support in its premade
kernel. Or you (and it) might not.
I run bttv under kernel 2.0.25. I forget which driver version. I seem
to recall 0.6.4 has a problem. 0.4.3 and 0.5.6 go fine.
--
=====================================================================
Peter
------------------------------
** 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.hardware) 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-Hardware Digest
******************************
