Linux-Hardware Digest #688, Volume #10 Wed, 7 Jul 99 01:13:27 EDT
Contents:
Re: Celeron, what's the catch? (kls)
Re: Celeron, what's the catch? (L.Angel)
Where to purchase Linux modems? ("ZombieSeed")
Re: Celeron, what's the catch? (L.Angel)
Re: Celeron, what's the catch? (L.Angel)
Re: Celeron, what's the catch? (Bud (the original))
Dual PPro SMP problems (Steven M. Gallo)
Re: Using the HP 722C DeskJet Printer in Redhat 6.0 (Larry Ozarow)
Re: Celeron, what's the catch? (Michael)
Re: To RAID or not to RAID? -that is the question... (Paul Colquhoun)
Re: Celeron, what's the catch? (Chris Robato Yao)
Re: *** Need modem recommendation for LINUX (wizard)
Re: Celeron, what's the catch? (Anthony Hill)
Re: Celeron, what's the catch? (Anthony Hill)
Re: modem mystery (Kenneth Been)
----------------------------------------------------------------------------
From: [EMAIL PROTECTED] (kls)
Crossposted-To: comp.sys.ibm.pc.hardware.chips,comp.sys.intel
Subject: Re: Celeron, what's the catch?
Date: Wed, 07 Jul 1999 02:28:12 GMT
In article <7lucq8$gi4$[EMAIL PROTECTED]>, [EMAIL PROTECTED] says...
>
>Andrzej Popowski ([EMAIL PROTECTED]) wrote:
>
>> These are times for compiling Linux kernel with different CPU set:
>
>> CPU time in seconds
>> Celeron 450 + P2 350 273
>> Cel. 450 + Cel. 550 233
>> Celeron 550 + P2 350 223
>
>Okay, I'm a little perplexed by these; not the benchmarks, but the platforms
>being tested. Are you saying you have dual CPU setups with asynchronous
CPUs?
>How was this accomplished?
I don't know what he was using but Abit's bp6 supplies seperate voltages
so you can oc each cpu seperatly.
------------------------------
From: a?n?g?e?l?@lovergirl-DOT.com (L.Angel)
Crossposted-To: comp.sys.ibm.pc.hardware.chips,comp.sys.intel
Subject: Re: Celeron, what's the catch?
Date: Wed, 07 Jul 1999 01:34:40 GMT
Reply-To: ?a?n?g?e?[EMAIL PROTECTED]
>> I have not see any benchmark or website that suggests a .55 ratio for
>> the P2 FPU vs. K6. You made this up. Back it up.
^^^^
>> On the contrary, there is plenty that suggests otherwise. Read the
>> K6-III review of http://www.combatsim.com. According to them, a K6-III
>> 450 is barely equivalent to a PII 400 on non 3DNow games.
>I am terribly confused. If a P2/400 is very roughly the same speed as a
>K63/450, _this_ is a .55 ratio? Really? My calculator sez 400/450 =
>~.889. What am I doing wrong?
The .55 ratio is for pure FPU comparisons. But in real apps which
according to the rest of the guys are hardly ever mostly FPU only,
the cache and integer performance of the K6-3 allows it to close the
gap quite a bit.
The little lost angel & her featherhead's 2 cents of dreaminess. :)
Email : Figure out what to remove, I'm getting tired of spam
------------------------------
From: "ZombieSeed" <[EMAIL PROTECTED]>
Subject: Where to purchase Linux modems?
Date: Tue, 6 Jul 1999 22:39:22 -0400
I basically give up on the current modem I have (PnP). Where could I buy a
Linux compatible modem (one with the jumpers where you set the IRQ and COM
yourself)? Any place online?
All the computer stores around here have nothing but Winmodems.
Thanks in advance
------------------------------
From: a?n?g?e?l?@lovergirl-DOT.com (L.Angel)
Crossposted-To: comp.sys.ibm.pc.hardware.chips,comp.sys.intel
Subject: Re: Celeron, what's the catch?
Date: Wed, 07 Jul 1999 01:44:16 GMT
Reply-To: ?a?n?g?e?[EMAIL PROTECTED]
>We're talking AMD here, not Intel :-) In fact Ironside is out and
I thot it was the Irongate? :P
The little lost angel & her featherhead's 2 cents of dreaminess. :)
Email : Figure out what to remove, I'm getting tired of spam
------------------------------
From: a?n?g?e?l?@lovergirl-DOT.com (L.Angel)
Crossposted-To: comp.sys.ibm.pc.hardware.chips,comp.sys.intel
Subject: Re: Celeron, what's the catch?
Date: Wed, 07 Jul 1999 01:43:48 GMT
Reply-To: ?a?n?g?e?[EMAIL PROTECTED]
>PS: The ratio will be approx. 4.5:1 for K7-600:PII-233
>:-)))) <- AMD fan watching Intels ass being kicked ...
You can't say tat based on a K7-600 vs a P2-233 :P
I normalized my figures by dividing by their clock speed to be fair.
For this comparison of K7 vs P2
K7 is 46% faster in FPU compared to the P3
P3 is 1% faster than P2 (according to Intel)
So the K7 should be about 47.4% faster than a P2 Mhz for Mhz.
The situation we see with the K6 vs P2 will be reversed :P
The little lost angel & her featherhead's 2 cents of dreaminess. :)
Email : Figure out what to remove, I'm getting tired of spam
------------------------------
From: [EMAIL PROTECTED] (Bud (the original))
Crossposted-To: comp.sys.ibm.pc.hardware.chips,comp.sys.intel
Subject: Re: Celeron, what's the catch?
Date: Wed, 07 Jul 1999 04:10:12 GMT
On Wed, 07 Jul 1999 01:43:48 GMT, a?n?g?e?l?@lovergirl-DOT.com (L.Angel)
wrote:
:You can't say tat based on a K7-600 vs a P2-233 :P
:I normalized my figures by dividing by their clock speed to be fair.
:
:For this comparison of K7 vs P2
:K7 is 46% faster in FPU compared to the P3
:P3 is 1% faster than P2 (according to Intel)
:So the K7 should be about 47.4% faster than a P2 Mhz for Mhz.
:The situation we see with the K6 vs P2 will be reversed :P
You have to remember, it was NOT a 'normal' P3, but a P3-Xeon that was
compared to the K7.
The regular P3 has half-CPU speed cache, the Xeon has full CPU speed cache.
TO reply via e-mail, change the I in the address to an A.
CA antispam laws went into effect Jan 1, 1999.
Violators may be liable for both fines AND jail time for spamming this account.
------------------------------
From: [EMAIL PROTECTED] (Steven M. Gallo)
Subject: Dual PPro SMP problems
Date: 7 Jul 1999 04:00:39 GMT
Hi,
I'm running RedHat 6.0 with a re-compiled 2.2.10 kernel. It appears that
both CPUs are found and enabled but only one of them seems to be getting
any of the work scheduled on it. I've selected SMP, as well as MTRR support.
My hardware is as follows:
IWill dp6ns motherboard with onboard aix7880 (ultra/wide SCSI)
2 PPro 166 cpus
128M memory
/proc/cpuinfo shows 2 CPUs found, but if I look at /proc/interrupts
I see that the timer for CPU1 is not getting any hits:
CPU0 CPU1
0: 258691 0 XT-PIC timer
1: 1916 2151 IO-APIC-edge keyboard
2: 0 0 XT-PIC cascade
5: 0 3 IO-APIC-edge soundblaster
7: 202747 1102 IO-APIC-edge eth0
12: 41546 49033 IO-APIC-edge PS/2 Mouse
13: 1 0 XT-PIC fpu
14: 5 2 IO-APIC-edge ide0
17: 12293 12778 IO-APIC-level aic7xxx
NMI: 0
ERR: 0
When running 2 CPU-intensive processes, top shows that each is utilizing
only 50% of the CPU. When I was running RH5.2, top showed each process taking
upwards of 95% of the CPU.
738 smgallo 19 19 12620 12M 140 R N 0 49.1 9.7 13:19 setiathome
739 smgallo 20 19 13588 13M 140 R N 0 47.7 10.5 13:25 setiathome
Am I missing something? How can I get the processes to utilize the 2nd CPU?
Steve
>From DMESG:
Intel MultiProcessor Specification v1.1
Virtual Wire compatibility mode.
OEM ID: OEM00000 Product ID: PROD00000000 APIC at: 0xFEE00000
Processor #0 Pentium(tm) Pro APIC version 17
Processor #1 Pentium(tm) Pro APIC version 17
I/O APIC #2 Version 17 at 0xFEC00000.
Processors: 2
mapped APIC to ffffe000 (fee00000)
mapped IOAPIC to ffffd000 (fec00000)
Detected 166194339 Hz processor.
Console: colour VGA+ 80x25
Calibrating delay loop... 165.48 BogoMIPS
Memory: 127672k/131072k available (1324k kernel code, 416k reserved, 1592k data,
68k init)
Checking 386/387 coupling... OK, FPU using exception 16 error reporting.
Checking 'hlt' instruction... OK.
POSIX conformance testing by UNIFIX
mtrr: v1.35 (19990512) Richard Gooch ([EMAIL PROTECTED])
per-CPU timeslice cutoff: 50.05 usecs.
CPU0: Intel Pentium Pro stepping 02
calibrating APIC timer ...
..... CPU clock speed is 166.1934 MHz.
..... system bus clock speed is 66.4771 MHz.
Booting processor 1 eip 2000
Calibrating delay loop... 165.89 BogoMIPS
OK.
CPU1: Intel Pentium Pro stepping 02
Total of 2 processors activated (331.37 BogoMIPS).
enabling symmetric IO mode... ...done.
ENABLING IO-APIC IRQs
init IO_APIC IRQs
IO-APIC pin 0, 20, 21, 22, 23 not connected.
..MP-BIOS bug: 8254 timer not connected to IO-APIC
...trying to set up timer as ExtINT... .. (found pin 0) ... works.
number of MP IRQ sources: 21.
number of IO-APIC registers: 24.
------------------------------
From: [EMAIL PROTECTED] (Larry Ozarow)
Subject: Re: Using the HP 722C DeskJet Printer in Redhat 6.0
Date: Wed, 07 Jul 1999 03:50:18 GMT
In article <[EMAIL PROTECTED]>,
Rob Clancy <[EMAIL PROTECTED]> writes:
> Can anyone provide me with either the location of the driver or the
> methods for using an existing driver to use the HP 722C DeskJet Printer
> in redhat 6.0? I've tried all the 500 and 600 series drivers that come
> with the operating system and haven't had any luck yet. Thanks in
> advance.
>
> Rob
> ([EMAIL PROTECTED])
>
The 722c is a "winprinter," and responds only to a fairly low-level
proprietary language that HP has not released specs for.
You can find a partial solution that'll enable you to print postscript
and anything reducible to postscript (at least in black-and-white) at
http://www.httptech.com/ppa/.
Larry
------------------------------
From: [EMAIL PROTECTED] (Michael)
Crossposted-To: comp.sys.ibm.pc.hardware.chips,comp.sys.intel
Subject: Re: Celeron, what's the catch?
Date: Wed, 07 Jul 1999 03:26:40 GMT
Jeffrey Karp <[EMAIL PROTECTED]> wrote:
>
>
>So what. If you want a gaming machine, then get an Athlon(AMD K7).
>The Athlon 550 has 46% FASTER FPU than a Pentium 3 550.
>http://www.amd.com/products/cpg/athlon/benchmarks.html
>It is also lower priced. I know, you want a new system NOW, and don't
>want to wait a few weeks for the Athlon. The K6-3 is a great CPU
>for business software, but not so great for games that are not
>optimized for 3D now, or for other heavy FPU.
>Choose the best tool for the job.
>
Amen to that. That's the problem in all these discussions. As a
business user, the K6 III runs rings around the Celeron 466 on the way
I use Windows. That's because of caches.....I can slow down the k6III
to Celeron level by turning off LII caches. The celeron is a good
processor for math oriented needs, and gamers who want general
versatility. As a general windows use, and from that perspective
only, the K6 III (not the II) is a fantastic processor. The celeron
shows it best not at multitasking but at single tasking, benchmarks be
damned. I know how long I have to wait to switch programs, watch how
quick IE 5.0 renders when I have multiple windows open, etc.
Gamers have different needs then general business users. That's why
the video card issues are not exciting to me... I could care less how
many times I have Kill per second in Quake (ok so I am using Hyperbole
here). But if how fast you can kill someone in a game is important to
you, or how fast you can drive on a computer screen is really a big
issue, then you need to balance your needs against slower
multitasking. If you use HEAVY filters in photoshop, then you need to
examine what setup gives you the best usage. I look at the processor
not as what's best for compiling, or fast scientific usage, or fast
gaming, but how well windows, the consumer versions run when I am
doing mail, spreadsheets, on the web, printing, all at the same time.
The celeron cannot give me the level of performace in these tasks that
the K6 III can, in fact, the k6 II can't either. The lack of big
caches really shows when you constantly flip back and forth.
Everything is instant on the k6III, and takes a second or more on the
celeron. The save dialogue boxes come up instantly, versus a second
or the celeron as an example. This is important to me, but not
someone else. As you said, choose the best processor for the job.
Intel has no monopoly on that, and neither does AMD. Quite frankly,
the level of FPU doesn't seem to impact most common users, except
gamers. But don't say a celeron is faster than an AMD k6 III in
general. Its not. That 128 level 2 cache has its limits, and the
principle of locality can only be stretched so far when you're
multitasking Word, IE 5.0, a mail program, and some other background
threads.
Mike
------------------------------
From: [EMAIL PROTECTED] (Paul Colquhoun)
Crossposted-To: comp.databases.informix,comp.os.linux.setup,comp.periphs.scsi
Subject: Re: To RAID or not to RAID? -that is the question...
Reply-To: <[EMAIL PROTECTED]>
Date: Wed, 07 Jul 1999 02:44:38 GMT
On 7 Jul 1999 00:44:21 GMT, Salem Lee Ganzhorn <[EMAIL PROTECTED]> wrote:
|[EMAIL PROTECTED] wrote:
|: Thanks y'all for your comments/suggestions. I guess
|: I must be missing something cause I did not get what
|: a lot of people talked about using RAID5. RAID5 will
|: no doubt be faster but then there is no redundancy.
|: Maybe two RAID5 channels and then database mirroring
|: across the two controllers?
|:
|: Will someone please comment on this: Suppose I have
|: a choice of having one single RAID1/0 (or even just
|: RAID1) controller OR having two Ultra2 SCSI channels
|: and then use Informix mirroring across the two cont-
|: rollers. Which one makes more sense?
|
|Raid 5 is mirrored. There is enough data mirrored on the disks so if any
|one drive goes down you can rebuild the drive completely from the remaining
|drives. You get the speed of striping plus the fault tolerance of mirroring.
|
|Of course you only get 1/2 of the space the drives have to offer.
|
|The only downside is you have to have at least 3 drives.
The only RAID varieties in common use are:
0 Plain striping, no redundancy.
+ Faster reading and writing ( spread over all drives )
- Loose 1 drive = loose all the data.
1 Mirroring. Most redundant, but needs 2x the drive space.
+ Faster reads ( can read from either drive )
+ Can loose a drive and still keep going.
- Writes are slightly slower, data has to be written twice.
5 Needs 3+ drives. Data is striped acros the drives, but 1 drive
worth of data is made up of checksums of the other data. This
is spread out over all the drives.
+ Faster reads, spread out over all drives,
+ Can loose a drive and keep going ( with reduced speed )
- Slower writes. Need to recompute and wite the checksum as well
A combination of Striping and Mirroring is sometimes called
RAID 1+0, RAID 0+1, or RAID 10
--
Reverend Paul Colquhoun, [EMAIL PROTECTED]
Universal Life Church http://andor.dropbear.id.au/~paulcol
-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-=*=-
xenaphobia: The fear of being beaten to a pulp by
a leather-clad, New Zealand woman.
------------------------------
From: [EMAIL PROTECTED] (Chris Robato Yao)
Crossposted-To: comp.sys.ibm.pc.hardware.chips,comp.sys.intel
Subject: Re: Celeron, what's the catch?
Date: 7 Jul 1999 03:02:38 GMT
Reply-To: [EMAIL PROTECTED] (Chris Robato Yao)
In <[EMAIL PROTECTED]>, [EMAIL PROTECTED] (kls) writes:
>In article <7ls4np$pp3$[EMAIL PROTECTED]>, [EMAIL PROTECTED]
>says...
>>
>>In <[EMAIL PROTECTED]>, [EMAIL PROTECTED] (kls) writes:
>>>In article <7lrrtr$hgc$[EMAIL PROTECTED]>, [EMAIL PROTECTED]
>says...
>>>>
>>>>In <[EMAIL PROTECTED]>, [EMAIL PROTECTED] (kls)
>writes:
>>>>>& you have a ver poor ability of looking up benchmark results.
>>>>
>>>>
>>>>Where are your benchmark results by the way?
>>>
>>>not MY benchmark results. Where are YOURS? Couldn't find any eh?
>That
>>>poor look up ability of yours I suppose...
>>
>>You are the one claiming a Dual Celeron setup (using 333s) can be
>faster
>>than a K6-III 450 on LInux. The Burden of Proof is on you.
>>
>>But you have not shown anything to back your claims.
>>
>>Have you seen http://www.cpureview.com?
>>
>>Article shows a K6-III 400 is faster than a Cel @ 450 in compiling.
>
>
>From http://www.cpureview.com/art_kernel_discuss_c.html,
>k63-400 vs single c400(4x66) k63 23.5% faster(though
>he gives different results in seperate reviews?).
>
>Remember 23.5%,
Which is basically good enough for two speed grades (a speed grade is
generally about 12%).
>
>http://perso.easynet.fr/~hugues.michel/gcc_time.html(use babefish)
>
>The times are not comparable(different build,..) but the % speed
>difference between single & smp is viable.
Yes, 34%. How does that make a couple of Cel 333s faster than a single
K6-3 450 (which is about the equivalent of a between Cel 500 and 550 in
application performance.) Unless you overclock them, but then of
course, you have a situation when Celerons are used out of their basic
specification (overclocked and SMP). To work these things out, you
need considerable time, trial, experience and preparation, not something
to recommend unless the person knows exactly what he is doing and is
committed to it.
Why should you recommend a *not supported* configuration to a buyer?
>
>Here's freebsd(pretty graphs even:),
>http://www.freebsd.org/~fsmp/SMP/akgraph-a/graph1.html
>
>
>>>>You have not yet answered my other question on your ratios of K6 vs.
>PII
>>>>FPUs.
>>>
>>>benchmarks, hardware & programming sites. That & I own a dog of a
>cpu(k6-2
>>
>>I have not see any benchmark or website that suggests a .55 ratio for
>>the P2 FPU vs. K6. You made this up. Back it up.
>>
>>On the contrary, there is plenty that suggests otherwise. Read the
>>K6-III review of http://www.combatsim.com. According to them, a
>K6-III
>>450 is barely equivalent to a PII 400 on non 3DNow games.
>>
>>
>>
>>>300. old core: .55 p2 fpu). On a personal note, if the performance
>were so
>>>ubar compared to p2's, as you keep on insisting, my frame rate
>wouldn't drop
>>>to 16-9fps in warbirds when things get even slightly heavy. I must
>be
>>>dreaming eh?
>>
>>I can get an Celeron overclocked to 500 to drop down to 16-9 FPS on
>>Jane's WWII Fighters when things get heavy. Does not mean anything.
>
>It does when a p2-300 in place of a k6-2 300 in the same situation is
>29-16fps instead.
>
>
Is this where you derive your .55?
That does not hold true on the average. If you want to compile as many
benchmarks you can find, the average is higher.
When I had my K6-2 300 and PII-300 before, when I benched with a Matrox
G200 last year, I got 3.25 on the K6-2 and 3.67 on the PII-300 on Final
Reality. Far from a .55 difference. A PII-450 on the old non 3DMark99
with a Riva TNT gets about 2800s, but a K6-2 450 with the same card gets
about 2400s without Detonator 3DNow drivers and about even with them.
2400 vs. 2800 isn't .55 as well. In both cases you got about 15 to 20%
differences.
Rgds,
Chris
(And the NUMBER ONE top oxy-MORON
1. Microsoft Works
---From the Top 50 Oxymorons (thanks to Richard Kennedy)
------------------------------
From: wizard <[EMAIL PROTECTED]>
Subject: Re: *** Need modem recommendation for LINUX
Date: Wed, 07 Jul 1999 00:09:09 -0400
Reply-To: [EMAIL PROTECTED]
__Fred Simmler wrote:
> I am running Redhat 6.0 on a PC and need to buy a 56K Flex/V90 modem. I am
> trying to get away from ISA slot modems because the newer motherboards
> are almost eliminating them. I would like to stay with an internal modem
> but I am not sure if there are any PCI slot modems that are not
> "software modems". Are there any PCI modems that are compatable with
> Linux? Am I locked into going to an external modem? What is a good
> value?
If you have a PC with an ISA slot go with a new ISA slot modem. The
technology of internet connection is changing at a rapid pace, a year from now
you may not even be using a modem.
You are correct in stating that some manufactures are leaving the ISA bus, the
problem is serial ports will be left behind about the same time. In the
long run you may be supportting a USB or FireWire ported modem. It will be
a couple of years before everyone switches over to the new technology.
I prefer the internal devices my self and would suggest you go that route.
You end up with a cleaner installation and fewer powersupplies to manage.
I would suspect that there wouold be a few PCI modems that would work but I
have not used any myself.
Dave
------------------------------
From: [EMAIL PROTECTED] (Anthony Hill)
Crossposted-To: comp.sys.ibm.pc.hardware.chips,comp.sys.intel
Subject: Re: Celeron, what's the catch?
Date: Wed, 07 Jul 1999 03:03:34 GMT
On 5 Jul 1999 03:19:33 GMT, [EMAIL PROTECTED] (Chris Robato
Yao) wrote:
>>Wrong again. ABit BP6 supports dual Socket 370 Celerons and UDMA/66
>>support using the Intel BX chipset.
>
>How? The BX does not officially support UDMA 66 unless you do tweaks
>that is not supported by Intel. I really doubt if it's true UDMA 66 at
>all since according to Intel, BX has no UDMA 66 capability. The only
>way I can see that is possible is by matching the i810's southbridge
>with the BX northbridge, and that is not sanctioned or allowed by Intel.
The new Abit boards with UDMA/66 apperently have a Promise
UDMA/66 IDE controller on-board. You are correct in that the PIIX4E
Southbridge used by all of Intel's current chipsets except the i810
(which doesn't use a north/south bridge design at all) does not
support UDMA/66.
Anthony Hill
[EMAIL PROTECTED]
------------------------------
From: [EMAIL PROTECTED] (Anthony Hill)
Crossposted-To: comp.sys.ibm.pc.hardware.chips,comp.sys.intel
Subject: Re: Celeron, what's the catch?
Date: Wed, 07 Jul 1999 03:03:36 GMT
On Sat, 3 Jul 1999 04:16:38 -0400, "FM" <[EMAIL PROTECTED]> wrote:
>I'm buying a PC soon for college (from college) and the standard package
>offered (http://www.dartmouth.edu/comp/newstudentinfo/buying/hardware.html)
>is Celeron 433, 6GB, 64MB, 15" (quite the worst point) etc. While I have not
>seen the specs of higher-end systems or the price, but so far it seems that
>this one will fit my budget best (well their systems seemed a bit overpriced
>despite alleged academic discounts;
>http://www.dartmouth.edu/~store/pricelist.html). But before making the
It's entirely possible that their prices aren't all that
great. I know that the computer store at my school has some pretty
poor prices on most things that they sell. I only shop there when I
need really small things (eg printer cartridges) where it's not worth
my while to go elsewhere.
>decision, I'd like some information on Celeron systems since I have some
>reservation about the chip (I guess due to some early criticism I didn't
>really heed but was exposed to nevertheless). First, does it perform as well
>as the benchmarks suggest?
For the most part, yes.
> I've seen some FPU benchmarks indicating that
>Celeron outperforms similarly clocked PII and some Integer benchmarks where
>it still holds fairly well. But do these benchmarks reflect the overall
>system performance considering the slower bus speed (66mhz) and other
>compromises? Second, are there any particular application areas where
>Celeron fares poorly? For example my hunch is that its design (smaller but
>faster L2 cache) wouldn't favor applications that require intensive but
>repetitive memory/disk access (server? compilers?) but is the difference
>worth noting?
The difference is only worth noting where the data set being
worked on tends to fit entirely in the L2 cache of the PII, but not
that of the Celeron. This sort of situation does occur from time to
time, but it's really quite rare. Genreally speaking the PII is
slightly faster, but usually you can figure on getting at least
equivilant performance from a Celeron 433 as from a PII 400 (assuming
all else is equal). If most of the calculations fit into the 128K
cache of the Celeron, it will be faster (hence the reason why
Celeron's sometimes win some FPU benchmarks, since these often use
smaller sets of data). When you're working with a LOT of memory
access that is really bandwidth intensive (this is mainly just heavy
graphics or sound processing), the PII will be faster due to the
100MHz memory bus vs. the 66MHz memory bus of the Celeron, however
again it's rare that the difference will be sufficient to make a PII
400 faster then a Celeron 433. It should also be noted that a lot of
"memory intensive" applications are really more dependant on latency
rather then the bandwidth of memory, and the faster memory bus of the
PII doesn't do much for improving latency.
>As for my use, it will be primarily used as a desktop Linux machine (Redhat
>6.0 with GNOME or KDE) with some casual server daemons like http, ftp,
>telnet, etc. Other tasks would include (ordered by approximate
>frequency/priority) wordprocessing, internet client apps,
>programming/compilation, image-manipulation (small-scale for my
>yet-to-be-purchased digital camera and yet-to-be published webpage), MP3
>(incase I can't afford a stereo), fractals, and maybe some chess programs
>(hmm maybe this is where Celeron might show its weaknesses?). I'm not
I can't see any application there were you're likely to find
Celeron 433 to be slower then a PII 400, and often I think you'll find
it to be faster. Considering that the Celeron 433 is cheaper then the
PII 400 (actually I think that even the Celeron 466 is cheaper then a
PII 400), I'd definately go for the Celeron out of those two.
Another chip you might want to consider would be the AMD
K6-III. This is an excellent chip for nearly all the applications you
listed above, and in most cases would probably be quite a bit faster
then either the Celeron or the PII at the same clock speed. Most of
the things you mentioned are going to be doing almost exclusively just
integer opperations, where the K6-III is very fast. The image
manipulation would be doing floating point work, but unless you're
doing a lot of fairly heavy duty graphics manipulation I don't think
that you'll find this to be a problem at all (when the difference is 5
seconds on a K6-III vs. 4 seconds on a PII, it's not nearly as bad as
10 minutes vs. 8 minutes). The only weakness I could see with the
K6-III would be for fractals, which I'm not altogether familiar with.
Anyway, just a thought.
>planning on playing games much and when I do, I'm unusually tolerant of
>low-res/low-framerate, not to mention that I'm not into modern
>graphic-intensive 3D shootemups (yeah I'm the guy who used to play starcraft
>on P75 overclocked to P100 and didn't find the setup disturbing at all). I
Hmm, I think I was the guy who got annoyed at your slow
computer for lagging up any Starcraft games I might have played
against you over Battle.net! :>
Anthony Hill
[EMAIL PROTECTED]
------------------------------
Date: Wed, 07 Jul 1999 00:23:24 -0400
From: Kenneth Been <[EMAIL PROTECTED]>
Subject: Re: modem mystery
Well, I got rid of the IRQ conflict, but the problem didn't go away.
First, I changed the jumpers on the old modem, so it is on COM2 and IRQ
3 (the new one is on COM3 and IRQ 4). (I also had to disable COM2 in
setup to get rid of an address conflict when booting.)
When this did not solve the problem, I tried removing the old modem
altogether. Still no luck.
Any more ideas?
Here is the output when I do
setserial -g /dev/ttyS?
/dev/ttyS0, UART: unknown, Port: 0x03f8, IRQ: 4
/dev/ttyS1, UART: 16550A, Port: 0x02f8, IRQ: 3
/dev/ttyS2, UART: 16550A, Port: 0x03e8, IRQ: 4
/dev/ttyS3, UART: unknown, Port: 0x02e8, IRQ: 3
Thanks.
Ken
M. Buchenrieder wrote:
>
> Kenneth Been <[EMAIL PROTECTED]> writes:
>
> >I just bought a Viking 56K external modem. I tried it with a friend's
> >computer, with win95, and it worked ok. On my Linux box, it is very
> >slow to even respond to commands, like the connect command or various AT
> >commands that I try sending it from kermit.
>
> >For example, I try this command (from the Serial-HOWTO):
>
> >ATE1Q0V1
>
> >and it takes probably 15 seconds or more just to echo the command back
> >to the terminal.
>
> Classical IRQ conflict. You can not have two ISA devices (or an ISA and
> a PCI device) sharing one IRQ on est. 95% of today's PC mainboards.
> You'll have to assign a separate IRQ to all serial lines you want to
> use.
>
> See the Serial-HOWTO for details.
------------------------------
** 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
******************************
