Archived from groups: alt.comp.hardware.pc-homebuilt (More info?)

Hi folks,

I'm about to start putting together my fourth build, and I'm at the point
where I can use mucho advice. The new mchine will be primarily for CPU-
intensive stuff, such as high quality audio compression and editing or
high definition fractals, and also general "house-office" stuff. I'm not
much into games or fast 3-D video, and am not planning to overclock, or
not extensively.

My present two year old machine, which is getting ncreasingly ragged
around the edges, has a Soyo KT400 Dragon and Athlon 1400. I am looking
for a decent real world speed improvement, but without the hassles of
building to the cutting edge. My highest priority is a stable and trouble
free machine made up of tried and tested components. Speed is my next
highest priority. Third priority is a quiet system. My budget can stand
some moderate upgrading of the components below, if it will mean a better
machine.

Here is my first and very tentative approximation of how to go about it.

Thank you all very much for any input and recommendations about better
alternatives, and for pointing out any bottleneck or overkill.

MOTHERBOARD:
ABIT NF7-S
or ASUS A7N8X-E DELUXE Rev.2.0 ?
(or possibly GIGABYTE GA-7N400 Pro2 ?)
CPU:
Barton XP2800
(or XP2500 OC to FSB 400 ?)
MEMORY: 2x512MB DDR400 operating as dual channel,
HD: Hitachi 250GB 7200RPM
DVD WRITER: Dual Layer, 12X or 16X
FLOPPY/MEMORY CARD READER: MITSUMI FA404A
VIDEO CARD: Chaintech FX20 (GeForce 5200, 128MB)
CASE: EVERCASE E4252

CPU SINK/FAN:
Speeze WHISPER ROCK II, #5F263B1M3G (2700rpm, 38CFM, 26.5dB)
or Speeze "FALCONROCK" #5F286B (2300rpm, 28cfm, 25db)
CASE FANS:
Back: 12cm, NMB 4710NL-04W-B29 (72cfm, 30dB, will use at 7V)
Front: 8cm: Speeze FDC08025S1M, (37.8cfm, 25.0 DB)

--- already have: ---
POWER SUPPLY: Pc Power and Cooling, 410W
HD #2: Hitachi 120GB
SOUND CARD: Revolution 7.1
WinXP SP1 Pro

Archived from groups: alt.comp.hardware.pc-homebuilt (More info?)

> building to the cutting edge. My highest priority is a stable and trouble
> free machine made up of tried and tested components. Speed is my next
> highest priority. Third priority is a quiet system. My budget can stand
> some moderate upgrading of the components below, if it will mean a better
> machine.
>
> Here is my first and very tentative approximation of how to go about it.
>
> Thank you all very much for any input and recommendations about better
> alternatives, and for pointing out any bottleneck or overkill.
>
> MOTHERBOARD:
> ABIT NF7-S
> or ASUS A7N8X-E DELUXE Rev.2.0 ?
> (or possibly GIGABYTE GA-7N400 Pro2 ?)
> CPU:
> Barton XP2800
> (or XP2500 OC to FSB 400 ?)
> MEMORY: 2x512MB DDR400 operating as dual channel,
> HD: Hitachi 250GB 7200RPM
> DVD WRITER: Dual Layer, 12X or 16X
> FLOPPY/MEMORY CARD READER: MITSUMI FA404A
> VIDEO CARD: Chaintech FX20 (GeForce 5200, 128MB)
> CASE: EVERCASE E4252

OK, you have some goals that conflict with each other, and some components
that conflict with your goals. If your power supply is fairly new and has
SATA and P4 connectors like this one here:

http://www.pcpowercooling.com/prod [...] /index.htm

Then that's a good start. Otherwise, I'd suggest a newer power supply.
It's generally not a good idea to recycle a power supply into a new system,
as the power supply is the most frequent failure point, and it is the only
component that is likely to "kill" other components WHEN (not if) it fails.

Stability is your first priority, but oddly you are entertaining the idea of
possibly overclocking. You can have stability or overclocking, one OR the
other. Don't listen to the idiots who will scream that it's possible to do
both. (don't count on being that lucky . . . )

Speed is your 2nd priority, but you are looking at possibly going from
Athlon XP to Athlon XP. Generally, an increase in clock speed will not
translate to a noticeably faster system. If you're going to spend money on
an upgrade, do it in a way that will translate to a faster system.

Quiet is your third goal, but if you stick with the Athlon XP platform
(don't do it), you've already honed in on two motherboards that have really
noisy cooling fans on their northbridge chipsets. A fan on the northbridge
is not necessary, not even for overclocking (which you are NOT going to do,
RIGHT?!?), and it just adds tons of noise to the system. Choose a mainboard
that is passively cooled, you will have plenty of choice there.

If you want to accomplish your goal of higher speed, you need to set your
sights just a tad higher. Look for an 800FSB P4 processor on just about ANY
(passively cooled) motherboard that will handle it. Or if your budget can
stretch that far, jump straight to an Athlon 64 in the ~3200 ~3500 range.
Again, for the Athlon64, find a passively cooled mainboard. While your
choice of video card was a good match for the system you originally
envisioned, it would be somewhat of a bottleneck for anything faster. I'd
suggest something like the Sappire Radeon 9600 or Sapphire Radeon 9600 XT
(be sure to get a passively cooled one!). While those aren't cutting edge
either, they won't be quite as much of a bottleneck as a 5200. Plus, they
are silent. Just be aware that different versions exist, and choose one
without a cooling fan. -Dave

Archived from groups: alt.comp.hardware.pc-homebuilt (More info?)

"Dave C." <mdupre@sff.net> said:

> You can have stability or overclocking, one OR the other. Don't
> listen to the idiots who will scream that it's possible to do both.

Dave, you need to educate yourself
on a topic before calling people idiots.
--
Mac Cool

Archived from groups: alt.comp.hardware.pc-homebuilt (More info?)

> Dave, you need to educate yourself
> on a topic before calling people idiots.
> --
> Mac Cool

If fifteen bazillion people do something stupid, that doesn't mean it's a
good idea for you to do it, also. If someone is looking for stability, they
should NOT be overclocking. Unfortunately, lots of people will probably
advise them to do JUST that. -Dave

Archived from groups: alt.comp.hardware.pc-homebuilt (More info?)

<yogi_berra@neverspamever.org> said:

> The new mchine will be primarily for CPU- intensive stuff, such as
> high quality audio compression and editing or high definition
> fractals, and also general "house-office" stuff. I'm not much into
> games or fast 3-D video, and am not planning to overclock, or not
> extensively.
>
> My present two year old machine ... Athlon 1400.

This was outdated two years ago.

> MOTHERBOARD:

MSI K8N NEO Platinum

> CPU: Barton XP2800 (or XP2500 OC to FSB 400 ?)

Bump up to an Athlon64 3000+ or 3200+
--
Mac Cool

Archived from groups: alt.comp.hardware.pc-homebuilt (More info?)

Dave C. wrote:

>>Dave, you need to educate yourself
>>on a topic before calling people idiots.
>>--
>>Mac Cool
>
>
> If fifteen bazillion people do something stupid, that doesn't mean it's a
> good idea for you to do it, also. If someone is looking for stability, they
> should NOT be overclocking. Unfortunately, lots of people will probably
> advise them to do JUST that. -Dave
>
>

Another case of knowing how to do something making all the difference.

When done properly, overclocked machines are just as stable as any other
system.

Archived from groups: alt.comp.hardware.pc-homebuilt (More info?)

David Maynard wrote:

> Dave C. wrote:
>
> >>Dave, you need to educate yourself
> >>on a topic before calling people idiots.
> >>--
> >>Mac Cool
> >
> >
> > If fifteen bazillion people do something stupid, that doesn't mean it's a
> > good idea for you to do it, also. If someone is looking for stability, they
> > should NOT be overclocking. Unfortunately, lots of people will probably
> > advise them to do JUST that. -Dave
> >
> >
>
> Another case of knowing how to do something making all the difference.
>
> When done properly, overclocked machines are just as stable as any other
> system.

Are you saying that processor makers are arbitrary and capricious when
assigning processor ratings? There are reasons for those ratings. There
are also reasons for assigning spped limits on roads. While someone
might be able to drive somewhat above the speed limit without having
an accident, the probability of an accident occuring is greater if they
exceed the speed limit.

Archived from groups: alt.comp.hardware.pc-homebuilt (More info?)

JK wrote:

>
> David Maynard wrote:
>
>
>>Dave C. wrote:
>>
>>
>>>>Dave, you need to educate yourself
>>>>on a topic before calling people idiots.
>>>>--
>>>>Mac Cool
>>>
>>>
>>>If fifteen bazillion people do something stupid, that doesn't mean it's a
>>>good idea for you to do it, also. If someone is looking for stability, they
>>>should NOT be overclocking. Unfortunately, lots of people will probably
>>>advise them to do JUST that. -Dave
>>>
>>>
>>
>>Another case of knowing how to do something making all the difference.
>>
>>When done properly, overclocked machines are just as stable as any other
>>system.
>
>
> Are you saying that processor makers are arbitrary and capricious when
> assigning processor ratings?

No, I didn't say that and there's nothing 'arbitrary' or 'capricious' about it.

While there are no guarantees, both Intel and AMD routinely sell processors
with speed ratings below the actual capability of the chip.

> There are reasons for those ratings.

Yes there are. Many, in fact, over a range of considerations, of which
clock speed is only one, and part of the overclocking art is knowing what
they are and the tradeoffs involved.

> There
> are also reasons for assigning spped limits on roads. While someone
> might be able to drive somewhat above the speed limit without having
> an accident, the probability of an accident occuring is greater if they
> exceed the speed limit.

Fortunately, processors are not automobiles driven by error prone humans on
a than perfect road system.

Archived from groups: alt.comp.hardware.pc-homebuilt (More info?)

>
> Another case of knowing how to do something making all the difference.
>
> When done properly, overclocked machines are just as stable as any other
> system.
>

If your mechanic tells you not to start your engine with no oil in it, WOULD
YOU? Now what if 15 bazillion other drivers swore up and down that it was
OK to start an engine with no oil in it, and that it runs just fine that
way? Hobbyists like to overclock because they don't know any better. How
do you tell the hacks from the pros? See who advises you to overclock and
who advises you to build a faster computer if that's what you need. Now YOU
guess who the professional is. Hint: It aint the guy telling you to start
your engine with no oil in it, and it for god damned sure isn't the guy
telling you that overclocking is OK because you CAN have a stable system "if
you know how". You won't find anyone with any degree of hardware knowledge
(REAL hardware knowledge) overclocking a system. I don't have to wonder
why, but maybe you should think about that. -Dave

Archived from groups: alt.comp.hardware.pc-homebuilt (More info?)

Dave C. wrote:

>>Another case of knowing how to do something making all the difference.
>>
>>When done properly, overclocked machines are just as stable as any other
>>system.
>>
>
>
> If your mechanic tells you not to start your engine with no oil in it, WOULD
> YOU? Now what if 15 bazillion other drivers swore up and down that it was
> OK to start an engine with no oil in it, and that it runs just fine that
> way?

No. Now you tell me why you ask silly irrelevant questions.

> Hobbyists like to overclock because they don't know any better. How
> do you tell the hacks from the pros? See who advises you to overclock and
> who advises you to build a faster computer if that's what you need. Now YOU
> guess who the professional is.

I always get a kick out of people who define "professional" as 'the ones
who agree with me'.

> Hint: It aint the guy telling you to start
> your engine with no oil in it, and it for god damned sure isn't the guy
> telling you that overclocking is OK because you CAN have a stable system "if
> you know how". You won't find anyone with any degree of hardware knowledge
> (REAL hardware knowledge) overclocking a system.

You're wrong.

> I don't have to wonder
> why, but maybe you should think about that. -Dave

While you're wondering maybe you should also wonder about learning
something of, as you put it, "REAL hardware knowledge."

Archived from groups: alt.comp.hardware.pc-homebuilt (More info?)

Hey David Maynard,
are you the David I know? Do you work in the MSSG?

==============
Posted through www.HowToFixComputers.com/bb - free access to hardware troubleshooting newsgroups.

Archived from groups: alt.comp.hardware.pc-homebuilt (More info?)

heartofBruce wrote:

> Hey David Maynard,
> are you the David I know? Do you work in the MSSG?

Nope. A different one.

Seems there's a fair number of us too because I get calls for various other
'wrong' versions of 'me' from time to time. hehe

>
> ==============
> Posted through www.HowToFixComputers.com/bb - free access to hardware troubleshooting newsgroups.

Archived from groups: alt.comp.hardware.pc-homebuilt (More info?)

> I always get a kick out of people who define "professional" as 'the ones
> who agree with me'.
>

And I get just as much of a kick out of people who disagree with
professionals as they are sure that they know better than the pros
o. -Dave

Archived from groups: alt.comp.hardware.pc-homebuilt (More info?)

Dave C. wrote:

>>I always get a kick out of people who define "professional" as 'the ones
>>who agree with me'.
>>
>
>
> And I get just as much of a kick out of people who disagree with
> professionals as they are sure that they know better than the pros
> o. -Dave
>

I have no disagreement at all with 'professionals'. It is YOU I have a
disagreement with.

Archived from groups: alt.comp.hardware.pc-homebuilt (More info?)

"Dave C." <mdupre@sff.net> said:

> Now what if 15 bazillion other drivers swore up and down that it was
> OK to start an engine with no oil in it, and that it runs just fine that
> way?

Dave, you are only making yourself look foolish. Educate yourself on
overclocking so that you can at least object intelligently. Yes, there are
potential drawbacks to overclocking, and yes, anyone whose primary concern
is stability should not run their hardware above spec. (They should stick
to intel chips exclusively IMO. Not that AMDs are unstable, but I have
never had a VIA based motherboard that was 100% stable. I don't know how
the nvidia boards are.)

In my experience, if a machine is stable to begin with, it will be stable
when moderately overclocked. I suppose saying so makes me an 'idiot', but
at least I know the difference between putting lubricant on moving parts
and running electronics beyond spec.
--
Mac Cool

Archived from groups: alt.comp.hardware.pc-homebuilt (More info?)

Mac Cool wrote:

> "Dave C." <mdupre@sff.net> said:
>
> > Now what if 15 bazillion other drivers swore up and down that it was
> > OK to start an engine with no oil in it, and that it runs just fine that
> > way?
>
> Dave, you are only making yourself look foolish. Educate yourself on
> overclocking so that you can at least object intelligently. Yes, there are
> potential drawbacks to overclocking, and yes, anyone whose primary concern
> is stability should not run their hardware above spec. (They should stick
> to intel chips exclusively IMO. Not that AMDs are unstable, but I have
> never had a VIA based motherboard that was 100% stable.

If you hate Via chipsets for any reason(even though the current ones are
quite good), you could choose a motherboard for an Athlon XP or Athlon 64
with an Nvidia chipset. So a dislike of Via chipsets should not be a reason
to avoid AMD processors.

> I don't know how
> the nvidia boards are.)

They are also good, as are the current Via chipsets. Intel has had issues
with past chipsets, yet you only want to have a boycott of Via for their
past mistakes, and not a boycott of Intel.

>
>
> In my experience, if a machine is stable to begin with, it will be stable
> when moderately overclocked.

not necessarily true.

> I suppose saying so makes me an 'idiot'

Not an idiot, but perhaps a risk taker? When you drive, do you like
to go above the speed limit.

> , but
> at least I know the difference between putting lubricant on moving parts
> and running electronics beyond spec.
> --
> Mac Cool

Archived from groups: alt.comp.hardware.pc-homebuilt (More info?)

"Mac Cool" <Mac@2cool.com> wrote in message
news:Xns9554DF5CBF56MacCool@24.25.9.43...
> "Dave C." <mdupre@sff.net> said:
>
> > Now what if 15 bazillion other drivers swore up and down that it was
> > OK to start an engine with no oil in it, and that it runs just fine that
> > way?
>
> Dave, you are only making yourself look foolish. Educate yourself on
> overclocking so that you can at least object intelligently. Yes, there are
> potential drawbacks to overclocking, and yes, anyone whose primary concern
> is stability should not run their hardware above spec.

Well, that's a good start. It's not just stability that you need to worry
about. Try doing a study on MTBF and how heat affects it. I'm not talking
system temperature or CPU temperature . . . I'm talking about internal IC
temperatures. Now keep in mind that overclocking affects every component in
your computer, not just the CPU. Even if you have a stable overclocked
system, you have shortened the life of every component in your PC by
overclocking it. That's WITHOUT messing around with voltages. Some
overclockers find it necessary to tweak voltages to achieve stability,
greatly increasing internally generated heat of vital components. So you
add extra cooling, right? Guess what . . . that only gives you a warm fuzzy
feeling when you watch your system temperatures drop. You need the extra
cooling because the ICs are generating more heat internally. In other
words, they are already damaged. The extra cooling does squat.

I know there are people who don't CARE how long their computer lasts, as
they are going to sell it in a year or (better yet) pass it off to their
mother-in-law whom they really don't like much. Overclocking is still a bad
idea for those folks. To "overclock correctly" (now there's an oxymoron),
you need to put a lot of time and effort into researching the proper
components that will not only overclock well, but work well with all other
components when THEY are overclocked. Plus you need to upgrade certain
components, like better cases, more expensive RAM, adding extra cooling fans
or more expensive HSF solutions for your CPU . . . and that's just to
mention a few. If you look at the time and effort (and money) you spend on
"overclocking correctly", you'll find you could have spent the same amount
of money (time is money also, remember) on building a faster system that
runs at factory specifications.

Or as I've said many times, if you need a faster system, build it. -Dave

Archived from groups: alt.comp.hardware.pc-homebuilt (More info?)

Dave C. wrote:

> "Mac Cool" <Mac@2cool.com> wrote in message
> news:Xns9554DF5CBF56MacCool@24.25.9.43...
>
>>"Dave C." <mdupre@sff.net> said:
>>
>>
>>>Now what if 15 bazillion other drivers swore up and down that it was
>>>OK to start an engine with no oil in it, and that it runs just fine that
>>>way?
>>
>>Dave, you are only making yourself look foolish. Educate yourself on
>>overclocking so that you can at least object intelligently. Yes, there are
>>potential drawbacks to overclocking, and yes, anyone whose primary concern
>>is stability should not run their hardware above spec.
>
>
> Well, that's a good start. It's not just stability that you need to worry
> about. Try doing a study on MTBF and how heat affects it. I'm not talking
> system temperature or CPU temperature . . . I'm talking about internal IC
> temperatures. Now keep in mind that overclocking affects every component in
> your computer, not just the CPU. Even if you have a stable overclocked
> system, you have shortened the life of every component in your PC by
> overclocking it. That's WITHOUT messing around with voltages. Some
> overclockers find it necessary to tweak voltages to achieve stability,
> greatly increasing internally generated heat of vital components. So you
> add extra cooling, right? Guess what . . . that only gives you a warm fuzzy
> feeling when you watch your system temperatures drop. You need the extra
> cooling because the ICs are generating more heat internally. In other
> words, they are already damaged. The extra cooling does squat.
>
> I know there are people who don't CARE how long their computer lasts, as
> they are going to sell it in a year or (better yet) pass it off to their
> mother-in-law whom they really don't like much. Overclocking is still a bad
> idea for those folks. To "overclock correctly" (now there's an oxymoron),
> you need to put a lot of time and effort into researching the proper
> components that will not only overclock well, but work well with all other
> components when THEY are overclocked. Plus you need to upgrade certain
> components, like better cases, more expensive RAM, adding extra cooling fans
> or more expensive HSF solutions for your CPU . . . and that's just to
> mention a few. If you look at the time and effort (and money) you spend on
> "overclocking correctly", you'll find you could have spent the same amount
> of money (time is money also, remember) on building a faster system that
> runs at factory specifications.
>
> Or as I've said many times, if you need a faster system, build it. -Dave
>
>

The problem I have with your 'analysis' is that you take a smidgen of
information you've acquired from lord knows where and then extrapolate it
to 'absolutist' hysterical assertions.

For example, you make the wild 'absolutist' claim that "overclocking
affects every component in your computer, not just the CPU." Now, I defy
you to explain how 'every component in the computer' has even the slightest
clue that the processor multiplier has been changed and that, say, an 1833
Mhz processor is being 'overclocked' to 2200 MHz instead of it being a
'legitimate' 2200 MHz processor. Or that a 300 Mhz processor, normally run
on a 66 MHz Bus, is being run 450 at 100MHz FSB isn't a 'legitimate'
450Mhz/100Mhz FSB processor operating at the perfectly normal and
'officially' specified FSB? In those examples the rest of the computer will
not care one WHIT, or even know, because it's only 'connection' is the
perfectly normal FSB.

Your 'absolutist' assertion is just plain not true.

And the rest of your assertions have just as many wildly flawed hysterical
extrapolations.