Why Pentium?

  • Thread starter Thread starter Talal Itani
  • Start date Start date
kony pravi:

Seriously, do you think the Pentium optimizations are better than the
AMD ones?

It doesn't matter, rather that the applications being tested
must have codepaths optimized for the generation of
processor tested. In many cases, they don't have the AMD
optimizations which is a reason for a workstation type use
of these apps to benefit from use of a P4, but
generalization about performance levels must factor for this
as well, that the buyer MUST also buy the software.

Different software will need assessed on it's own, not
vaguely assumed to have similar performance per CPU
architecture.


AMD's 64bit optimizations aren't used very well yet either. But all it
takes is to download properly compiled software (Linux) and you do.

The operating system itself is not at issue, either of these
alternatives are plenty fast enough for the OS. Rather the
applications are the key and no OS optimization is enough
without the app optimization.

And AMDs are practically faster than Intels. I know the numbers look
neater on Intel processors, but AMDs very simply make a computer more
responsive. I'm guessing it's something to do with overall design.

There are several factors, and it's not that Intel's
offerings aren't a good alternative, only that the specific
use must be weighed against the performance/_total_price of
each.
 
kony said:
Different software will need assessed on it's own, not
vaguely assumed to have similar performance per CPU
architecture.


Criminy, Korny, this is an English NG. What the hell
language was that? If you'd slow down and stop spewing,
people might know what the hell you're trying to say.

*TimDaniels*
 
Criminy, Korny, this is an English NG. What the hell
language was that? If you'd slow down and stop spewing,
people might know what the hell you're trying to say.

*TimDaniels*

The subject is not that big of a deal, either. The original poster
could have avoided cross posting.
 
John Doe pravi:
So do fluid dynamic bearing fans, but longer.

Are those those that don't really have a bearing but only a pocket of
fluid taped in?

They last about a month as far as my experience goes. The upper side of
the bearing is unprotected, which means that surrounding dust (attracted
by the electric charge of the fan engine) can have direct contact with
the fluid. Once there, the dust turns the fluid glue-solid and the fan
comes to a sticky stop, until you remove the fluid with some kind of
ethanol-based solution. Then it spins again, but with no real bearing.

--
Primary function: Coprocessor
Secondary function: Cluster commander

http://www.thought-beacon.net

Pay once per lifetime webhosting:
http://farcomm-it.com/?ref=jsah

We are the paragon of humanity. You may worship us. From afar.

01010010 01100101 01110011 01101001 01100100 01100101 01101110 01110100
01000010 01000001 01010011 01001001 01000011
 
kony pravi:
It doesn't matter, rather that the applications being tested
must have codepaths optimized for the generation of
processor tested. In many cases, they don't have the AMD
optimizations which is a reason for a workstation type use
of these apps to benefit from use of a P4, but
generalization about performance levels must factor for this
as well, that the buyer MUST also buy the software.

The vast majority of software developers does not use a language that
would allow for these specific optimizations. The use of the
optimizations arises from the way their compiler works.

For example the GNU C Compiler can compile the software for any
enstablished CPU of choice.. software is generally compiled for the
Intel 386 processor, with added optional optimizations for MMX and the
various SSEs. Similar applies for how Microsoft compiles it's software
(okay, short of the DirectX drivers). Yet all this software appears to
perform better on AMD processors.
Different software will need assessed on it's own, not
vaguely assumed to have similar performance per CPU
architecture.

That only really applies for benchmarking software and certain
performance-critical code here and there. Nothing to do with general
performance.

Beside CPUs are primarily built to perform well with software compiled
with a specific compiler. Compilers are only adapted to utilize special
features of various CPUs.
The operating system itself is not at issue, either of these
alternatives are plenty fast enough for the OS. Rather the
applications are the key and no OS optimization is enough
without the app optimization.

And your knowledge of software construction is nearing -20% of what it
should be to make that assessment.

Programs depend heavily on the code present in the OS, thus it is
important that the OS is built for speed and compiled to the proper
architecture.

Also, you would probably like to know that in Linux, once the OS of
choice compiled to the proper architecture is installed, practically all
consequent software installed is also of the exact same architecture
(either selected from available sources or automatically compiled for
the exact hardware the Linux system is running on).
There are several factors, and it's not that Intel's
offerings aren't a good alternative, only that the specific
use must be weighed against the performance/_total_price of
each.

The AMD processors I was looking at were cheaper than the Intel ones
I've compared the performance to.


--
Primary function: Coprocessor
Secondary function: Cluster commander

http://www.thought-beacon.net

Pay once per lifetime webhosting:
http://farcomm-it.com/?ref=jsah

We are the paragon of humanity. You may worship us. From afar.

01010010 01100101 01110011 01101001 01100100 01100101 01101110 01110100
01000010 01000001 01010011 01001001 01000011
 
kony pravi:
Sure, but the apps themselves can.
Actually it can be a benefit, if for example you wanted to
apply some filters and capture to MPEG4 while using a
system, it'll be good to give that enough priority.

Nearly all modern OSs are rather unsuitable for signal processing. Their
internal system for sharing CPU time amongst programs is inefficient
when you run processes that need a lot of CPU in a fluid manner
alongside those who need little. That's why today you have buffers
everywhere you look.

Your 5 settings of Windows priority will simply not let you OR your
system provide all programs with the optimal share of CPU power. You'd
need the OS doing that for you automatically if you wanted it to work.

--
Primary function: Coprocessor
Secondary function: Cluster commander

http://www.thought-beacon.net

Pay once per lifetime webhosting:
http://farcomm-it.com/?ref=jsah

We are the paragon of humanity. You may worship us. From afar.

01010010 01100101 01110011 01101001 01100100 01100101 01101110 01110100
01000010 01000001 01010011 01001001 01000011
 
Jure said:
John Doe pravi:

Are those those that don't really have a bearing but only a pocket
of fluid taped in?

Sounds like a troll, considering you pretend to have experience with
them. They are called "fluid bearings" or "fluid dynamic bearings".

http://en.wikipedia.org/wiki/Fluid_bearing
They last about a month as far as my experience goes. The upper
side of the bearing is unprotected, which means that surrounding
dust (attracted by the electric charge of the fan engine) can have
direct contact with the fluid. Once there, the dust turns the
fluid glue-solid and the fan comes to a sticky stop, until you
remove the fluid with some kind of ethanol-based solution. Then it
spins again, but with no real bearing.

My guess is your "experience" is fantasy, or you bought a cheap fan,
or maybe you tried to lubricate it before your problems started, or
maybe you just couldn't find one with a blue LED.

Whatever.
 
John Doe pravi:
Sounds like a troll,

Tell me, is everything that you disagree with a troll to you?
considering you pretend to have experience with
them. They are called "fluid bearings" or "fluid dynamic bearings".

http://en.wikipedia.org/wiki/Fluid_bearing

Sounds like quite what I had in mind.
My guess is your "experience" is fantasy, or you bought a cheap fan,
or maybe you tried to lubricate it before your problems started, or
maybe you just couldn't find one with a blue LED.

Or maybe you're looking for valid excuses to proove yourself right. All
I'm saying is that the fluid bearing is unsuitable for a microchip
cooling fan and thus do not last long. I have both the experience and
the theory to proove it.

I have seen these fluid bearings in many CPU fans, graphic card fans and
some slot fans. The general characteristics are always the same: The
tiny circular circuit board that manages the magnets that keep the fan
spinning is set directly atop the bearing, the axis is led trough a hole
in the circuit board, the fluid bearing being located inside the plastic
casing supporting the tiny circuit board. The underside of this plastic
casing is taped over with a factory sticker to seal the fluid inside the
bearing. The axis is terminated on this side with a small metal ring
embedded in a shallow gap near the end of the axis, which is floating in
the fluid and is there to prevent the axis from being pulled out of the
bearing vertically.

The upper side of the bearing is thus not sealed shut, as that would be
impossible given the point that the axis needs to be able to rotate
freely. Air travels along the fan and at any point along the fan's
construction where the airflow is slowed down, dust accumulates. The
tiny circuit board's electrical contacts are bare, exposed and charged
with a ~12V electric charge, which is powering the fan. Dust is
attracted by the ionization and tiny dust particles find their way right
to the axis. Very slowly, over time, enough dust accumulates there and
along with the effects of the heat, the fluid turns into a glue-like
substance, preventing the axis from moving. At this point it is very
hard to rotate the fan even with your fingers and sometimes when it
cools off it becomes stuck solid.

Sometimes, you don't even need that as the glue on the back sticker no
longer seals it completely shut and that combined with the heat of the
cooling apparatus causes the fluid to leave the bearing.



--
Primary function: Coprocessor
Secondary function: Cluster commander

http://www.thought-beacon.net

Pay once per lifetime webhosting:
http://farcomm-it.com/?ref=jsah

We are the paragon of humanity. You may worship us. From afar.

01010010 01100101 01110011 01101001 01100100 01100101 01101110 01110100
01000010 01000001 01010011 01001001 01000011
 
John Doe pravi:
In Windows XP, it's 6 settings.

In any Linux or UNIX it is 20 settings (it's called "Nice", not
"Priority" tho). And in any real-time OS there are no settings as the OS
fine-tunes the priorities automatically at the precision of the CPU clock.

--
Primary function: Coprocessor
Secondary function: Cluster commander

http://www.thought-beacon.net

Pay once per lifetime webhosting:
http://farcomm-it.com/?ref=jsah

We are the paragon of humanity. You may worship us. From afar.

01010010 01100101 01110011 01101001 01100100 01100101 01101110 01110100
01000010 01000001 01010011 01001001 01000011
 
So do fluid dynamic bearing fans, but longer.

"Fluid dynamic bearing" is in itself just a nonsense
marketing term. They all fall into one of two categories,
ball bearing or sleeve. We can further divide into dual
ball, ball + sleeve, and certain types of sleeve designs but
overall the type of fluid bearings used in PC fans is low
quality rather than the high quality sealed bearings used in
hard drives.
 
My guess is your "experience" is fantasy, or you bought a cheap fan,
or maybe you tried to lubricate it before your problems started, or
maybe you just couldn't find one with a blue LED.

Whatever.


Fluid bearing fans are usually very low quality junk. A
very few manufacturers make decent ones, like Panasonic,
Papst, Comair, a few Sunons and Deltas are passible.

Any serious design needing longest fan life uses dual ball
bearing fans if at all possible. Intel's fans are this, but
intel spec'd for a particularly small center hub and to
achieve that the bearings themselves were downsized from
what the manufacturer normally uses (for better results).

The manufacturer didn't randomly choose larger bearings on a
whim, they knew more about fans than Intel.
 
kony pravi:

Nearly all modern OSs are rather unsuitable for signal processing.

Nonsense.

They may be sub-optimal, but we're not doing one hardcoded
task on a PC, the versatility is important too. Unsuitable
in this context would not apply as they certainly can and do
get the job done.
Their
internal system for sharing CPU time amongst programs is inefficient
when you run processes that need a lot of CPU in a fluid manner
alongside those who need little. That's why today you have buffers
everywhere you look.

Yes they're not "as" efficient as they could be, but that is
the nature of a multipurpose system. It couldn't work any
other way and be versatile enough, and inexpensive enough,
to do so many jobs.


Your 5 settings of Windows priority will simply not let you OR your
system provide all programs with the optimal share of CPU power. You'd
need the OS doing that for you automatically if you wanted it to work.

Sure it will, most programs don't need more than a *normal*
priority level, they can sit in the background. If the
workload is so high that multiple realtime activities can't
be done, this is a case that would be a problem for any type
of hardware design, not just a PC.
 
kony said:
Fluid bearing fans are usually very low quality junk.

Says who?
A very few manufacturers make decent ones, like Panasonic, Papst,
Comair, a few Sunons and Deltas are passible.

Any serious design needing longest fan life uses dual ball bearing
fans if at all possible.

That's backwards. Fluid dynamic bearings last many times longer than
ball bearings.
 
A troll who needs to provide citations instead of babbling.


Path: newssvr27.news.prodigy.net!newsdbm04.news.prodigy.com!newsdst01.news.prodigy.net!prodigy.com!newscon06.news.prodigy.com!prodigy.net!newsfeed.cw.net!cw.net!news-FFM2.ecrc.de!news.germany.com!newsfeed.utanet.at!news-hub.siol.net!news.siol.net!not-for-mail
From: Jure Sah <admin thought-beacon.net>
Organization: MesonAI
User-Agent: Thunderbird 1.5.0.4 (Windows/20060516)
MIME-Version: 1.0
Newsgroups: alt.comp.hardware,alt.comp.hardware.homebuilt,alt.comp.hardware.pc-homebuilt,alt.computer,comp.sys.ibm.pc.hardware.chips
Subject: Re: Why Pentium?
References: <4gusksF1p657fU1 individual.net> <pbhsa21cjiqivcam6mpqnsqvhph40rnkb9 4ax.com> <6mjsa2h1li63b92ae46qtpppeh4135kcvf 4ax.com> <74osa2tdjivbnlfecp2d3vnce99065folt 4ax.com> <qvnta2prmvkn5s62kq0i7vgo56uhvcgr3n 4ax.com> <4h862mF1qdbajU1 individual.net> <432ua2pjl8hi3avm66u8joi9lkdeu8g1ri 4ax.com> <4h8nviF1or785U1 individual.net> <f2m0b2ljvsk60a26gtvfk72phlc9m7d0fv 4ax.com> <4hb9hrF1qpn08U1 individual.net> <a361b21d9h02olr8bbispuosrh9jodrfim 4ax.com> <4hbl1aF1qlql5U1 individual.net> <icq9b2ldci81q6es9jupvemokrrhprshee 4ax.com> <4hkr8hF2bseU1 individual.net> <kHNvg.4362$oj5.1511399 news.siol.net> <Xns9806854A4F3580123456789 207.115.17.102> <2jSvg.4371$oj5.1512740 news.siol.net> <Xns9806ABC3D3F3B0123456789 207.115.17.102>
In-Reply-To: <Xns9806ABC3D3F3B0123456789 207.115.17.102>
Content-Type: text/plain; charset=ISO-8859-2; format=flowed
Content-Transfer-Encoding: 7bit
Lines: 82
Message-ID: <UkTvg.4375$oj5.1513262 news.siol.net>
Date: Fri, 21 Jul 2006 00:20:35 +0200
NNTP-Posting-Host: 193.77.182.145
X-Complaints-To: abuse siol.net
X-Trace: news.siol.net 1153434036 193.77.182.145 (Fri, 21 Jul 2006 00:20:36 MET DST)
NNTP-Posting-Date: Fri, 21 Jul 2006 00:20:36 MET DST
Xref: prodigy.net alt.comp.hardware:314510 alt.comp.hardware.homebuilt:223679 alt.comp.hardware.pc-homebuilt:471501 alt.computer:270315 comp.sys.ibm.pc.hardware.chips:460716

John Doe pravi:
Sounds like a troll,

Tell me, is everything that you disagree with a troll to you?
considering you pretend to have experience with
them. They are called "fluid bearings" or "fluid dynamic bearings".

http://en.wikipedia.org/wiki/Fluid_bearing

Sounds like quite what I had in mind.
My guess is your "experience" is fantasy, or you bought a cheap fan,
or maybe you tried to lubricate it before your problems started, or
maybe you just couldn't find one with a blue LED.

Or maybe you're looking for valid excuses to proove yourself right. All
I'm saying is that the fluid bearing is unsuitable for a microchip
cooling fan and thus do not last long. I have both the experience and
the theory to proove it.

I have seen these fluid bearings in many CPU fans, graphic card fans and
some slot fans. The general characteristics are always the same: The
tiny circular circuit board that manages the magnets that keep the fan
spinning is set directly atop the bearing, the axis is led trough a hole
in the circuit board, the fluid bearing being located inside the plastic
casing supporting the tiny circuit board. The underside of this plastic
casing is taped over with a factory sticker to seal the fluid inside the
bearing. The axis is terminated on this side with a small metal ring
embedded in a shallow gap near the end of the axis, which is floating in
the fluid and is there to prevent the axis from being pulled out of the
bearing vertically.

The upper side of the bearing is thus not sealed shut, as that would be
impossible given the point that the axis needs to be able to rotate
freely. Air travels along the fan and at any point along the fan's
construction where the airflow is slowed down, dust accumulates. The
tiny circuit board's electrical contacts are bare, exposed and charged
with a ~12V electric charge, which is powering the fan. Dust is
attracted by the ionization and tiny dust particles find their way right
to the axis. Very slowly, over time, enough dust accumulates there and
along with the effects of the heat, the fluid turns into a glue-like
substance, preventing the axis from moving. At this point it is very
hard to rotate the fan even with your fingers and sometimes when it
cools off it becomes stuck solid.

Sometimes, you don't even need that as the glue on the back sticker no
longer seals it completely shut and that combined with the heat of the
cooling apparatus causes the fluid to leave the bearing.



--
Primary function: Coprocessor
Secondary function: Cluster commander

http://www.thought-beacon.net

Pay once per lifetime webhosting:
http://farcomm-it.com/?ref=jsah

We are the paragon of humanity. You may worship us. From afar.

01010010 01100101 01110011 01101001 01100100 01100101 01101110 01110100
01000010 01000001 01010011 01001001 01000011
 
A troll who doesn't know what a fluid dynamic bearing is.


Path: newssvr27.news.prodigy.net!newsdbm04.news.prodigy.com!newsdst01.news.prodigy.net!prodigy.com!newscon06.news.prodigy.com!prodigy.net!border1.nntp.dca.giganews.com!nntp.giganews.com!local02.nntp.dca.giganews.com!nntp.insightbb.com!news.insightbb.com.POSTED!not-for-mail
NNTP-Posting-Date: Thu, 20 Jul 2006 21:06:50 -0500
From: kony <spam spam.com>
Newsgroups: alt.comp.hardware,alt.comp.hardware.homebuilt,alt.comp.hardware.pc-homebuilt,alt.computer,comp.sys.ibm.pc.hardware.chips
Subject: Re: Why Pentium?
Date: Thu, 20 Jul 2006 22:07:53 -0400
Reply-To: spam spam.com
Message-ID: <6id0c25a3gbdc36fc2m5s349f1sg78kp2d 4ax.com>
References: <qvnta2prmvkn5s62kq0i7vgo56uhvcgr3n 4ax.com> <4h862mF1qdbajU1 individual.net> <432ua2pjl8hi3avm66u8joi9lkdeu8g1ri 4ax.com> <4h8nviF1or785U1 individual.net> <f2m0b2ljvsk60a26gtvfk72phlc9m7d0fv 4ax.com> <4hb9hrF1qpn08U1 individual.net> <a361b21d9h02olr8bbispuosrh9jodrfim 4ax.com> <4hbl1aF1qlql5U1 individual.net> <icq9b2ldci81q6es9jupvemokrrhprshee 4ax.com> <4hkr8hF2bseU1 individual.net> <kHNvg.4362$oj5.1511399 news.siol.net> <Xns9806854A4F3580123456789 207.115.17.102>
X-Newsreader: Forte Agent 1.93/32.576 English (American)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Lines: 23
NNTP-Posting-Host: 74.129.166.247
X-Trace: sv3-jNSaqENM9z60jKNi7G8UjAtTJdw49JkiWFPeF2HXCwvIhaKgFCTGtZBTA6mD8fcU7cnjM/6/auZjYaX!heqAPkJ66Pqkz611rFlMhciYt/+oqbVSu7W0Tb3u5y8sMqeW/nxpvH8Yq5/APtxxbUCJreXDRojF!6zVs+L0LIeEOgQ==
X-Complaints-To: abuse insightbb.com
X-DMCA-Complaints-To: abuse insightbb.com
X-Abuse-and-DMCA-Info: Please be sure to forward a copy of ALL headers
X-Abuse-and-DMCA-Info: Otherwise we will be unable to process your complaint properly
X-Postfilter: 1.3.32
Xref: prodigy.net alt.comp.hardware:314541 alt.comp.hardware.homebuilt:223681 alt.comp.hardware.pc-homebuilt:471523 alt.computer:270332 comp.sys.ibm.pc.hardware.chips:460726

So do fluid dynamic bearing fans, but longer.

"Fluid dynamic bearing" is in itself just a nonsense
marketing term. They all fall into one of two categories,
ball bearing or sleeve. We can further divide into dual
ball, ball + sleeve, and certain types of sleeve designs but
overall the type of fluid bearings used in PC fans is low
quality rather than the high quality sealed bearings used in
hard drives.
 
I knew this troll was going to have a tough time with my simple
correction.


Path: newssvr27.news.prodigy.net!newsdbm04.news.prodigy.com!newsdst01.news.prodigy.net!prodigy.com!newscon06.news.prodigy.com!prodigy.net!newsfeed-00.mathworks.com!kanaga.switch.ch!switch.ch!newsfeed00.sul.t-online.de!t-online.de!newsfeed.t-com.hr!news-hub.siol.net!news.siol.net!not-for-mail
From: Jure Sah <admin thought-beacon.net>
Organization: MesonAI
User-Agent: Thunderbird 1.5.0.4 (Windows/20060516)
MIME-Version: 1.0
Newsgroups: alt.comp.hardware,alt.comp.hardware.homebuilt,alt.comp.hardware.pc-homebuilt,alt.computer,comp.sys.ibm.pc.hardware.chips
Subject: Re: Why Pentium?
References: <icola2h0lp4fgk07iesgssutsje3bgp1ot 4ax.com> <4h0alnF1ocnf4U1 individual.net> <qo2ma21va6e60oj3rq3ptaec4v9i6gomks 4ax.com> <3cisa29nrsvn4tfa0095u50g50hqcsqr1k 4ax.com> <1rata2h3jltsj1pr0n9fc85r0r4cdhj33a 4ax.com> <1olta29uos0r2d078pfhag46umbu0naep1 4ax.com> <5dota2t45mh3pbjm3afa9428as13sm8vtu 4ax.com> <ojoua2559khkqvetbhcas1b3lq4ub7vi0o 4ax.com> <65n0b2l57goe2jmcc0ptfmrnlvunmcssr9 4ax.com> <3hk1b25be3b159d7vjfk65csdk38nufh25 4ax.com> <rcr9b2h54lfp3ukl4nhq0vn7dgnov85ken 4ax.com> <KMSvg.4373$oj5.1513161 news.siol.net> <Xns9806AE68B15C0123456789 207.115.17.102>
In-Reply-To: <Xns9806AE68B15C0123456789 207.115.17.102>
Content-Type: text/plain; charset=ISO-8859-2; format=flowed
Content-Transfer-Encoding: 7bit
Lines: 37
Message-ID: <bzTvg.4376$oj5.1513513 news.siol.net>
Date: Fri, 21 Jul 2006 00:35:52 +0200
NNTP-Posting-Host: 193.77.182.145
X-Complaints-To: abuse siol.net
X-Trace: news.siol.net 1153434951 193.77.182.145 (Fri, 21 Jul 2006 00:35:51 MET DST)
NNTP-Posting-Date: Fri, 21 Jul 2006 00:35:51 MET DST
Xref: prodigy.net alt.comp.hardware:314511 alt.comp.hardware.homebuilt:223680 alt.comp.hardware.pc-homebuilt:471502 alt.computer:270317 comp.sys.ibm.pc.hardware.chips:460718

John Doe pravi:
In Windows XP, it's 6 settings.

In any Linux or UNIX it is 20 settings (it's called "Nice", not
"Priority" tho). And in any real-time OS there are no settings as the OS
fine-tunes the priorities automatically at the precision of the CPU clock.

--
Primary function: Coprocessor
Secondary function: Cluster commander

http://www.thought-beacon.net

Pay once per lifetime webhosting:
http://farcomm-it.com/?ref=jsah

We are the paragon of humanity. You may worship us. From afar.

01010010 01100101 01110011 01101001 01100100 01100101 01101110 01110100
01000010 01000001 01010011 01001001 01000011
 
Says who?


Anyone that's had a few years experience dealing with failed
fans.

That's backwards. Fluid dynamic bearings last many times longer than
ball bearings.

There is no reputable source to back up your claim. Even
the premium fan manufacturers themselves acknowledge
otherwise so you're either quite ignorant and stupid for
drawing a conclusion while ignorant, or just trolling.
 
Back
Top