Linux-Hardware Digest #16, Volume #13 Sat, 10 Jun 00 00:13:05 EDT
Contents:
Re: Athlon recomendations (Marshall Blythe)
Re: Athlon recomendations (Marshall Blythe)
Re: Redhat 6.1 compatibility with Abit ka7 or Asus k7V? (Marshall Blythe)
Re: Drivers for a) modem b) sound card (Dances With Crows)
Re: Need your help again... (Duane)
Re: MATSHITA LS120 Ver5 ("J T")
Re: IDE Zip disk faster via ide-scsi ??? (Duane)
Re: zip: lost interrupt (was Re: Problems booting on AGP G4) (David Haines)
Re: Linux now supports ATA/100... and versus Ultra-160 SCSI ("J. Clarke")
Re: Matrox and Als 100+ (Edward Lee)
Re: 2.2.15 + USB patch hangs on my Athlon (Konstantinos Agouros)
Re: PS2 Mouse + Graphical mode (Mark Bratcher)
Re: New tape drive... Now what? (Mark Bratcher)
Silent Power Supply? ("Me")
Re: Silent Power Supply? (Paul Rubin)
Q&A archived by subject 10 Jun ("K.Tsakaloglou")
Re: Help identify sound card chip set... (Steve Wampler)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (Marshall Blythe)
Subject: Re: Athlon recomendations
Date: Sat, 10 Jun 2000 00:11:34 GMT
On Thu, 08 Jun 2000 05:01:13 +0000, Christopher Segot <[EMAIL PROTECTED]> wrote:
>Hello, I am about to upgrade my system, I am considering an athlon with
>redhat, however on their hcl they say that the athlon may not work
>depending on the system config. Can anybody recomend a good
>mboard,cpu,hd combo ?
>thanks alot.
I've got Redhat 6.2 loaded on the following system:
700 MHz Athlon
Abit KA7 motherboard
128 MB Crucial 133 MHz RAM
ASUS V6800 GeForce DDR video card
300 watt power supply
Adaptec AHA2940 SCSI adapter
3 IBM hard drives: one EIDE (ATA 33) and two SCSI II
US Robotics 33.6 modem (ISA)
The system is stable and bloody fast! I have Xfree86 4.0 installed and tonight
I compiled and installed version 2.2.16 of the kernel (in just 2 minutes!).
--
Marshall Blythe
---> remove 'devnull' from my email address to reply <---
______________________________________________________________________________
------------------------------
From: [EMAIL PROTECTED] (Marshall Blythe)
Subject: Re: Athlon recomendations
Date: Sat, 10 Jun 2000 00:17:36 GMT
On Thu, 08 Jun 2000 21:01:20 +0200, Gero H. Marten <[EMAIL PROTECTED]> wrote:
>Athlons are never used as servers, because they are not 100% i386 compatible.
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
...and the facts backing up this claim may be found where?
--
Marshall Blythe
---> remove 'devnull' from my email address to reply <---
______________________________________________________________________________
------------------------------
From: [EMAIL PROTECTED] (Marshall Blythe)
Subject: Re: Redhat 6.1 compatibility with Abit ka7 or Asus k7V?
Date: Sat, 10 Jun 2000 00:32:09 GMT
On Fri, 09 Jun 2000 19:46:01 GMT, [EMAIL PROTECTED] <[EMAIL PROTECTED]>
wrote:
>I would like to either get the Abit ka7 or the Asus k7v (with k7-650).
>Has anyone had problems installing Redhat 6.1 on either of the above
>motherboards?
I've got the Abit KA7 and Redhat 6.2. The only problem I had was trying to
install and use the system while the "memory hole" was enabled in the BIOS.
This is an option need by some older expansion cards which it turns out I
didn't need, but in my ignorance I left it enabled. What a mess: signal 11's
and filesystem corruption out the ying-yang! It all stopped when I disabled
the memory hole. Stable, fast smooth sailing ever since!
--
Marshall Blythe
---> remove 'devnull' from my email address to reply <---
______________________________________________________________________________
------------------------------
From: [EMAIL PROTECTED] (Dances With Crows)
Subject: Re: Drivers for a) modem b) sound card
Date: 09 Jun 2000 20:42:13 EDT
Reply-To: [EMAIL PROTECTED]
On Fri, 9 Jun 2000 15:34:44 GMT, M. Buchenrieder
<<[EMAIL PROTECTED]>> shouted forth into the ether:
>Jansen <[EMAIL PROTECTED]> writes:
>>My sound card is an integrated chipset (I believe) made my crystal...
>>Windows 95 detects it as 'crystal audio codec'. I've attempted looking
>>for drivers for it, but I've had no luck as yet.
>
>Crystal - isn't that one supported by the OSS drivers ?
Maybe, maybe not. I've seen two very different sound cards (an ES1371 and
a CS4236) detected as "Crystal Audio System Codec" by the Device Mangler.
I'm wondering if it isn't just a catch-all term for "This is some kind of
sound card, and the 3rd-party driver looks like it supports it."
--
Matt G / Dances With Crows \###| You have me mixed up with more
There is no Darkness in Eternity \##| creative ways of being stupid?
But only Light too dim for us to see \#| Beer is a vegetable. WinNT
(Unless, of course, you're working with NT)\| is the study of cool. --MegaHAL
------------------------------
From: Duane <[EMAIL PROTECTED]>
Crossposted-To: alt.comp.hardware,de.comp.os.unix.linux.hardware
Subject: Re: Need your help again...
Date: Fri, 09 Jun 2000 17:04:47 -0700
Pieter Breugelmans wrote:
>
> Yep.. you are absolute correct ... and in my opinion, 30MB/s is quite a lot cos
> other people that i know get around 24Mb/s with UDMA66. Anyway... your PCI-bus is
> running at 33MHz, and this will give you a maximum transferrate of 133MB/s (don't
> ask me any math.. cos i seem to have forgotten it :)...See why the manufactures
> are making the UMDA100? ... just optimizing the speed....
The bus is 32 bits, or 4 bytes wide.
4 bytes per clock, so 4 x 33 = 132MB/s
Or in high end machines you might get a slot that is PCI 66/64 which is
64 bits at 66 MHz:
8 x 66 = 528MB/s
--
My real email is akamail.com@dclark (or something like that).
------------------------------
From: "J T" <[EMAIL PROTECTED]>
Subject: Re: MATSHITA LS120 Ver5
Date: Sat, 10 Jun 2000 01:59:48 +0000
> You should not make a '/ls120' dir, but a '/mnt/ls120' directory.
It shouldn't make any difference where your mount-point is, it is just
some way of referencing the device. I have my mount-point for my LS-120
as '/fd'
> Once you've done that you should be able to mount a pc formatted 1.44
> meg floppy by:
>
> mount -t vfat /dev/hdc /mnt/ls120
>
> You should now see the contents of the disk at /mnt/ls120. umount
> /mnt/ls120, and put a 120 meg disk in it in place of the floppy and
> repeat the above mount command. Again you should be able to see the
> contents of the disk, and in both cases you should be able to write a
> file, and then delete that file from the mounted disk.
>
> It works equally well for those which I've put the ext2 system on by
> leaving out the '-t vfat' option.
>
Sounds about right. This may be stating the obvious but you have compiled
support for 'IDE/ATAPI Floppy Drives' in to the kernel (or as a module)
haven't you?!?
Jay.
------------------------------
From: Duane <[EMAIL PROTECTED]>
Subject: Re: IDE Zip disk faster via ide-scsi ???
Date: Fri, 09 Jun 2000 17:26:19 -0700
Igor Boukanov wrote:
>
> When I enabled SCSI interface emulation to my 100 MB IDE Zip drive
> (via modprobe ide-scsi) I noticed significant performance improvements,
> i.e. "dd if=/dev/zip of=/dev/null" started to run 20% faster with
> zip=/dev/sda4 than it had been with zip=/dev/hdd4.
>
> Are there any reasons for that? I mean if you so easily can improve
> performance by 20% with this trick why it is not on by default???
>
> Regards, Igor
dd speed is very dependent on block size. Try again without ide-scsi,
but use:
dd if=/dev/zip of=/dev/null bs=65536
That will probably also speed things up. Pure guesswork ahead -> Perhaps
the ide-scsi driver uses a different default blocksize from the standard
ide driver.
In general, dd is not a good way to measure performance. I also cannot
really think of a reason why you would want to use it. Linux will do the
same thing using the device file as an argument to "cp". Try this and
compare to the speed of dd:
cp /dev/zip /dev/null
And no, that is not copying the "files" from the zip drive, but is doing
a byte by byte raw device copy, just like dd. That is because /dev/zip
is a raw device, not a filesystem.
This is very handy in these days of large cheap disks. Buy an identical
copy of your root disk, plug it in, boot single user mode, and do "cp
/dev/hda /dev/hdb". It will copy boot partitions, windoze partitions,
Linux partitions, and everything. Stick it on a shelf for the day when
your disk crashes - it is always handy to have a spare disk around
anyway. On my system with 32 bit UDMA33 disk accesses enabled, it took
35 minutes to copy a 20 GB disk.
--
My real email is akamail.com@dclark (or something like that).
------------------------------
Date: Fri, 09 Jun 2000 21:45:10 -0400
Subject: Re: zip: lost interrupt (was Re: Problems booting on AGP G4)
From: David Haines <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.powerpc
in article 8h66n4$7cc$[EMAIL PROTECTED], Billy Foss at [EMAIL PROTECTED]
wrote on 6/1/00 01:32 PM:
>
>
> I noticed that 'lost interrupt' was one of the remaining problems with
> the 2.4 series. It was listed under the 'boot time' failures, and the
> comment from Alan Cox indicated that he did not know which platforms
> were affected.
>
> So it seems this affects x86 platforms with older ATA (not ATAPI) zip
> drives. I know from previous problems with this drive that the
> heads/cylinders/sectors are not the same at the ATAPI version. I was
> looking through the ide kernel source an noticed that 2.2.14+ seems to
> assume that all ide drives can follow some standard that remaps the
> C/H/S to a certian value.
>
> Does anyone know if the PowerPC zip is ATAPI or just ATA? The PowerPC
> also had a BIOS issue that made the problem appear when one upgraded
> the BIOS.
>
> Thanks,
> Billy
>
I think this whole thread was started, when I couldn't boot my G4 using
BootX. The posting that you replied was a kind of commiseration with my
"lost interrupt" experience. What fixed the problem was using Yaboot,
an open-firmware booting utility - it's required for any of the "New World"
Macs that have 9.04 installed.
The Zip drive in my G4 is ATA. I believe that the older PPC Macs that
have (internal) Zip-drives are ATAPI. Not sure about the original G3's or
the "Blue & White" G3's.
------------------------------
Reply-To: "J. Clarke" <[EMAIL PROTECTED]>
From: "J. Clarke" <[EMAIL PROTECTED]>
Crossposted-To: comp.periphs.scsi,comp.sys.ibm.pc.hardware.storage
Subject: Re: Linux now supports ATA/100... and versus Ultra-160 SCSI
Date: Fri, 9 Jun 2000 20:24:57 -0400
"Bob Willard" <[EMAIL PROTECTED]> wrote in message
news:[EMAIL PROTECTED]...
> Michael Stajduhar wrote:
> >
> > The new problem will be the 132MB/s limit of the PCI bus. We'll need a
> > new PCI standard (2x,4x,???), convert to the 64-bit PCI bus, or a new
> > bus (ala AGP).
> >
> The 132 MB/s limit of PCI went away long ago. Starting with the
> v2.1 PCI spec (ca. 1995), these PCI variants have been defined:
>
> 33 MHz, 32bit, 133 MB/s
> 33 MHz, 64bit, 266 MB/s
> 66 MHz, 32bit, 266 MB/s
> 66 MHz, 64bit, 532 MB/s
FWIW, not a lot of boards shipping with these. The 840 boards seem to all
have 66/64, the Intel and SuperMicro GX boards have 66/32 IIRC, the AMI GX
board has 33/64, and quite a few servers do as well.
Looked for a host adapter that supports 66/64, and the only ones I found
were the Compaq 129803-B21 and 154457-B21. Found a bunch that were 33/64:
the Adaptec 29160, 39160, and DPT Century RAID controller; Mylex ExtremeRAID
2000; the IBM ServerRAID-4L, 4M, and 4H; QLogic QLA12160, but no 66/64 RAID
controllers.
, > And, the PCI-X variant doubles this again, to 1064 MB/s. While
> no systems with PCI-X are shipping yet, expect a few systems and
> a few I/O widgets before the end of this year.
>
> Most I/O widgets are 33 MHz, 32bit, widgets. That's because
> most I/O widgets are cost-focused, not performance-focused; the
> don't need much bandwidth. But, for power users and servers, PCI
> is pretty capable.
> --
> Cheers, Bob
--
---
--- John
Reply to jclarke at eye bee em dot net
------------------------------
From: Edward Lee <[EMAIL PROTECTED]>
Subject: Re: Matrox and Als 100+
Date: Fri, 09 Jun 2000 19:23:40 -0700
A typical problem is setting the sound card for 8 bit rather than the 16
bit needed for mp3.
Louis Cyfer wrote:
> Hy guys,can u tell me if Matrox g400 Dual Head is supported under
> Mandrake 7.0?Anyway,I have another problem:my sound card ALS 100+
> can't play mp3.How can I solve it?
> Byeeeeeeeeeeeeeeeeee :-)
------------------------------
From: [EMAIL PROTECTED] (Konstantinos Agouros)
Subject: Re: 2.2.15 + USB patch hangs on my Athlon
Date: 9 Jun 2000 21:27:54 +0200
In <[EMAIL PROTECTED]> Anders Skovsted Buch <[EMAIL PROTECTED]> writes:
>Hi,
>I am using RedHat 6.1 on my Athlon 500, and have recently installed a
>2.2.15
>kernel with the USB backport patch. After that my machine has had
>occational
>lockups. Before it was running rock-solid. Does anybody know about a
>fix? (If
>it matters I have my ATAPI CD-RW configured as an scsi device.)
Hmmm I think you are not alone. I have the same problem running on a HP-Omnibook
4150. It has a PIII.
Can it be, that you have the lockups happening during high memoryload?
Konstantin
>Any help will be greatly appreciated!!
>Anders Buch ([EMAIL PROTECTED])
--
Dipl-Inf. Konstantin Agouros aka Elwood Blues. Internet: [EMAIL PROTECTED]
Otkerstr. 28, 81547 Muenchen, Germany. Tel +49 89 69370185
============================================================================
"Captain, this ship will not sustain the forming of the cosmos." B'Elana Torres
------------------------------
From: Mark Bratcher <[EMAIL PROTECTED]>
Subject: Re: PS2 Mouse + Graphical mode
Date: Fri, 09 Jun 2000 22:41:35 -0400
Stuart Fotheringham wrote:
>
> I have a PS/2 mouse problem: when I run startx, the cursor does
> not appear, the keyboard does not respond, and a reset needs to be
> performed.
>
> Prior to startx, in text mode, the cursor "block" responds to mouse
> movement.
>
> After a number of reboots and power cycles, the problem goes away.
>
> When a serial mouse is used, there is no problem.
It sounds like you may have an interrupt conflict.
The PS/2 mouse uses a specific interrupt (I think it's 12, but double
check that).
It may be that something else in your system is using that interrupt.
--
Mark Bratcher
mailto:[EMAIL PROTECTED]
=========================================================
Escape from Microsoft's proprietary tentacles: use Linux!
------------------------------
From: Mark Bratcher <[EMAIL PROTECTED]>
Subject: Re: New tape drive... Now what?
Date: Fri, 09 Jun 2000 22:47:54 -0400
George Walford wrote:
>
[snip]
> My question is, that now I have this tape drive, now what?
> I need to use it to generate backups on the server, but I have no idea
> how to do that. I took a look at the ftape how to (it does not seem to
> apply) and the man files for tar. They are, to say the least, a bit
> obsfucated.
[snip]
'tar' has already been suggested as one option.
Also, look at 'dump' and 'restore'. These tools provide more traditional
backup features, like keeping track of what was previously backed up.
You can do several backup "levels". The man pages describe this.
--
Mark Bratcher
mailto:[EMAIL PROTECTED]
=========================================================
Escape from Microsoft's proprietary tentacles: use Linux!
------------------------------
From: "Me" <[EMAIL PROTECTED]>
Crossposted-To: alt.comp.hardware,alt.comp.hardware.overclocking
Subject: Silent Power Supply?
Date: Fri, 9 Jun 2000 23:16:45 -0400
Hello, I am looking for a silent power supply for a machine that will be
going with me to college. So far I have narrowed choices to the 300-Watt PS
at www.quietpc.com (which I gander is the same as those offered at
www.silentpc.nl - Mr. Spammer need not tell me to go there unless you can
give me a really good reason to. Your resellers seem a little shady to me)
and the Silencer 275 ATX from www.pcpowercooling.com . Has anybody had any
experience with either of these PSU's? Though the PC Power & Cooling unit
has slightly less power, would I be better off in piece of mind with their
quality? The only issue with them is that their Silencer PS's aren't listed
as AMD approved, should I upgrade that route in the future, and the 400-Watt
unit is way out of my price range. The Quietpc 300-Watt is AMD approved,
however, but I'm not sure at what cost of noise this extra wattage comes at,
or its reliability.
While I'm on silent systems, does anyone have info on the Molex silent
heatsink/fans? Are they really that quiet, but do they provide enough
cooling to dissipate enough heat for, say, a Pentium-III 550E @ 733MHz?
Any info that anybody could provide would be greatly appeciated. Thank you
very much.
Me
------------------------------
From: [EMAIL PROTECTED] (Paul Rubin)
Crossposted-To: alt.comp.hardware,alt.comp.hardware.overclocking
Subject: Re: Silent Power Supply?
Date: 10 Jun 2000 03:46:24 GMT
In article <8hsc2k$al5$[EMAIL PROTECTED]>,
Me <[EMAIL PROTECTED]> wrote:
>Hello, I am looking for a silent power supply for a machine that will be
>going with me to college. So far I have narrowed choices to the 300-Watt PS
>at www.quietpc.com (which I gander is the same as those offered at
>www.silentpc.nl - Mr. Spammer need not tell me to go there unless you can
What's this about shady resellers and spam? I'm thinking of getting
one of their disk sleeves (probably from the US reseller, to save on
shipping) but don't want to buy from spammers.
>give me a really good reason to. Your resellers seem a little shady to me)
>and the Silencer 275 ATX from www.pcpowercooling.com . Has anybody had any
>experience with either of these PSU's? Though the PC Power & Cooling unit
>has slightly less power, would I be better off in piece of mind with their
>quality?
I have the AT version of the Silencer and it is great. It appears though
to be a fairly normal P/S with a quieter-than-usual fan. PC P&C will
sell these fans for around $10 if you want to put one in another P/S.
>The only issue with them is that their Silencer PS's aren't listed
>as AMD approved, should I upgrade that route in the future, and the 400-Watt
>unit is way out of my price range. The Quietpc 300-Watt is AMD approved,
>however, but I'm not sure at what cost of noise this extra wattage comes at,
>or its reliability.
What the heck is AMD Approved about? If AMD made CPU's that were
incompatible with standard P/S's, that would be the end of them. So I
wouldn't worry about this issue.
>While I'm on silent systems, does anyone have info on the Molex silent
>heatsink/fans? Are they really that quiet, but do they provide enough
>cooling to dissipate enough heat for, say, a Pentium-III 550E @ 733MHz?
This is probably asking for trouble.
------------------------------
From: "K.Tsakaloglou" <[EMAIL PROTECTED]>
Subject: Q&A archived by subject 10 Jun
Date: Sat, 10 Jun 2000 06:54:13 +0300
Questions and answers from this newsgroup (archived by subject) can be found
at http://server.hellug.gr/LUGistics/en/pub/QA_articles_main.php3
Links suggested in those messages are categorized at
http://links.hellug.gr
K.Tsakaloglou
[EMAIL PROTECTED]
------------------------------
From: Steve Wampler <[EMAIL PROTECTED]>
Subject: Re: Help identify sound card chip set...
Date: Fri, 09 Jun 2000 20:54:02 -0700
Dances With Crows wrote:
>
> On Fri, 09 Jun 2000 07:30:56 -0700, Steve Wampler
> <<[EMAIL PROTECTED]>> shouted forth into the ether:
> >
> >I recently ordered a new computer and installed Linux on
> >it - now I need help getting the sound card to work.
> >The packing slip says the card is a "Skywalker Yamaha
> >PCI 3D Sound W/Wave" - but, since the packing slip was
> [snipp]
> >If you recognize this card and know the Linux driver to
> >use with it, please let me know!
>
> Two things to try:
> 0. "cat /proc/pci" will tell you a lot of info about the stuff on the PCI
> bus. Since the soundcard is PCI, this will give you some info about what
> the kernel thinks the chipset on the card is.
Hey, this is getting fun, in a perverted sort of way...
/proc/pci reports that this is an "OPTi unknown device" (Rev 16),
with vendor ID 1045 and device ID c935 (not even close to anything
that windows reports, but a visit to the OPTi site reveals that
they sold off their audio chip devision several years ago (the last
audio chip they mention is the 82c931, which is suspiciously close
to the c935 device ID...). opensound.com doesn't mention this
chip (or anything for OPTi) in there lists, but the alsa project
does have an opti93x driver. I guess that's what I'll try next.
I also stumbled on a way to really identify the actual card manufacturer
that others may find useful (not that many people seem to have this
problem):
Locate the FCC ID: printed on the card and visit the site:
http://www.fcc.gov/oet/fccid
You can locate the manufacturer that was granted that FCC ID.
I'm going to try that the next time I'm close enough to that computer
(it's there, and I'm here) to pull the card...
Of course, I'd love to know where the "Mach 1 PCI Sound System" was
coming from...
--
---
Steve Wampler {[EMAIL PROTECTED]}
The gods that smiled upon your birth are laughing now. -- fortune cookie
------------------------------
** 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
******************************
