Linux-Hardware Digest #436, Volume #12 Wed, 8 Mar 00 20:13:13 EST
Contents:
Re: VIA vs Intel chipsets - which is better? ("Neil Davis")
Re: Idiot of the Year for c.s.i.p.h.c [was Re: VIA vs Intel chipsets - which is
better? (Jim Cochrane)
Re: VIA vs Intel chipsets - which is better? (tjasz)
Re: best graphics card? (Piercarlo Grandi)
Re: Odd cua0 vs. ttyS0 bug from gpm w/kernel 2.2.14 (m buller)
Re: SCSI? IDE? Opinions please (Greg Leblanc)
Re: VIA vs Intel chipsets - which is better? (Sebastian)
Re: best graphics card? (Dan Law)
Re: HP cdwriter (Greg Leblanc)
Re: Impasse with 2 SCSI controllers, kernel mods required? (teri)
Re: VIA vs Intel chipsets - which is better? (tjasz)
----------------------------------------------------------------------------
From: "Neil Davis" <[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: Wed, 8 Mar 2000 19:14:08 -0500
I'm sorry to see this thread deteriorate so quickly into name-calling,
because there are some legitimate chipset and driver questions I'd love to
see addressed. I've put together a number of VIA-based systems, but
recently I got fed up with them and have switched to the BX boards instead.
One of the problems I've suffered with the VIA chipset is IRQ sharing. It's
easy to end up with nicely configured systems that have more peripherals
than available IRQ's, and my experience is that the VIA IRQ miniport driver
doesn't allow IRQ sharing as well as the BX boards. I've seen VIA boards
share IRQ's, but it's not the norm, and I haven't figured out how to get it
to work as repeatably as the BX boards. For example, one of my machines has
video, TV-tuner, video capture, SCSI, MODEM, and Soundblaster Live (requires
2 IRQ's), USB, two IDE's, and they are all shared sociably on a BX board
(one IRQ actually has 4 devices, and one IRQ is still listed as free). So
question #1: is there something in the design of the IRQ hardware in the
Intel chipset that allows it to work better with IRQ-sharing drivers, or is
Intel just better at writing the IRQ-sharing driver, or is my experience
with IRQ sharing not consistent with what others have seen?
The second problem I've had with VIA boards is poor support for certain
devices using their busmastering drivers. For example, the Nakamichi 5x16
CD changer hangs up when you try to play audio CD's (using the drivers from
VIA that were available last month). I also had problems with one of the
Creative DVD players--lots of dropouts even when busmastering enabled,
although the Hitachi G2500 DVD player worked perfectly on the same computer.
So question #2: is there something unique about the VIA IDE hardware that
makes it difficult to get compatibility with more devices, or is this just a
software maturity problem, and when will VIA finally get it right?
My recommendation to potential motherboard buyers has been to go with the
lower-cost VIA boards if they aren't going to have lots of devices that
require IRQ's or if they are going to stick with IDE devices that are known
to work with the VIA drivers. The VIA boards are great for people who don't
experiment much with different peripherals--I believe this is exactly what
Michael Dell was referring to when he recently said: "We found the AMD
environment to be much more fragile ... than equivalent Intel systems." I'm
not interested in starting a flame war--I'm curious whether others had the
same negative experiences with the VIA chipset. Even more to the point, I
am currently interested in buying an Athlon system, but I'm gun-shy about
going with the KX-133 chipset. Any reports on how well it supports IRQ
sharing?
Dean_Kent <[EMAIL PROTECTED]> wrote in message
news:#41yLVMi$GA.96@cpmsnbbsa02...
> Jim Cochrane <[EMAIL PROTECTED]> wrote in message
> news:8a4l62$[EMAIL PROTECTED]...
> > While calling a couple computer shops today to look into upgrading my
> > current system to a PIII motherboard, I came across an interesting
> > dilemna - essentially, two shops I talked to gave two different
> > opinions on chipsets available for Pentium III motherboards.
>
> Let me guess...
>
> >
> > I'm posting here to get y'all's opinions on these issues, but I think
> > it may also serve as an interesting (and possibly controversial) topic
> > of discussion. (Perhaps it has already been covered here and is old
> > hat; but unfortunately, my main machine is in the shop and I can't
> > access the web to look at dejanews with my 486.)
>
> Um, I'll bet one said VIA is buggy and the other said VIA is not-so-bad...
>
> >
> > The disagreement is on the quality of the non-Intel VIA chipset
> > versus the quality of the Intel chipsets. According to one shop, the
> > VIA chipset is much more buggy than the Intel chipsets (BX3 and I820, I
> > believe) and that this can potentially cause problems running a Linux
> > kernel (my intended OS). The fellow with this point of view stated
> > that the kernel has a lot of patches applied to work around bugs in the
> > VIA chipset. He acknowledged that the Intel chipsets had some bugs,
> > but not nearly as many as the VIA chipset.
> >
> > The fellow with the opposite opinion stated that this was false. He
> > essentially stated that there may be some bugs in both the VIA chipset
> > and the Intel chipsets, but that these bugs should not affect the linux
> > kernel and thus will not cause problems. (He also said he had
> > installed Linux on several such systems (VIA) and had not run into any
> > problems with respect to the chipset.)
>
> Yup. Sounds like one of the religious wars fought here in the past...
>
> >
> > Which of these opinions is correct? Or are they both partly right and
> > partly wrong? (With respect to my hardware upgrade, the questions, of
> > course, boils down to whether to purchase a VIA-based board, such as
> > Tyan, or an Intel-based board, such as a Supermicro.)
>
> Here is my opinion - VIA had some problems with AGP implementation (don't
> know all the details) with their SS7 chipset, but worked them out
> eventually, AFAIK. Unfortunately, once a reputation is gained, it is
near
> impossible to shake (just ask AMD). The current crop of Apollo Pro
> chipsets (133 and 133A) also had some AGP issues, but that was before they
> were released to the public (OEM samples only). These have been all but
> completely fixed. The only issue left is that the video transfer rates
are
> slightly slower than Intel's (BX and i820).
>
> When you consider that almost 50% of all Pentium II/III motherboards
shipped
> today to OEMs are VIA Apollo Pro based, it should tell you something...
>
> Now, here are your choices, IMO:
>
> Intel 440BX - tried and true. Currently fastest of all chipsets, overall.
> Limited to 2x AGP and UDMA/33 (though some manufacturers have 3rd party
IDE
> controllers to add UDMA/66). Officially supports only 66MHz and 100MHz
> FSB, and takes PC100 SDRAM (though many manufacturers implement 133MHz
FSB -
> some with 1/4 PCI divisor). If overclocking, AGP speed is 2/3 of CPU
speed,
> so that could cause problems.
>
> Intel i820 - almost as fast as the BX and has 4x AGP and UDMA/66 support.
> Supports 133MHz FSB. Unfortunately, it only accepts RDRAM in it's
'natural'
> state. To support SDRAM, it requires a 'Memory Translator Hub' chip,
which
> slows down perfomance and only officially supports 100MHz FSB. More
> expensive than BX solutions.
>
> VIA Apollo Pro133 or Pro133A - Almost as fast as BX (about the same as the
> i820, even at 100MHz FSB). Supports AGP 4x and UDMA/66. Also
officially
> supports 133MHz FSB, and has a better 'PCI/AGP divisor' scheme for
> overclockers. If the system is primarily for 3D games, it will be
slightly
> slower than a BX based board because the video transfer rate is a bit
> slower. Depending upon application, it could be 5% to 10% slower than BX
> based board.
>
> Can't speak to the Linux issue. I can't imagine why it would be a
problem,
> considering Intel licensed the P6 design to VIA so there shouldn't be any
> compatibility problems to speak of...
>
> >
> > [For those interested, I have also posted a question about another issue
> > I came across with the same two shops - see the subject line "PIII vs
PIII
> > E - which is faster?"]
>
> I'll render an opinion on that here as well...
>
> The Coppermine processors have a 256K full speed L2 cache that is 'wider'
> than the Klamath processors 512K 1/2 speed cache. It has a higher initial
> latency, but better overall throughput. Net result is that the Cu
> processors are faster. FSB is the same (memory to CPU), it is the cache
> speed that is different (BSB).
>
> Intel went to Socket 370 for Celeron early, because that was the 'low-end'
> processor, and Socket 370 saves bucks over the Slot 1 design. They are
now
> going to Socket 370 for *all* processors - and very soon, reportedly.
> PIII processors using Socket 370 have a lower voltage than the Celerons
> (1.6v vs. 2.0 v, I believe).
>
> >
> > [Apologies for any errors in terminology - as you can tell, I'm not a
> > hardware dude.]
>
> I'm sure that others will give their input, and correct any errors I may
> have made (as well as give any opinions that differ from mine)... :-)
>
> Regards,
> Dean
>
> > --
> > Jim Cochrane
> > [EMAIL PROTECTED]
>
>
------------------------------
From: [EMAIL PROTECTED] (Jim Cochrane)
Crossposted-To: comp.sys.ibm.pc.hardware.chips,comp.sys.ibm.pc.hardware.systems
Subject: Re: Idiot of the Year for c.s.i.p.h.c [was Re: VIA vs Intel chipsets - which
is better?
Date: 8 Mar 2000 17:21:30 -0700
Sorry - didn't mean to start a flame war.
But I'd like to thank those that posted responses to my questions. Your
responses were very helpful and informative.
Thanks -
--
Jim Cochrane
[EMAIL PROTECTED]
------------------------------
From: tjasz <[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: Wed, 08 Mar 2000 15:54:00 -0800
Reply-To: nospam@dontbother
ignore the stalking ninny...
On Wed, 08 Mar 2000 17:07:43 GMT, "Ron Reaugh"
<[EMAIL PROTECTED]> wrote:
This guy sells VIA and is....blind.
------------------------------
Crossposted-To: alt.os.linux,comp.os.linux.x,uk.comp.os.linux
Subject: Re: best graphics card?
From: [EMAIL PROTECTED] (Piercarlo Grandi)
Date: Thu, 09 Mar 2000 00:38:19 GMT
>>> On Mon, 06 Mar 2000 13:53:35 +0000, Dan Law <[EMAIL PROTECTED]> said:
Dan> Can anyone recommend a graphics card supported by Linux (well,
Dan> XFree86) that would have a high quality mpeg decoder that can be
Dan> used in DVD playback on a pc monitor? [ ... ] My computer is a
Dan> PIII 450 running RH 6.1. Also, all of my DVDs are region 1 & I
Dan> live in the UK, just to make the problem more difficult... [ ... ]
Well, a PIII 450 can easily do software only decode, you don't really
need any special purpose hardware assist. Cinemaster for Windows (in its
many incarnations) seems to be the most efficient DVD software decoder.
You also have the TV version of the Voodoo3, so you are sorted on that
front too.
As to (in your case, probably pointless) cards with hardware DVD assist,
those that have it are allegedly the ATI 128, Matrox G400 and Savage4.
I am not sure whether the GeForce has got it too, the TNT/TNT2 haven't
or have minimal support.
The ATI 128, G400 and Savage 4 all have Linux X servers; the best one is
the G400 one. The G400 also is one of the few cards supported right now
for OpenGL 3D acceleration under Linux. This might suggest having a look
at the G400. The Savage 4 and many ATI 128 models are however usually
much cheaper.
------------------------------
From: m buller <[EMAIL PROTECTED]>
Crossposted-To: comp.os.linux.development.system
Subject: Re: Odd cua0 vs. ttyS0 bug from gpm w/kernel 2.2.14
Date: Thu, 09 Mar 2000 00:30:20 GMT
Hi Peter,
Thanks for the info. You are correct. The major # for the Digi ports
ttyD0-ttyD16 are at 22, the same as cud0-cud31. The funny thing is the rest
of the defined ports (as created by the digi rpm install) are at major 23.
Also the creation date for D0-D16 is April 17, 1999, for D17-D31 and cud0 -
cud31 is March 5, 2000 the date of my digi rpm install. It appears that the
digi install program works correctly but there are predefined (by redhat ?)
ttyD0-ttyD17 devices that do not get over written. Now I need to find out
about device creation. Is it safe to just delete ttyD0-ttyD17 and use
mknod to recreate them? Or is there more to it then that ? Know of any good
howto's ?
Thanks again
H. Peter Anvin wrote:
>
> Followup to: <[EMAIL PROTECTED]>
> By author: m buller <[EMAIL PROTECTED]>
> In newsgroup: comp.os.linux.development.system
> >
> > I checked the version of getty (getty_ps 2.0.7j-7) looks like it is the
> > latest. I contacted Digi tech support they suggested I try deleting
> > cud0. I did this, but still get the same message. Note the port (dev)
ttyD0
> > works fine.
> >
> > Any Ideas ?
> >
>
> You probably have the majors on /dev/cud* and /dev/ttyD* reversed.
> Digiboard have had problems with it... at least some versions of the
> drivers put the smaller major as /dev/cud*, which violates the
> invariant that you find the callout device by adding 1 to the major of
> the callin one.
>
> -hpa
> --
> <[EMAIL PROTECTED]> at work, <[EMAIL PROTECTED]> in private!
> "Unix gives you enough rope to shoot yourself in the foot."
--
Posted via CNET Help.com
http://www.help.com/
------------------------------
From: Greg Leblanc <[EMAIL PROTECTED]>
Subject: Re: SCSI? IDE? Opinions please
Date: Thu, 09 Mar 2000 00:44:30 GMT
In article <[EMAIL PROTECTED]>,
"Lyle R. Taylor" <[EMAIL PROTECTED]> wrote:
> It may depend on what you plan to do with your system. If you plan to
> perform tasks that are very disk intensive, then you may want SCSI,
> because it handles it better. If you plan to do normal, everyday
tasks
> (web browsing, etc.), then get IDE, because it's much cheaper. I do a
> lot of development, and SCSI seems to work better for me (if only I
could
> afford to replace my crashed SCSI drive! - was a problem with the
brand,
> not SCSI). If you're going to run a server that will be in high
demand,
> definately go with SCSI.
That pretty much sums it up.
>
> I also think you will run into less headaches if you use SCSI for
other
> peripherals like tape drives, etc. Especially since you can only have
4
> IDE devices hooked up to your motherboard (I really wish they would
> change that, it's a really stupid and unnecessary limitation).
There are several motherboards that I've seen with 4 IDE channels (which
means 8 drives, if you're out of your mind). I usually try to run just
1 IDE drive per channel, unless they're just there to have large
ammounts of space, and I don't care how slow they go.
Greg
--
It's pronounced "sexy" not "scuzzy"!
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: Sebastian <[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: Thu, 09 Mar 2000 00:55:19 GMT
John Howland wrote:
>
> Think MHz, not timing - it runs at full CPU MHz speed.
OK. So it runs at full core speed but is capable of transfering data every
second clock tick. So while it's maybe clocked at full core speed (it probably
takes clock puleses form the same internal pulse generator (pumped by external
clock at FSB rate) as whole rest of the chip) but it behaves exactly as half
speed cache with 256bit bus and low latency (~8 cycles).
That all full core speed vs half core speed becomes quite meaningless when
caches start to significantly differ in bus width, latency and all that mess.
Regards
--
Sebastian Kaliszewski
--
"Never underestimate the power of human stupidity" - from Notebooks of L.L.
------------------------------
From: Dan Law <[EMAIL PROTECTED]>
Crossposted-To: alt.os.linux,comp.os.linux.x,uk.comp.os.linux
Subject: Re: best graphics card?
Date: Thu, 09 Mar 2000 00:42:26 +0000
Piercarlo Grandi wrote:
> >>> On Mon, 06 Mar 2000 13:53:35 +0000, Dan Law <[EMAIL PROTECTED]> said:
>
> Dan> Can anyone recommend a graphics card supported by Linux (well,
> Dan> XFree86) that would have a high quality mpeg decoder that can be
> Dan> used in DVD playback on a pc monitor? [ ... ] My computer is a
> Dan> PIII 450 running RH 6.1. Also, all of my DVDs are region 1 & I
> Dan> live in the UK, just to make the problem more difficult... [ ... ]
>
> Well, a PIII 450 can easily do software only decode, you don't really
> need any special purpose hardware assist. Cinemaster for Windows (in its
> many incarnations) seems to be the most efficient DVD software decoder.
> You also have the TV version of the Voodoo3, so you are sorted on that
> front too.
>
> As to (in your case, probably pointless) cards with hardware DVD assist,
> those that have it are allegedly the ATI 128, Matrox G400 and Savage4.
> I am not sure whether the GeForce has got it too, the TNT/TNT2 haven't
> or have minimal support.
>
> The ATI 128, G400 and Savage 4 all have Linux X servers; the best one is
> the G400 one. The G400 also is one of the few cards supported right now
> for OpenGL 3D acceleration under Linux. This might suggest having a look
> at the G400. The Savage 4 and many ATI 128 models are however usually
> much cheaper.
This is the best advice I've had. Thanks VERY much!
-Dan
------------------------------
From: Greg Leblanc <[EMAIL PROTECTED]>
Subject: Re: HP cdwriter
Date: Thu, 09 Mar 2000 00:48:16 GMT
In article <[EMAIL PROTECTED]>,
Filippo <[EMAIL PROTECTED]> wrote:
> I'm going to buy an HP 9210 scsi cdwriter. I'm attaching it to a
2940U2w
> scsi controller Do you know of any problems under linux?
There shouldn't be any problems. The 2940U2W is a good card, and I've
had good luck with my HP drives.
Greg
--
It's pronounced "sexy" not "scuzzy"!
Sent via Deja.com http://www.deja.com/
Before you buy.
------------------------------
From: [EMAIL PROTECTED] (teri)
Crossposted-To:
comp.os.linux.development.system,comp.os.linux.setup,alt.os.linux.caldera
Subject: Re: Impasse with 2 SCSI controllers, kernel mods required?
Date: Wed, 8 Mar 2000 20:01:20 EST
In article <[EMAIL PROTECTED]>,
John Belew <[EMAIL PROTECTED]> wrote:
>I have a similar problem (Adaptec 7880 and 2940UW adapters) with a
>RedHat 6.1 installation.
>
>RedHat describes the problem at:
>
> http://www.redhat.com/support/docs/gotchas/6.1/gotchas-6.1-6.html#ss6.27
>
>with the words
>
> Systems with multiple SCSI cards will find that the SCSI modules
> have been loaded in the opposite order than specified. This can
> cause drives letters to change due to Linux's SCSI way of dealing
> with drives.
>
>but I followed their suggested fix, without solving the problem.
Well, considering that their suggestion is a fixed RPM package and I'm
not running RedHat, I think I won't risk damaging my system, in an area
as critical as the boot process with non-caldera packages.
Furthermore, I have to point out that the boot order problem I see is when
both drivers (the 2940UW with the boot drive attached to it, and the 1542CF
that only has a scanner on it) are *compiled into the kernel*.
In this case the kernel repeatedly tries to acces disk 0 on the wrong
controller, and prints plenty of messages while doing so. This is with
a freshly compiled 2.2.14.
When I try to load the 1542CF driver is when I get the device busy error.
When the 1542CF driver is loaded (as a module) at boot time the boot
process locks up, but there are no messages indicating that the kernel
is doing anything. This is with caldera's kernel (2.2.10)
Anyway, thanks for taking the time to respond.
------------------------------
From: tjasz <[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: Wed, 08 Mar 2000 15:56:27 -0800
Reply-To: nospam@dontbother
not a chance...not even I am that gullible...LOL
On Wed, 8 Mar 2000 11:54:31 -0800, "John Howland"
<[EMAIL PROTECTED]> wrote:
I run across him on a regular basis... my concern is someone who
doesn't know better may actually
believe him.
------------------------------
** 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
******************************
