Linux-Hardware Digest #463, Volume #12           Sun, 12 Mar 00 15:13:09 EST

Contents:
  Re: COM1 or COM2 for the ext. modem; whats optimal? (Keith R. Williams)
  Re: Real audio, /dev/mixer (Chris J/#6)
  Re: OnStream ADR50 (Sami Shaaban)
  Re: 1: Xvidtune, 2: fix freq monitor (Henrik Carlqvist)
  Re: caller id solutions (Henrik Carlqvist)
  A way to determine uptime (was Re: Cyrix CPUs) (Henrik Carlqvist)
  Re: VIA vs Intel chipsets - which is better? (Dan Wojciechowski)
  Re: ftape with 2.2.x kernel (Robie Basak)
  Zoom modem configuration (rjuarezp)
  MDA and Linux? (Milton Johnson)
  Re: Linux sucks (Pjtg0707)
  Re: modems and linux... ("plato")
  Laptop problem (Chris Davis)
  Driver Geometry
  Re: Anybody really good at monitor-settings? ([EMAIL PROTECTED])
  Re: memory problem ("John J. Budd III")
  Re: memory problem (Hal Burgiss)
  ATA66 and Linux ("news-server.optonline.net")
  UDMA66 HPT366 linux driver ("Damon Tsang")
  Re: XFree86 with Intel i810 chipset ([EMAIL PROTECTED])
  Re: Best Distribution for Laptop (Jerry McBride)
  Re: VIA vs Intel chipsets - which is better? ("Remy Moulin")

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

From: [EMAIL PROTECTED] (Keith R. Williams)
Reply-To: [EMAIL PROTECTED]
Crossposted-To: 
alt.comp.periphs.mainboard.asus,comp.sys.ibm.pc.hardware.chips,comp.sys.ibm.pc.misc
Subject: Re: COM1 or COM2 for the ext. modem; whats optimal?
Date: 12 Mar 2000 14:21:23 GMT

On Sun, 12 Mar 2000 12:34:33, "Wayne Monteath" 
<[EMAIL PROTECTED]> wrote:

> COM2 uses IRQ3.
> COM1 uses IRQ4
> 
> IRQ3 gets serviced (has a higher priority) before IRQ4.

This is backwards (which is why COM1 is on IRQ4 by default), but 
it doesn't matter in any case.

----
  Keith 



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

From: [EMAIL PROTECTED] (Chris J/#6)
Subject: Re: Real audio, /dev/mixer
Date: 12 Mar 2000 14:27:01 -0000
Reply-To: [EMAIL PROTECTED]

Find the URL of the RA file, then use wget to fetch it...I've used this
method before now, with some success. Similarly works for midi files,
mods, wavs, and anything else that gets thrown on a web page.

Chris...

Martin Booth  <[EMAIL PROTECTED]> wrote:
>Dear Folks,
>
>Sorry for the cross-posting, I'm a bit pushed for time - that is if
>anyone knows the answer that is. I have a problem in that I would like
>to download a real audio file and copy it to CD ( a present for
>someone). I've tried to download the real audio file but it just has a
>link to pnm://..... . I've installed real audio on my Linux system (more
>
>of that later) and tried to save it to a file. I've, naively, tried to
>replace the /dev/dsp with a file and hope that would catch the data
>before it hit my sound card. Needless to say it didn't work. Anyone know
>
>how to get the data to disk in a readable format?
>

-- 
@}-,'--------------------------------------------------  Chris Johnson --'-{@
    / "(it is) crucial that we learn the difference / [EMAIL PROTECTED]  \
   / between Sex and Gender. Therein lies the key  /                       \ 
  / to our freedom" -- LB                         / www.nccnet.co.uk/~sixie \ 

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

From: Sami Shaaban <[EMAIL PROTECTED]>
Subject: Re: OnStream ADR50
Date: Sun, 12 Mar 2000 15:00:11 GMT

Hmmm ... it seems that turning off the drive's compression (using mt)
makes this problem go away.

I hope this helps anyone with a similar problem.  However, if anyone can
offer an insight on how to turn compression back on without receiving
this error, that would be great.

Sami Shaaban wrote:
> 
> I just ordered an ADR50 and installed it on a RedHat 6.1 system.  When
> using it with some common backup software (Amanda), certain backup
> volumes
> written to the tape contain unexpected EOFs (tar volumes written to tape
> with dd).
> 
> Upon further exploration, I found that the kernel was complaining of the
> following:
> 
> Mar 11 10:02:23 nurelm kernel: st0: Error with sense data: Info
> fld=0x40, Current st09:00: sense key Medium Error
> Mar 11 10:02:23 nurelm kernel: Additional sense indicates Block sequence
> error
> 
> I am using a Tekram DC-390F SCSI card, which seems quite well supported
> and is automatically detected and configured by the OS.  I tried another
> backup tape in case the media was hosed.
> 
> I would be grateful for any hints as to what's going on.

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

From: Henrik Carlqvist <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.x
Subject: Re: 1: Xvidtune, 2: fix freq monitor
Date: Fri, 10 Mar 2000 20:05:23 +0100

[EMAIL PROTECTED] wrote:
> Has anyone has some experience to share about Xvidtune.  I am new
> with this and would like to hear some thing before I use this.

It is a program which you could use to adjust the image on your monitor.
Once you have found a good image you could make it print a modeline
which you later can use by editing you XF86Config file.
 
> (2)  Connecting a Fix Freq Monitor

You can use fixed freq monitors with XFree86. Instead of giving syncs
like "30-70" you should give them like "31.5, 35.2".

> I am seriously considering to pick up some
> cables, connectors and a soldering iron to make a set of BNC cable to
> plug into the VGA (Dsub15graphics) connector of my graphic card. 

Instead of soldering the cable yourself it is probably a better idea to
buy a cable. Reflections in a bad made cable will show up as shadows in
the image.

> Q Further to this is it correct that in Xconfigurator I should
> select.. "Non-Interlaced SVGA, 1024x768 @ 60 Hz, 800x600 @ 72 Hz" ? 

Probably not. I have never used Xconfigurator, try xf86config instead.
There you will be given an option to give the frequencies yourself. Then
give them as comma separated lists.

> Q (BTW, what is a clockchip asked by Xconfigurator?  Is this related
> in anyway to the motherboard?

No, it is a feature of some graphic cards.

> Q Can someone tell me what are te possible causes when, within the
> server pressing Ctrl, Alt and '+' simultaneously to cycle video
> resolutions do not work in my box?

If you get the same old image it is probably because some of your
resolutions has been removed. This is probably because the settings of
your monitor or graphics card would be unable to handle them.

If your image gets black or shaky it is probably becouse you have given
settings that your monitor really is unable to handle.
 
> P/S I am hoping that the image will eventually fill up the screen to
> within 5 mm or so from the sides/edges. :-)

If you are unable to adjust your monitor you might be able to fix this
with xvidtune.
 
regards Henrik
-- 
spammer strikeback:
root@localhost
[EMAIL PROTECTED]

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

From: Henrik Carlqvist <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.misc
Subject: Re: caller id solutions
Date: Fri, 10 Mar 2000 20:31:51 +0100

matthew neil garman wrote:
> The problem with this "solution" though is that minicom locks the
> modem.  So to use ppp, I have to quit minicom, then restart minicom
> after ending ppp.
> 
> I was wondering if there is a simple way to use built-in Linux
> functionality to monitor caller id.  What I would like is to have
> something (e.g. syslog) log caller id information from the modem,
> without locking the device.

If you write a program which reads from the serial port without first
locking it you will be able to use ppp at the same time as you program
is reading from the serial port. Unfortunately characters from the
serial port will not be sent to both programs. Some of the characters
will go to ppp and some will go to your program in a random order. Then
both your program and ppp will be unusable.

You will have to write some kind of wrapper which first kills the
caller-id application, then starts ppp. After that you could start your
caller-id application from the ip-down script again.

> Ideally, I would have a log such as
> /var/log/caller_id.log, then open up a small xterm and run "tail" on
> /var/log/caller_id.log.  This way, the most recent caller is always
> displayed in the xterm, and I have scrollback history.
> 
> Does anyone know if I can implement such a thing relatively simply?

You might just as well keep using minicom. Make an alias or write a
shell script which kills minicom before starting ppp. Or better than
minicom, write a program which does 
system("xterm -title caller-id -e cat /dev/modem");

regards Henrik
-- 
spammer strikeback:
root@localhost
[EMAIL PROTECTED]

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

From: Henrik Carlqvist <[EMAIL PROTECTED]>
Subject: A way to determine uptime (was Re: Cyrix CPUs)
Date: Fri, 10 Mar 2000 21:34:40 +0100

FWFR wrote:
> This is a longest since a reboot, or major power outage. 
> 1:03am  up 137 days,  9:36,  1 user,  load average: 0.00, 0.00, 0.00

I have an old 486 with a long uptime. One easy way to check the uptime
was to use "ruptime" as I have rwhod running. Unfortunately this has
given an erronious value for a long time. 

Since a few weeks I also get a wrong value by running uptime locally.
Maybe the uptime got wrapped at 512 days.

Checking with ps also make it seem as if the init process and a lot of
other processes were started only a few weeks ago.

There are only two ways I can tell that the machine really hasn't
rebooted:

A process (rc564) which isn't started automatically by any startup
script is still running.

There are no traces of any reboot in the syslogs. However, I'm still
unable to determine the exact uptime as I only keep my logs for about a
year.

There is one more thing which looks a little strange:

hydran:~> cat /proc/interrupts 
 0:  208007750   timer
 1:     167002   keyboard
 2:          0   cascade
 4:      32883 + serial
 5:   13861418   3c503-PIO
13:          1   math error
14:  120958239 + ide0

Some week ago the number of ide0 interrupts were bigger than the timer
interrupts. I suppose that the number of timer interrupts has wrapped
over.

Does anyone know how to determine my uptime? I'm running Slackware 3.1
with kernel upgraded to 2.0.30. 

regards Henrik
-- 
spammer strikeback:
root@localhost
[EMAIL PROTECTED]

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

From: Dan Wojciechowski <[EMAIL PROTECTED]>
Crossposted-To: comp.sys.ibm.pc.hardware.chips,comp.sys.ibm.pc.hardware.systems
Subject: Re: VIA vs Intel chipsets - which is better?
Date: Sun, 12 Mar 2000 09:41:20 -0600

Ron Reaugh wrote:
...
> Typical of the kind who has nothing to contribute of technical value to the
> NG.  Your type always deals in personalities and not the facts.  Why bother
> even posting.

I've dropped in kind of late... What's the beef Ron?  I don't sell anything.
I've built and used some 15 or 20 systems over the last 15 years.  (So my
experience is statistically meaningless.)  I've had some Intel motherboards,
some VIA motherboards, and a couple SIS motherboards.  I've never had one
that didn't work.  Currently, two of my home systems use FIC VA503+
motherboards.  (One is the original 1.0a, the other is a new 1.2a.)  I have
bus mastering eide drivers loaded for both boards.  Both work flawlessly.

So, once again, what's the beef?  Contribute something of technical value
to me:  tell me which VIA chipsets you are talking about; tell me what the
problems are.
-- 
                   Dan (Woj...) [EMAIL PROTECTED]
" Did you ever wonder why we had to run for shelter when the promise
  of a brave new world unfurled beneath the clear blue sky? "

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

From: [EMAIL PROTECTED] (Robie Basak)
Subject: Re: ftape with 2.2.x kernel
Reply-To: [EMAIL PROTECTED]
Date: 13 Mar 2000 00:01:53 GMT

It compiled perfectly for me (2.2.14) with ftape-4.x-1999_09_11 for an Iomega
Ditto Max - I did have timeout errors with the latest stable version.

On Sat, 11 Mar 2000 19:30:13 GMT, scott <[EMAIL PROTECTED]> wrote:
>     This group has saved me before, and I hope it can again.
>
>     I have read all of the discussions concerning ftape, as well as the 
>websites devoted to ftape. I understand that when compiling ftape with the 
>2.2.x kernels, you get various fdc-io.c timeout errors. It was suggested 
>to use the latest unstable version of ftape (4.03) when compiling it with 
>kernel version 2.2.x and later. I have done this and the results are the 
>same. I even tried installing an rpm version 4.02 without success. Though 
>there were no errors, it didn't seem to install anyways. So my question is 
>has anyone had any success installing ftape on a 2.2.x kernel? I've got so 
>many residual ftape files on my system, I don't know which one is which 
>anymore. I'm afraid some of these might confuse a potential successful 
>installation at a later date.
>
>     OK, to recap. I would like to install ftape on a 2.2.x kernel WITHOUT 
>any errors. I've got RedHat 6.0, and a Iomega Ditto 3200 1.6 GB tape drive 
>that uses the floppy controller. I've tried the latest version of ftape. 
>Where should I go next? And please inform me if and how I should clean out 
>my previous failed ftape files to avoid confusion. Thanks everyone.
>
>
>--
>Posted via CNET Help.com
>http://www.help.com/


-- 

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

From: rjuarezp <[EMAIL PROTECTED]>
Subject: Zoom modem configuration
Date: Sun, 12 Mar 2000 15:30:07 GMT

  The KPPP configuration program finds my modem (on COM 2 when in Windows 
it is on COM 4), but when it tries to get the ATI information, it gets 
nothing (everything is blank). When connecting the modem hangs on when 
initializating. I've got a Zoom internal 56K Faxmodem. What do I have to 
do?

   Please help me!! Thanks.

--
Posted via CNET Help.com
http://www.help.com/

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

From: Milton Johnson <[EMAIL PROTECTED]>
Subject: MDA and Linux?
Date: Sun, 12 Mar 2000 16:20:45 GMT

I have an old monochrome display adapter and mber monitor I would like
to use on a linux/samba server. I cant seem to get linux to boot up with
a mono card. any help?
Thanks,
Milt Johnson


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

From: [EMAIL PROTECTED] (Pjtg0707)
Crossposted-To: comp.os.linux.networking,comp.os.linux.portable,linux.redhat.install
Subject: Re: Linux sucks
Reply-To: [EMAIL PROTECTED]
Date: Sun, 12 Mar 2000 16:22:46 GMT

On Sun, 12 Mar 2000 11:11:59 GMT, Rusty Lingenfelter <[EMAIL PROTECTED]> wrote:
>Will all of you please take this rant to an appropriate newsgroup. I 
>would suggest:
>
>comp.os.linux.advocacy or
>alt.linux.sux
>
>As you all seem to have plenty of time to read, read a bit about 
>newsgroup etiquette. Please get the hint.

Finally! A reasonable post on this stupid thread that ought to die a
LOOOOOOOOOONG time ago.


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

From: "plato" <[EMAIL PROTECTED]>
Subject: Re: modems and linux...
Date: Sun, 12 Mar 2000 18:33:35 +0200

sorry..thanks :))



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

From: Chris Davis <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.setup,comp.os.linux.misc,comp.os.linux.help
Subject: Laptop problem
Date: Sun, 12 Mar 2000 11:46:26 -0500

This is a multi-part message in MIME format.
==============3AED724070EBCB802F16161D
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit


Anyone out there running RH6.0 on a laptop with a Trident CyberBlade i7
video interface?
My system is a Compaq Presario 1277.

I'm having trouble with XF86 and would greatly appreciate some help. X
server crashes reporting problems with "ModeLine" entries.
Perhaps an XF86Config file?
I'm new to the laptop scene so bear with me.

TIA,
chris


==============3AED724070EBCB802F16161D
Content-Type: text/x-vcard; charset=us-ascii;
 name="cdavis.vcf"
Content-Transfer-Encoding: 7bit
Content-Description: Card for Chris Davis
Content-Disposition: attachment;
 filename="cdavis.vcf"

begin:vcard 
n:Davis;Christopher
x-mozilla-html:FALSE
org:Embedded Systems Design, Inc.
adr:;;;;;;
version:2.1
email;internet:[EMAIL PROTECTED]
title:Embedded Specialist
x-mozilla-cpt:;0
fn:Christopher Davis
end:vcard

==============3AED724070EBCB802F16161D==


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

From: <[EMAIL PROTECTED]>
Subject: Driver Geometry
Date: Sun, 12 Mar 2000 17:30:22 GMT

I was a loyal Mandrake user when I downloaded an image of Redhat 6.1.  When 
I tried to install it I somehow got the driver geometry screwed up and 
Redhat would read my 4gig as an 8 gig.  This carried over into this error 
showed up in NT4 and 2000.  I finally got this cleared up with a couple of 
unconditional formats.  I thought I had this issue fixed when I downloaded 
an image of Mandrake 7.  When I tried to install this the drive geometry 
was being miscalcualted once again.  Please Help!!!!!!

Tom

--
Posted via CNET Help.com
http://www.help.com/

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

From: [EMAIL PROTECTED]
Subject: Re: Anybody really good at monitor-settings?
Date: Sun, 12 Mar 2000 18:33:15 GMT

In article <[EMAIL PROTECTED]>, Lars Olsson <[EMAIL PROTECTED]> wrote:
> I run Linux with XFree86. I have a no-name monitor.
> It is 17" and is capable of 85Hz @ 1024x768, and
> run at that in Windows98 (shiver).
> In Linux it runs at 75Hz. I´ve located the line in
> XF86Config that the server uses to set that setting,
> and I´ve tried to edit it. No luck. My question:
> There´s a value on that line called "DotClock", and
> it´s not mentioned in my low-budget manual for the
> monitor. All that´s covered is hor. and ver. freq. for the
> different settings. Here´s what it says for the setting I would like
> to use:
> Horizontal freq = 68.677 kHz
> Vertical freq = 84.997 Hz
> It also mentions some info on the timing chart (T1-T5)
> and Sync Polar. Is there any way to calculate the "DotClock"
> from this info? Any help is greatly appr.
> /Lars
> 
All you have to do is increase the dot clock value, at any given 
resolution, and the refresh rate will increase.  Dot clock = pixels
per second = refresh rate times resolution.



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

From: "John J. Budd III" <[EMAIL PROTECTED]>
Reply-To: [EMAIL PROTECTED]
Subject: Re: memory problem
Date: Sun, 12 Mar 2000 13:00:20 -0600

I have been in the bios many times but it does have any option to do
anything to the memory.  Is there something I am missing?  JJB

Hal Burgiss wrote:

> On Sun, 12 Mar 2000 00:25:33 -0600, John J. Budd III
> <[EMAIL PROTECTED]> wrote:
> >40 megs according to bios.  13 Megs according to Linux.  40 Megs
> >according to Windoze.  Researched Deja and tried the append=memxxx32M
> >etc but it just crashes the machine.  This is a Packard Bell Pentium
> >200.  I went to their website looking for a different bios but I have
> >the latest they offer.  Does anyone have any suggestions?  JJB
> >
>
> Disable 'memory hole at 16M' in BIOS.
>
> --
> Hal B
> [EMAIL PROTECTED]
> --


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

From: [EMAIL PROTECTED] (Hal Burgiss)
Subject: Re: memory problem
Reply-To: [EMAIL PROTECTED]
Date: Sun, 12 Mar 2000 19:15:38 GMT

On Sun, 12 Mar 2000 13:00:20 -0600, John J. Budd III
<[EMAIL PROTECTED]> wrote:
>I have been in the bios many times but it does have any option to do
>anything to the memory.  Is there something I am missing?  JJB

That's unfortunate because that is just the kind of thing the 'memory
hole' setting does on Linux. Maybe the BIOS is fubarred. What is the
exact statement for 'append' you are using for lilo? If you are using
'memxxx32M', I do not get the 'xxx' part of this. Try this at the LILO
prompt:

 linux mem=40M

If this works, stick 

 append="mem=40M"

in lilo.conf. If this works, but crashes later, back off the 40M to 39M.
Some BIOSes tie up a bit of memory.

>Hal Burgiss wrote:
>
>> On Sun, 12 Mar 2000 00:25:33 -0600, John J. Budd III
>> <[EMAIL PROTECTED]> wrote:
>> >40 megs according to bios.  13 Megs according to Linux.  40 Megs
>> >according to Windoze.  Researched Deja and tried the
>> >append=memxxx32M etc but it just crashes the machine.  This is a
>> >Packard Bell Pentium 200.  I went to their website looking for a
>> >different bios but I have the latest they offer.  Does anyone have
>> >any suggestions?  JJB
>>
>> Disable 'memory hole at 16M' in BIOS.


-- 
Hal B
[EMAIL PROTECTED]
--

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

From: "news-server.optonline.net" <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux,linux.redhat.install
Subject: ATA66 and Linux
Date: Sun, 12 Mar 2000 19:45:14 GMT

I have a dell 700mhz system with an ata66 controller card, and I am trying
to install redhat linux 6.1.  When I get to the part where I select what
kind of installation I will perform, it says that it cannot find any media
to install to, and then leaves me hanging.  What is an ata66 controller card
and how do I get it to work with linux (I am very comfortable with computers
in general, but only so-so with linux).



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

From: "Damon Tsang" <[EMAIL PROTECTED]>
Subject: UDMA66 HPT366 linux driver
Date: Mon, 13 Mar 2000 03:51:42 +0800

Do anyone know where to find Highpoint HPT366 UDMA66 linux driver.

-- Damon



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

From: [EMAIL PROTECTED]
Crossposted-To: alt.os.linux.mandrake,comp.os.linux.misc,comp.os.linux.setup
Subject: Re: XFree86 with Intel i810 chipset
Date: Sun, 12 Mar 2000 19:47:25 GMT

The intel driver does not support the audio.
Check the opensound website where you can find
the open sound system program which is said to
support i810 chips. In any case, you'll get to
know which driver really works since it is not
really clear if it is actually the i810 or
another controller that deals with the sound.

In article <895f9a$[EMAIL PROTECTED]>,
  "Mees Mehepoeg" <[EMAIL PROTECTED]> wrote:
> HELP
>
> I have RPMs for intel and graphic works fine
> but sound not. Sndconfig found card but not
> seting work. Some problem with modem.
>
>


Sent via Deja.com http://www.deja.com/
Before you buy.

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

From: [EMAIL PROTECTED] (Jerry McBride)
Crossposted-To: comp.os.linux.portable,comp.os.linux.setup
Subject: Re: Best Distribution for Laptop
Date: Sun, 12 Mar 2000 13:38:06 -0500

In article <893n8r$bj5$[EMAIL PROTECTED]>,
"Kachi Armony" <[EMAIL PROTECTED]> wrote:
>Hi. I have a NEC 2700 series laptop (P133, 32MB RAM, 2.1GB HD, CDROM, Modem,
>PCMCIA Ethernet, etc).
>I will like to know what the best distribution for laptops would be. I have
>a 450MB partition ready for Linux to be installed (dual boot with WIN95). I
>guess my main use for Linux will be Development (C++, Java, and others - I'm
>a CS student). I won't need any fancy apps. If everything (hardware) works
>fine, I may eventually reformat and have my entire drive for Linux only (and
>get rid of Windoze). Please advise.
>

Caldera Open Linux 2.3 installed without problems using LISA. It detected
everything on my OLD Toshiba 2135 including the PCMCIA attached cdrom...


--

*******************************************************************************

Knowledge without wisdom is useless.

*******************************************************************************
* NetRexx - The onramp to the Internet - http://www2.hursley.ibm.com/netrexx  *
*******************************************************************************


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

From: "Remy Moulin" <rmoulin*NoSpam*@club-internet.fr>
Crossposted-To: comp.sys.ibm.pc.hardware.systems,comp.sys.ibm.pc.hardware.chips
Subject: Re: VIA vs Intel chipsets - which is better?
Date: Sun, 12 Mar 2000 21:10:17 +0100

Keith R. Williams <[EMAIL PROTECTED]> a écrit dans le message :
[EMAIL PROTECTED]

> Stand behind the assertion (yours BTW) that there are no
> busmastering problems?  Nope.  I can't do that, though I haven't
> seen one for some long time. I know you keep reporting them,
> though you admit that you never use anything but Intel.  So be
> it.  I've always liked SiS. It's too bad they lost focus.
> However, I've never had any problems with my ALi or Via based
> boards. ...yes, teething pains, but no more.

Well, I do have some "interesting features" with ALi chipset that I don't
have with Intel or VIA. I won't speak of SiS as I don't have enough feedback
to talk with these.

It only concern Windows OS as Linux is a totally different thing... (See the
crosspost addresses)

Sometimes, the included Windows Bus-Mastering driver for the ALi don't
recognise a DVD-rom or a CD-Rom. When it happen, the controller (primary or
secondary) is in error with a nice yellow exclamation point on the icon in
the device manager. So, when this happen, I just have to install the ALi
bus-master driver into windows et voilà, no more yellow exclamation point.

Where's the feature ? Here it is:
 I have the (bad ?) habit of installing the dos Cd driver in config.sys and
autoexec.bat for my customer ease of use (some of them do use dos games in
true dos sessions), and this works well either with windows default driver
and  with Intel or VIA driver. However, with the ALi driver, it always
end-up with a system lock at windows boot time...

Is this a bus-mastering problem ? Not at all ! I just avoid loading
mscdex.exe when I need to install the ALi bus-mastering driver. It's rather
a driver issue.

Notice that the VIA driver readme recommend to not install any dos
Cd-driver, even though it seems to always works as a charm. Maybe Intel says
the same things, I never read the Intel readme ;-)...

Also, Microsoft do admit there can be problems in Windows 95 (OSR 2+) when
ticking the DMA box for a device and when it's U-DMA 33 (66 also ? I've not
verified), so they have done a patch to correct ESDI_506.PDR and
VOLTRACK.VXD. Interestingly, that patch can be used also on Windows 95 Gold
(OSR 0) and OSR 1... Now, each time I have to reinstall Win95 whatever
version, I install that patch: REMIDEUP.EXE.

Last note, ESDI_506.PDR is upgraded again when using the K6 patch, but not
VOLTRACK.VXD. So I think it's better to still install REMIDEUP before using
AMDK6UPD. Also, I've noticed it doesn't hurt when it's used with an Intel
CPU (even though the posts I've read here said the way round), when I do
have a bit of time, I install it also for Intel CPUs... It upgrade many
system files !

Just my 0.02?,
Remy.
Tech in a after-sales service.



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


** 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
******************************

Reply via email to