Does a high FSB really matter??

Archived from groups: alt.comp.periphs.mainboard.abit (More info?)

Hi Folks,
There's a thread going in this group asking about the performance gains
of overclocking an NF7. I'm starting a new thread along these lines as
it may be of interest generally.
I'm basically stating that IMHO, one shouldn't get too worked up about
having super fast FSB speeds and pushing your RAM to the limit (and
possibly forking out loads of dosh on ultra high speed stuff that you
might not really need). THE most important speed in a given system is
the actual internal CPU clock speed. (I obviously appreciate that in
some scenarios, the two are linked)
As a very rough and ready test, (I didn't have too much time to spend on
this when I did it) I wanted to use a real life application that could
use all the bags of processing power it could get. I chose 'Reaktor' as
I was working on it at the time. (For those that don't know, its an
audio synthesiser package that allows you to build your own virtual
instruments that work in real time).
As I had an unlocked Barton, it was easy to alter the multiplier and
FSB.
Reaktor has a '%CPU utilisation' meter. A good real life indicator of
what's going on under the hood. Try too complex a patch, get up to 100%
and you can't go any further!
Take a look a the figures below. A summary of the findings is this
however:
As you increase the CPU clock, the %CPU utilisation goes down as you
would expect. It varies quite a lot, an 800MHz original athlon is
peaking out at 100%. Switch to the Barton at 1.46GHz and its gone down
to 32%.
Now change the Barton speed. At 1.8GHz its only using 25% CPUU and at
2.31GHz, its only 19%. This is at a DDR420MHz FSB speed.
Useful differences.
Now if you switch to 2.3GHz, but use only a DDR206 FSB (under half), and
crank the multiplier right up, you only lose 1%
Bearing in mind that's one extreme to the other, can you see my point??

I also thought it would be interesting to have a quick look at how games
would be affected, so I ran 3Dmark2001 on these two extremes.
I was using a Ti4200.
Remember that I'm not really interested in synthetic benchmarks here, I
want real life speed improvements. Lots of the tests do rely on pure
data throughput. The only meaningful numbers are the FPS readings. Game
1 (dragothic?) does shift a fair amount data. Nature is heavily GPU
based.
Again, bear in mind that I'm only using the extreme FSB speeds.
How much difference would there be using say a gig of normal £130 PC3200
compared to £340 of PC4400 ?? a few percent ??? Are you really going to
notice this outside the benchmark sheet??
Remember, I DO realise sometimes you have to get high FSB speeds to push
the CPU speed up. But sometimes you don't ;-)

Hope this was interesting.


*************************************************************************
Quick test to see the effect of raw CPU megahertz compared to
varying the RAM/FSB speed.
The %CPU score is the processor utilisation running a standard (complex)
patch in Reaktor.

My Asus A7V600 with unlocked Barton XP2500+

Actual CPU spd GHz FSB Mult RAM %CPU
2.31 420 11 210 19
2.32 333 14 166 19
2.30 256 18 128 20
2.30 206 22.5 103 20

1.8 400 9 200 25
1.83 333 11 166 25
1.8 266 13.5 133 26

1.46 266 11 133 32

Interestingly, on My Athlon slotA 800MHz
0.8 200 100 approx 100%


I also compared:
result 1 (flat out FSB)
result 4 (really throttled back slow RAM)
With some 3Dmark2001 benchmarks:
3DMark Game1 low Game1 high Nature
1 9811 158.3 60.5 41.1
4 11388 179.4 77.8 42.5
--
__________________________________________________
Personal email for Gareth Jones can be sent to:
'usenet4gareth' followed by an at symbol
followed by 'uk2' followed by a dot
followed by 'net'
__________________________________________________
6 answers Last reply
More about does high matter
  1. Archived from groups: alt.comp.periphs.mainboard.abit (More info?)

    "Gareth Jones" <usenet@nospam.demon.co.uk> wrote in message...
    > I'm basically stating that IMHO, one shouldn't get too worked up
    > about having super fast FSB speeds and pushing your RAM to the
    > limit (and possibly forking out loads of dosh on ultra high speed stuff
    > that you might not really need). THE most important speed in
    > a given system is the actual internal CPU clock speed.

    Erm, nope. It's far too simplistic to claim that CPU speed is the be-all and
    end-all, and that memory speed is less important.

    > (I obviously appreciate that in some scenarios, the two are linked)

    In reality all parts of the PC are linked more often than not. In this case,
    the bandwidth between the CPU and memory bus, and the bandwidth between the
    memory bus and the memory both contribute directly to the overall
    performance, although it's fair to say that the reliance on memory bandwidth
    is greater with some applications than it is with others.

    > Hope this was interesting.

    Lol, a jumping-off point for discussion if nothing else. :-)
    --


    Richard Hopkins
    Cardiff, Wales, United Kingdom
    (replace .nospam with .com in reply address)

    The UK's leading technology reseller www.dabs.com
    Get the most out of your digital photos www.dabsxpose.com
  2. Archived from groups: alt.comp.periphs.mainboard.abit (More info?)

    As I've said previously I largely agree with you but:
    1)PC4400 (275fsb) on an AMD system is ludicrous & worse will have really
    loose timings - better a lower fsb & tighter timings (& wouldn't cost £340
    either).
    2) I imagine that your gpu will be a bottleneck for some of the tests &
    therefore not able to show a difference that may exist.

    Most of us know that we are talking of performance differences probably <5%
    & it's getting the balance between clockspeed & fsb not all of 1 & none of
    the other : )

    "Gareth Jones" <usenet@nospam.demon.co.uk> wrote in message
    news:5qidfhEIkktAFwWy@nospam.demon.co.uk...
    > I also compared:
    > result 1 (flat out FSB)
    > result 4 (really throttled back slow RAM)
    > With some 3Dmark2001 benchmarks:
    > 3DMark Game1 low Game1 high Nature
    > 1 9811 158.3 60.5 41.1
    > 4 11388 179.4 77.8 42.5

    If I'm reading this correctly there is something very odd going on here
  3. Archived from groups: alt.comp.periphs.mainboard.abit (More info?)

    I think this article supports your position.

    http://www.tomshardware.com/motherboard/20040119/index.html

    I would wonder about the application(s) being used for evaluating
    memory performance. Wouldn't audio/video/photo editing would be more
    appropriate. Or at least, an application (or application operation)
    where performance is directly related to memory speed.

    I think you would find memory speeds to be more critical in non home
    scenarios like: (1) internet router/switch (2) telephone switch.

    What about hard disk access speed? Seems like it would be the most
    important system spec - as it is the slowest component by several
    orders of magnitude. Many have converted to Raid 0 set ups.

    I would agree that others are probably chasing there tails when it
    comes to memory. Perhaps its just for the fun of it :-)

    Forrest

    Motherboard Help By HAL web site:
    http://home.comcast.net/~hal-9000/


    On Thu, 27 May 2004 21:01:12 +0100, Gareth Jones
    <usenet@nospam.demon.co.uk> wrote:

    < snip >
    > THE most important speed in a given system is
    >the actual internal CPU clock speed.
    < snip >
  4. Archived from groups: alt.comp.periphs.mainboard.abit (More info?)

    In message <40b68262$0$20509$cc9e4d1f@news-text.dial.pipex.com>, Richard
    Hopkins <richh@dsl.nospam.co.uk> writes
    >"Gareth Jones" <usenet@nospam.demon.co.uk> wrote in message...
    >> I'm basically stating that IMHO, one shouldn't get too worked up
    >> about having super fast FSB speeds and pushing your RAM to the
    >> limit (and possibly forking out loads of dosh on ultra high speed
    >>stuff that you might not really need). THE most important speed in
    >> a given system is the actual internal CPU clock speed.
    >
    >Erm, nope. It's far too simplistic to claim that CPU speed is the
    >be-all and end-all, and that memory speed is less important.

    Erm.... Well prove me wrong then. In a typical modern day PC, I'd be
    interested to see an application where the memory speed is MORE
    important than CPU speed.

    >
    >> (I obviously appreciate that in some scenarios, the two are linked)
    >
    >In reality all parts of the PC are linked more often than not. In this
    >case, the bandwidth between the CPU and memory bus, and the bandwidth
    >between the memory bus and the memory both contribute directly to the
    >overall performance, although it's fair to say that the reliance on
    >memory bandwidth is greater with some applications than it is with others.
    >
    >> Hope this was interesting.
    >
    >Lol, a jumping-off point for discussion if nothing else. :-)

    That's why I posted!
    The simple tests I ran were something I did a couple of months ago as I
    needed to have some ballpark idea on what was the cheapest option to get
    a good performance machine (specifically for music apps in this case).

    Don't get too worked up because the weather isn't as nice today where
    you are ;-)


    --
    __________________________________________________
    Personal email for Gareth Jones can be sent to:
    'usenet4gareth' followed by an at symbol
    followed by 'uk2' followed by a dot
    followed by 'net'
    __________________________________________________
  5. Archived from groups: alt.comp.periphs.mainboard.abit (More info?)

    In message <tjvtc.341$%31.140@newsfe1-win>, BUFF <NOSPAM@HERE.COM>
    writes
    >> 3DMark Game1 low Game1 high Nature
    >> 1 9811 158.3 60.5 41.1
    >> 4 11388 179.4 77.8 42.5
    >
    >If I'm reading this correctly there is something very odd going on here

    Oops... Yea, I must have reversed the 3Dmark score. The slower (number4)
    should be the 9811.
    Well spotted.


    --
    __________________________________________________
    Personal email for Gareth Jones can be sent to:
    'usenet4gareth' followed by an at symbol
    followed by 'uk2' followed by a dot
    followed by 'net'
    __________________________________________________
  6. Archived from groups: alt.comp.periphs.mainboard.abit (More info?)

    In message <svldb0t1qvlavv0vsgn59qe46l79e3i2fr@4ax.com>, - HAL9000
    <gumpy@mail.org> writes
    >I think this article supports your position.
    >
    >http://www.tomshardware.com/motherboard/20040119/index.html

    Cheers for that, I'd missed that one. Interesting.

    >
    >I would wonder about the application(s) being used for evaluating
    >memory performance. Wouldn't audio/video/photo editing would be more
    >appropriate. Or at least, an application (or application operation)
    >where performance is directly related to memory speed.

    You're right. I didn't do the benchmark specifically for the posting
    though, I just happened to do it a while back for personal data. Maybe
    is I get a chance I'll do it for MPEG encoding or something similar, but
    I still maintain that for the vast majority of people, most tasks don't
    really care about RAM speeds and you're unlikely to notice any
    difference (outside the benchmark result sheet!)

    >What about hard disk access speed? Seems like it would be the most
    >important system spec - as it is the slowest component by several
    >orders of magnitude. Many have converted to Raid 0 set ups.

    That's something else and outside the scope of the posting.
    Again I agree and all my speed critical machines are now using striped
    raid.
    I've really grown to have a dislike for all these published benchmarks.
    I find most of them are misleading and need a further stage of mental
    calculation to get them into any meaningful perspective.
    FWIW, the sustained data transfer on my Asus A7V600 has gone up from
    approx 55MB/s using single PATA to around 100MB/s using Raid0
    At some stage I'll want to know what this means in terms of how many
    simultaneous audio tracks can be handled by Cubase or ProTools on my
    machine as that is the bottom line in terms of my personal benchmarking
    needs (apart from every app opening so darn fast :-)

    Whatever, Raid0 is certainly worth it speed wise, as you pay very little
    for the benefit.


    >
    >I would agree that others are probably chasing there tails when it
    >comes to memory. Perhaps its just for the fun of it :-)
    >

    I suspect we agree again!
    And while there's nothing wrong with that (and I have to confess I'm a
    demon tweaker myself ;-) I'm sure some other people read all this stuff
    about how vital a high FSB is and then spend loads of cash or fret
    unnecessarily if they can't achieve it.

    --
    __________________________________________________
    Personal email for Gareth Jones can be sent to:
    'usenet4gareth' followed by an at symbol
    followed by 'uk2' followed by a dot
    followed by 'net'
    __________________________________________________
Ask a new question

Read More

Motherboards