Linux-Hardware Digest #585, Volume #10 Fri, 25 Jun 99 03:13:37 EDT
Contents:
Re: Windows easy to install? BULLSHIT! (Brian Hartman)
2.2.9 raid5 with 4 ide WD 20gig drive system lockup (Doug McClendon)
Re: SCSI drive was on linux SPARC problem. (Tim Lines)
Re: Windows easy to install? BULLSHIT! (Brian Hartman)
Re: NICs needed badly (TurboTex)
Re: Identifying USR modem (Giorgio Casinovi)
isdn card for linux... ([EMAIL PROTECTED])
Re: Please help: RedHat6.0 Iomega 250 Parallel Zip install and config (The RZA)
help! installing ftape drive (Alan Mead)
Re: Windows easy to install? BULLSHIT!
Re: Monitor dies while installing Redhat 5.2 ("Andrew J. Norman")
Re: Recommendation needed for Tape Backup drive (David C)
Re: Monitor dies while installing Redhat 5.2 ("Gene Heskett")
----------------------------------------------------------------------------
From: Brian Hartman <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.advocacy,comp.os.linux.setup
Subject: Re: Windows easy to install? BULLSHIT!
Date: Thu, 24 Jun 1999 23:19:15 -0400
[EMAIL PROTECTED] wrote:
> On Thu, 24 Jun 1999 15:17:27 -0400, Brian Hartman <[EMAIL PROTECTED]> wrote:
> >Michal Jaegermann wrote:
> >
> >> Brian Hartman ([EMAIL PROTECTED]) wrote:
> >> : Michal Jaegermann wrote:
> >>
> >> : > Brian Hartman ([EMAIL PROTECTED]) wrote:
> [deletia]
> >> Yes, yes! I know the theory. In theory I can even do that myself.
> >> Only a practice is a bit different and I have seen a similar stuff on
> >> a number of occasions. :-) This is called "Reboot, reboot, reboot and
> >> pray". The card in question, which was getting in, was one of Diamond
> >> S3 cards - BTW. When I have seen that "procedure" for the first time
> >> I was surprised myself. I did not realize that things are __that__ bad.
> >>
> >
> >The procedure I outlined above involves two reboots. One to acknowledge the new
>card,
> >another to complete the installation of the drivers. And if the drivers are any
>good (I've
> >got a Trident 9440 card in my PC) there won't be a problem. Amd amyway. Windows
>supports
> >more hardware than Linux, so you're more likely to have a problem on the Linux side.
>
> NO, you're more likely that it just plain won't work. However,
> if you're adding hardware that's a non-issue. The same effor
> you expend avoiding the ATIs can be put into ensuring that the
> new device will work well under Linux.
>
It's not the same effort. You can pick nearly any hardware device off the shelf in any
computer store and get it to work with Win 95 in some way. (There are exceptions of
course,
but less than there are with Linux. ) As far as it being "more likely that it "just
plain
won't work", the Linux problem with hardware is well-documented. From PC Magazine's
recent
review of the 2.2.x kernel:
First, support for plug-and-play is still poor. Of course, you
could say
the same about Microsoft's
PnP, but try installing even a commercial distribution of Linux
(from
Caldera or Red Hat, for
instance) and Windows 98 on the same machine and Win 98 will
automatically
detect and configure
far more devices. If you're coming to Linux from the Mac, which
has always
had superb
plug-and-play support (so good, in fact, it didn't even have to be
called
PnP), you'll find this even
more frustrating. Getting the video system to work with X Window
is better
than before but still too
difficult for many--indeed possibly most users--as are the
disk-partitioning requirements at
installation.
You can read the whole article at:
http://www.zdnet.com/pcmag/pctech/content/18/11/tf1811.001.html
The passage I qoted from can be found at:
http://www.zdnet.com/pcmag/pctech/content/18/11/tf1811.005.html
The article also notes that the kernel does not support USB. As I said earlier,
anecdotes not
withstanding, the idea that Linux supports hardware as well as Windows just doesn't
hold up.
> [deletia]
> >> : Win98 is not an OS for serious computer people doing intensive tasks. It's an
>OS for
> >> : novices and those who don't really understand the sofware side.
> >>
> >> It came preloaded on the laptop and it was not an option. It also has
> >> a number of nice toys to play with TV pcmcia cards which do not exist
> >> under Linux (can you spell "proprietary undocumented interfaces"?).
> >> There is also quite a few people convinced that you need it for
> >> "office productivity" software. Ahem...
> >>
> >
> >Office productivity software is quite a different thing from "intensive tasks".
>It's no
>
> That is the grand fallacy of the IBM PC Clone mentality. It's all
> the same sort of work. None of if is really any less valuable or
> any less 'real' for the person doing it. The expectation that
> your document won't disappear while you're editing it is no
> different than the expectation that your Alpha will stay up long
> enough to render this next frame.
It's not that the work is any less valuable. Obviously, the work has the same amount
of value
to the people doing the work. However, word processing, average office spreadsheets,
and
presentations don't take up the resources that data-intensive tasks (like video
editing) do.
In such intense environments, it matters a great deal how the OS works with the data.
If
you're saving one small file (as most word processing documents are, for example)
there's a lot
less chance of something going wrong, and the OS balking.
>
>
> [deletia]
> --
>
> It helps the car, in terms of end user complexity and engineering,
> that a car is not expected to suddenly become wood chipper at some |||
> arbitrary point as it's rolling down the road. / | \
>
> Seeking sane PPP Docs? Try http://penguin.lvcm.com
------------------------------
From: Doug McClendon <[EMAIL PROTECTED]>
Subject: 2.2.9 raid5 with 4 ide WD 20gig drive system lockup
Date: Fri, 25 Jun 1999 03:11:51 GMT
I'm running RedHat6.0, on an intel MB, p2-233
(also tried an fic-2013 + amd k6-2-450)
kernel 2.2.9
old raidtools-0.41
and I have 4 WD 20.4 (5400 rpm) ide drives.
(udma)
I have no problems doing mkfs on all the partitions
(each drive has a 1gig partition, and a 17gig partition).
And I have no problem doing mkraid on the 4 17gig partitions.
But as soon as I try to do anything with /dev/md0, it locks
up after a short period of time.
mkfs goes till either 54/2??? or 369/2???,
and hdparm -t /dev/md0 goes for a second or so.
After that point the system is completely wedged.
One oddity is that although all 4 drives are identical,
(identically set in BIOS as well)
hda and hdb have geometry "2482/foo/foo"
while hdc and hdd have geometry "35999/foo/foo"
(all numbers are sloppily approximate).
Anyway, I used fdisk to force hdc and hdd to have the same
geometry as hda and hdb, and this caused no problems when using
the partitions individually. Only when I use them as a single
device does it wedge up the machine.
Anyway, I'll go ahead and use the partitions individually and try
the raid thing once again once all those wonderful raid0145 patches
have actually made it into the stable kernel. (or I might give it
one more go disabling udma and seeing what happens.).
-dmc
------------------------------
From: Tim Lines <[EMAIL PROTECTED]>
Subject: Re: SCSI drive was on linux SPARC problem.
Date: Fri, 25 Jun 1999 03:50:49 +0000
Is this supposed to work? I have had trouble taking tapes back and forth
between SPARC and Intel architectures before because Intel machines are
little endian and SPARC chips are big endian. The bytes just don't make
sense if they're written backwards. There are a number of solutions to
this when dealing with a tape but I don't know how to solve this for a
filesystem.
root wrote:
> Hi All:
> was running linux on a SPARC with external SCSI disk drives.
> Have now moved the externals to a Linux PC box. Even though
> the drives were running ext2 when on the SPARC I can't
> read them now.:
>
>
------------------------------
From: Brian Hartman <[EMAIL PROTECTED]>
Crossposted-To: comp.os.ms-windows.advocacy,comp.os.linux.setup
Subject: Re: Windows easy to install? BULLSHIT!
Date: Thu, 24 Jun 1999 23:34:09 -0400
Brian Hartman wrote:
> [EMAIL PROTECTED] wrote:
>
> > On Thu, 24 Jun 1999 15:17:27 -0400, Brian Hartman <[EMAIL PROTECTED]>
>wrote:
> > >Michal Jaegermann wrote:
> > >
> > >> Brian Hartman ([EMAIL PROTECTED]) wrote:
> > >> : Michal Jaegermann wrote:
> > >>
> > >> : > Brian Hartman ([EMAIL PROTECTED]) wrote:
> > [deletia]
> > >> Yes, yes! I know the theory. In theory I can even do that myself.
> > >> Only a practice is a bit different and I have seen a similar stuff on
> > >> a number of occasions. :-) This is called "Reboot, reboot, reboot and
> > >> pray". The card in question, which was getting in, was one of Diamond
> > >> S3 cards - BTW. When I have seen that "procedure" for the first time
> > >> I was surprised myself. I did not realize that things are __that__ bad.
> > >>
> > >
> > >The procedure I outlined above involves two reboots. One to acknowledge the new
>card,
> > >another to complete the installation of the drivers. And if the drivers are any
>good (I've
> > >got a Trident 9440 card in my PC) there won't be a problem. Amd amyway. Windows
>supports
> > >more hardware than Linux, so you're more likely to have a problem on the Linux
>side.
> >
> > NO, you're more likely that it just plain won't work. However,
> > if you're adding hardware that's a non-issue. The same effor
> > you expend avoiding the ATIs can be put into ensuring that the
> > new device will work well under Linux.
> >
>
> It's not the same effort. You can pick nearly any hardware device off the shelf in
>any
> computer store and get it to work with Win 95 in some way. (There are exceptions of
>course,
> but less than there are with Linux. ) As far as it being "more likely that it "just
>plain
> won't work", the Linux problem with hardware is well-documented. From PC Magazine's
>recent
> review of the 2.2.x kernel:
>
> First, support for plug-and-play is still poor. Of course, you
>could say
> the same about Microsoft's
> PnP, but try installing even a commercial distribution of Linux
>(from
> Caldera or Red Hat, for
> instance) and Windows 98 on the same machine and Win 98 will
>automatically
> detect and configure
> far more devices. If you're coming to Linux from the Mac, which
>has always
> had superb
> plug-and-play support (so good, in fact, it didn't even have to
>be called
> PnP), you'll find this even
> more frustrating. Getting the video system to work with X Window
>is better
> than before but still too
> difficult for many--indeed possibly most users--as are the
> disk-partitioning requirements at
> installation.
>
> You can read the whole article at:
> http://www.zdnet.com/pcmag/pctech/content/18/11/tf1811.001.html
>
> The passage I qoted from can be found at:
> http://www.zdnet.com/pcmag/pctech/content/18/11/tf1811.005.html
>
> The article also notes that the kernel does not support USB. As I said earlier,
>anecdotes not
> withstanding, the idea that Linux supports hardware as well as Windows just doesn't
>hold up.
>
> > [deletia]
> > >> : Win98 is not an OS for serious computer people doing intensive tasks. It's
>an OS for
> > >> : novices and those who don't really understand the sofware side.
> > >>
> > >> It came preloaded on the laptop and it was not an option. It also has
> > >> a number of nice toys to play with TV pcmcia cards which do not exist
> > >> under Linux (can you spell "proprietary undocumented interfaces"?).
> > >> There is also quite a few people convinced that you need it for
> > >> "office productivity" software. Ahem...
> > >>
> > >
> > >Office productivity software is quite a different thing from "intensive tasks".
>It's no
> >
> > That is the grand fallacy of the IBM PC Clone mentality. It's all
> > the same sort of work. None of if is really any less valuable or
> > any less 'real' for the person doing it. The expectation that
> > your document won't disappear while you're editing it is no
> > different than the expectation that your Alpha will stay up long
> > enough to render this next frame.
>
> It's not that the work is any less valuable. Obviously, the work has the same
>amount of value
> to the people doing the work. However, word processing, average office
>spreadsheets, and
> presentations don't take up the resources that data-intensive tasks (like video
>editing) do.
> In such intense environments, it matters a great deal how the OS works with the
>data. If
> you're saving one small file (as most word processing documents are, for example)
>there's a lot
> less chance of something going wrong, and the OS balking.
>
> >
> >
> > [deletia]
> > --
> >
> > It helps the car, in terms of end user complexity and engineering,
> > that a car is not expected to suddenly become wood chipper at some |||
> > arbitrary point as it's rolling down the road. / | \
> >
> > Seeking sane PPP Docs? Try http://penguin.lvcm.com
I just looked at the message I posted and saw the messy line breaks. Sorry about that.
------------------------------
From: TurboTex <[EMAIL PROTECTED]>
Subject: Re: NICs needed badly
Date: Thu, 24 Jun 1999 23:48:30 -0500
Never pay more than 9.00 for 10baset
If you need some let me know.. 15.00 +s&h
"W. Kiernan" wrote:
>
> Vince wrote:
> >
> > Hello, I am in desperate need for 10 NICs for my RedHat Linux 5.2
> > boxes.
> >
> > I have read the http://www.redhat.com/hardware page so please don't
> > tell me to look there.
> > I have had no luck in finding supported NICs for my RedHat Linux
> > boxes.
> > Please, if someone could point me to a site that has supported NICs I
> > would be very very very thankful.
>
> If you'd be satisfied with 10 mbps ISA cards you can get 3C509s from
> www.buy.com for less than $30. That's for the fancy shrink-wrapped
> package too. They work great in Linux once you disable Plug-n-Play.
>
> Unfortunately they ship with Plug-n-Play enabled, which is a pain in
> either Linux or Windows NT. I'm told there is a way to disable
> Plug-n-Play from Linux, but I used the supplied 3C5X9CFG.EXE
> configuration program from MS-DOS to do that. (Actually there is a
> bacth file on the configuration diskette, PNPDSABL.BAT, which calls
> 3C5X9CFG.EXE with command-line switches to disable PNP.) You also must
> use 3C5X9CFG.EXE to set the I/O port and IRQ for this card; it has no
> jumpers, alas.
>
> Yours WDK - [EMAIL PROTECTED]
--
----
M.H. Collins < LINUX: The Official OS >
****** < for the New Millennium >
Powered by TurboLinux 3.4 http://www.linuxlink.com
Driven by XFCE2 http://www.austinlug.org
------------------------------
From: [EMAIL PROTECTED] (Giorgio Casinovi)
Crossposted-To: comp.os.linux.questions,comp.os.linux.setup
Subject: Re: Identifying USR modem
Date: 25 Jun 1999 03:54:25 GMT
In article <[EMAIL PROTECTED]>, Goodell <[EMAIL PROTECTED]> wrote:
>This may not be the best place to ask, but does anybody know how to
>identify USR modems by model/product numbers? The USR/3Com site is
>worthless. I need to know so that I can set up my ppp.
Have you tried http://consumer.3com.com/modem/upgrades/index.html ?
Once you have identified your product code, click on
"I already know my product code" and follow the
instructions. Clicking on "View Product Information"
will identify the type of modem with that product
code.
--
Giorgio Casinovi, Senior Research Engineer
Information Technology & Telecommunications Lab
Georgia Tech Research Institute, Atlanta, GA 30332-0832
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED]
Crossposted-To: comp.dcom.isdn
Subject: isdn card for linux...
Date: Fri, 25 Jun 1999 04:18:04 GMT
I want to replace my Ascend Pipeline75 with my linux server. I'd like to
put a bri card in the linux machine and run IP masquerading for my
network, and also have an analog modem in the linux box for ppp dialin to
my network. I wen't to redhat's website, but the only card they say
they've tested is the Spellcaster Datacommute. I've never heard of
Spellcaster, let alone the datacommute adapter. Has anyone used this card
with linux? Or any ISDN adapter with linux? I prefer a U-interface card
with at least one pots jack (prefer 2). Any information is appreciated.
Brent J. Miller
[EMAIL PROTECTED]
------------------------------
From: The RZA <[EMAIL PROTECTED]>
Crossposted-To: linux.redhat.misc,comp.os.linux.misc
Subject: Re: Please help: RedHat6.0 Iomega 250 Parallel Zip install and config
Date: Fri, 25 Jun 1999 01:31:29 -0400
I found out why 'insmod imm' alone results in unresolved symbols.
Instead
of the 'modprobe imm' what I've done is added the line 'insmod parport'
followed by 'insmod imm' to my rc.local file. The reason for the
unresolved
symbols is because imm needs parport loaded first.
Craig J Copi wrote:
>
> In article <[EMAIL PROTECTED]>,
> The RZA <[EMAIL PROTECTED]> writes:
> > Since you have the ZIP 250 drive, you should be using IMM rather than
> > the
> > older PPA. Also, instead of 'insmod imm' do a 'modprobe imm'... it
> > worked for me and I didn't even have to edit my conf.modules file.
>
> But what if you want kmod to do the work for you. Then what do you put in
> conf.modules? I tried setting scsi_hostadapter (and some other things I made
> up as I went along) but none of them worked. Just curious. A modprobe in
> rc.local works for now but I just hate to see "all that" memory wasted when
> the drive isn't being used;)
>
> --
> Craig J Copi | [EMAIL PROTECTED]
> Case Western Reserve University | http://erebus.phys.cwru.edu/~copi/
> Department of Physics | (216) 368-8831
------------------------------
From: [EMAIL PROTECTED] (Alan Mead)
Subject: help! installing ftape drive
Date: Fri, 25 Jun 1999 05:16:41 GMT
I have an exabyte eagle tr-3 with a floppy interface and when I
install it according to the docs my bios says something like "floppy
drive not detected correctly" and neither the tape nor the floppy are
accessable.
Any ideas?
------------------------------
From: [EMAIL PROTECTED] ()
Crossposted-To: comp.os.ms-windows.advocacy,comp.os.linux.setup
Subject: Re: Windows easy to install? BULLSHIT!
Date: Thu, 24 Jun 1999 21:10:10 -0700
On Thu, 24 Jun 1999 23:19:15 -0400, Brian Hartman <[EMAIL PROTECTED]> wrote:
>[EMAIL PROTECTED] wrote:
>
>> On Thu, 24 Jun 1999 15:17:27 -0400, Brian Hartman <[EMAIL PROTECTED]>
>wrote:
>> >Michal Jaegermann wrote:
>> >
>> >> Brian Hartman ([EMAIL PROTECTED]) wrote:
>> >> : Michal Jaegermann wrote:
>> >>
>> >> : > Brian Hartman ([EMAIL PROTECTED]) wrote:
>> [deletia]
>> >> Yes, yes! I know the theory. In theory I can even do that myself.
>> >> Only a practice is a bit different and I have seen a similar stuff on
>> >> a number of occasions. :-) This is called "Reboot, reboot, reboot and
>> >> pray". The card in question, which was getting in, was one of Diamond
>> >> S3 cards - BTW. When I have seen that "procedure" for the first time
>> >> I was surprised myself. I did not realize that things are __that__ bad.
>> >>
>> >
>> >The procedure I outlined above involves two reboots. One to acknowledge the new
>card,
>> >another to complete the installation of the drivers. And if the drivers are any
>good (I've
>> >got a Trident 9440 card in my PC) there won't be a problem. Amd amyway. Windows
>supports
>> >more hardware than Linux, so you're more likely to have a problem on the Linux
>side.
>>
>> NO, you're more likely that it just plain won't work. However,
>> if you're adding hardware that's a non-issue. The same effor
>> you expend avoiding the ATIs can be put into ensuring that the
>> new device will work well under Linux.
>>
>
>It's not the same effort. You can pick nearly any hardware device off the shelf in
>any
>computer store and get it to work with Win 95 in some way. (There are exceptions of
>course,
That simply doesn't cut it. No consumer should tolerate it either.
That any of them would shows that the current computing market is
very broken.
[deletia]
Nevermind that if you bother with such an artifically wide
definition of hardware support, Linux derives a similar
'benefit'.
--
It helps the car, in terms of end user complexity and engineering,
that a car is not expected to suddenly become wood chipper at some |||
arbitrary point as it's rolling down the road. / | \
Seeking sane PPP Docs? Try http://penguin.lvcm.com
------------------------------
From: "Andrew J. Norman" <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Crossposted-To: comp.os.linux.advocacy
Subject: Re: Monitor dies while installing Redhat 5.2
Date: Fri, 25 Jun 1999 04:51:22 GMT
Okay here's the deal.......you have an old S3 video card right? Most
likely a Diamond Speedstar or similar....
These cards (I have one in an old machine myself) typically have a bank of
dip switches on the back of the card (poking through a hole next to the
video output) These control the way that the card syncs and addresses
memory and bios.....ahh memory and bios you say (lightbulbs go on all over
the place)......yes!
Reading from my own manual......on the page marked "DIP Switch SW1
Settings" (note you will want to use your own manual since the chances of
us both have the exact same ten year nine year old card are slim to none)
1 Turbo display memory access (ON)
1 Normal display memory access (OFF)
2 Turbo display memory access (ON)
2 Turbo display memory access (OFF)
3 16-bit bios access (ON)
3 8-bit bios access (OFF)
.
.
.
yadda yadda........
Point is that newer kernels (by that I mean the 2.x series) seem to have a
problem with certain combinations of "turbo" access both to memory and
bios. Additionally if you bios are cached you might have a problem (or if
you are trying ot cache the video memory, which some new boards allow)
FYI this also occures in a more dramatic way with an older model (well one
of last summer's) of Creative Labs Graphics Blaster Extreme, except with
it, the bios issue a jump (which I am assuming mistakenly goes to
0xFFFF:0000 when the bios are cached) and the whole machine reboots. Best
solution in that case was to give the card to my brother (it plays Quake
rather well) and get a cheap VT420 terminal....don't ask....some of us like
amber.
In this case however you should be able to tweek the dips (or jumpers if
you don't have dips) and get the card to work by turning off the turbo
modes or reducing access to 8-bit modes (yes this will be slow but you're
using an old card to begin with). Note: this is a video card problem not a
monitor problem. The fact that your monitor goes blank is only an effect
of it loosing signal from the video card. Should all else fail you can
pick up a new (an probably much nicer) card for some where in the
neighborhood of $20-$30 mail order.
If you have more problems let me know and I can probably give you the
setting that I finally set my card to.
Barry Smith wrote:
> I have posted this on other groups, but have had no solution.
>
> I am trying to install RedHat 5.2 on a very old but much upgraded Dell
> computer. The BIOS is Phoenix, vintage 1993, and not upgradeable. I am
> using the free distribution which came from the magazine PC Plus, and
> created the boot disk. I have tried booting from the install floppy in
> default and expert mode but each time the monitor dies, and the light
> that indicates connection with the PC extinguishes. This happens soon
> after I press ENTER from the welcome screen. I see "Loading initrd.img";
> the monitor dies, but the disc keeps busy for a while.
>
> The computer uses ordinary system memory for display, and in Windows 9x
> it runs as a default S3 type. I have searched the documentation but
> can't find the parameter to define my display type in expert mode.
>
> Someone suggested fitting a video board. I don't see what difference
> this would make and is not really a viable option. Others have suggested
> twiddling X-Windows, but the monitor dies long before X-Windows is
> loaded. Oddly enough, the same thing happens when I boot SuSE 5.2 from
> the floppy. However, when I boot SuSE from Linload, everything looks
> fine, and I have got as far as the point where you partition the disc.
>
> What can I do?
>
> --
> Barry Smith
> Email: sax (at) wychcraft.demon.co.uk <-- I don't want ANY spam!
--
Andrew J. Norman
Dept. of Physics
College of William & Mary
_____________________________________________________________
"I am afeard there are few die well that die in a battle;
for how can they charitably dispose of any thing,
when blood is their argument?" -- Williams
Henry V--Act 4 Scene 1
_____________________________________________________________
------------------------------
From: David C <[EMAIL PROTECTED]>
Crossposted-To:
alt.os.linux,comp.os.linux,comp.os.inux.admin,comp.os.linux.misc,comp.os.linux.redhat
Subject: Re: Recommendation needed for Tape Backup drive
Date: 25 Jun 1999 02:04:21 -0400
Martin Kiely <[EMAIL PROTECTED]> writes:
>
> I'm planning to co-locate an Intel box running redhat 6, can someone
> recommend a tape drive and software combination.
Anything SCSI (SCSI-2 or better) is almost certain to be compatible with
any operating system, assuming you've got the SCSI card's device driver
installed OK. (The one exception being Win95 - it's bundled backup
program doesn't support SCSI tapes, but third-party backup programs do.)
Aside from that, go with whatever mechanism your budget/needs dictate.
The nice thing about SCSI-2 is that a generic command-set for tapes is
defined, so all drives should be equally compatible. I've used
QIC/Travan drives, 4mm DAT and 8mm drives with no problems.
CD-R and CD-RW may also be viable alternatives, depending on what your
backup needs are. For me, I need tape - so I can insert one tape and do
an unattended back up of my entire file system (currently about 1.5GB).
-- David
------------------------------
Date: 25 Jun 99 03:02:08 -0500
From: "Gene Heskett" <[EMAIL PROTECTED]>
Subject: Re: Monitor dies while installing Redhat 5.2
Crossposted-To: comp.os.linux.advocacy
Gene Heskett sends Greetings to Donovan Rebbechi;
DR> On Thu, 24 Jun 1999 16:41:05 -0700, Alex Lam wrote:
>>Dump Redhat.. Redhat killed one of my monitor before during installation
>>probing. I mean totally killed it.
DR> killed it ? The monitor shouldn't die unless you sync it out of spec
DR> ( and even then, it should be able to shut itself down when it's
DR> overdriven )
DR> I'd suggest you return your monitor for a prompt refund.
>From a tech:
A plumb well thought out response, NOT. What if the monitor is already 5
years old? Or more?
While its possible the newer ones, particularly those that can display
their sweep speed parameters might well shut themselves down in self
defense, or simply refuse to go that low and drop out of sync to protect
themselves, older ones quite often do not.
If the sync circuitry is capable of slowing it down enough to allow the
outout transformers ferrite core to become magneticly saturated as the
beams approaches the right border of the screen, and it has no other
protective circuitry that can act litterally in a couple of
microseconds, then the output transistor, and probably a lot else in the
general power supply area will be destroyed, and about 100x faster than
a fuse can blow. Once the core is saturated, the current can rise to
100 amps or more in just another microsecond, depending on the condition
of the filter capacitor supplying the circuit. And old, high ESR
capacitor MAY save the transistor for a few seconds and repeats IF
the transistor can successfully shut itself off at the right border of
the screen from such an overcurrent, but the capacitor will probably
blow itself all over the insides of the monitor doing it. Messy.
In that case, the user is out of luck, and looking at a fairly hefty
repair charge from his neighborhood tech (if he knows about such, many
don't appreciate the finer nuances of the results of 'good enough for
the girls I go with' parts replacements), or the maker if they are
within a reasonable shipping charge distance, and still want to support
the oldies but goodies.
During any install, if you do not know the exact min-max speed
parameters for your monitor, and the monitor is not on the builtin list,
do NOT let it probe, but choose whatever default that's currently working,
and leave it that way until you can obtain that information from some
source, and THEN go edit the /etc/X11/XF86Config file to set these
limits properly.
Horizontal overspeed is not usually so dangerous, but underspeed can be
an instant killer. Grab the nearest magic marker (permanent type) and
write it on the wall.
My condolences to the gentlemans dead monitor.
Cheers, Gene
--
Gene Heskett, CET, UHK |Amiga A2k Zeus040 50 megs fast/2 megs chip
Ch. Eng. @ WDTV-5 |A2091,GuruRom,1g Seagate,CDROM,Multiface III
|Buddha + 4 gig WDC drive, 525 meg tape
|Stylus Pro, EnPrint, Picasso-II, 17" vga
RC5-Moo! 690kkeys/sec isn't much, but it all helps
email gene underscore heskett at iolinc dot net
--
------------------------------
** 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
******************************
