Date: Tue, 01 Oct 1996 11:03:32 -0700 From: "Michael L. VanLoon -- HeadCandy.com" <michaelv@MindBender.serv.net> To: "Rodney W. Grimes" <rgrimes@gndrsh.aac.dev.com> Cc: narvi@haldjas.folklore.ee, obrien@nuxi.cs.ucdavis.edu, freebsd-hardware@freebsd.org Subject: Re: H/W recommendation Message-ID: <199610011803.LAA18291@MindBender.serv.net> In-Reply-To: Your message of Tue, 01 Oct 96 09:08:50 -0700. <199610011608.JAA19605@GndRsh.aac.dev.com>
next in thread | previous in thread | raw e-mail | index | archive | help
>> Even Memory PCI bus
>> Multiple bus speed speed
>> MHz MHz MHz
>> -------- --------- -------
>> 33.3 66.67 33.3
>> 30.0 60.00 30.0
>> 25.0 50.00 25.0
>If you guys are going to talk about this at least get the FACTS right
>before you make 3 or 4 erronious posts...
I have all the FACTS right -- I know exactly what I am talking about.
I typo'd the original post (when I accidentally typed PCI where I
meant memory bus), which is different from not getting the "facts"
right.
>``Even multiplier misses
>the fact that there are 1.5, and 2.5 multipliers!!!''.
I never once mentioned the word "multiplier". I did not discuss the
motherboard clock multiplier setting. I discussed multiples of a
value. 100 is an even multiple of 33 1/3. It's true that a 100MHz
Pentium generally runs at 1.5 * 66 2/3 MHz, but that's not what I was
discussing. The original post specifically asked about chip speeds
and bus speeds, not multiplier settings.
>Memory bus
>speed should generally not be specified with 2 digits beyond the
>decimal point as it depends heavely on the clock chip and/or oscillator
>used in a design.
I was specifically trying to make a general observation. That's why
the first time I wrote the numbers, I wrote "33 1/3" and "66 2/3".
Unfortunately, it's very awkward to write fractions in ascii test,
especially when following them by a unit, such as "MHz". So, I used
33.33 and 66.67 to approximate 33 1/3 and 66 2/3. Moreover, if those
aren't fairly close to the real figure, and we're truly talking about
33 and 66MHz, then we're also talking about 132MHz, 165MHz, and
198-199MHz Pentiums. I doubt there are any decent motherboard
manufacturers out there really doing something stupid like that.
Using "166" and "133" is merely a convenient way for them to say "133
1/3" and "166 2/3".
>Here, express it the way that most motherboard documents do:
That's fine. Very nice table, but not as summary-oriented as I had
attempted to be. I figured if they wanted multiplier settings for
their motherboard, it would be just as easy for them to look it up in
their motherboard's manual, as to read it here.
>Rated External Clock External to PCI Bus
> CPU and Memory Bus Internal Clock Clock
> MHZ MHZ** Multiplier MHZ
>
> 75 50 1.5 25
> 90 60 1.5 30
> 100 50* 2 25
> 100 66 1.5 33
> 120 60 2 30
> 133 66 2 33
> 150 60 2.5 30
> 166 66 2.5 33
> 180 60 3 30
> 200 66 3 33
>
>* The Pentium 100 can be run at either 50MHz external clock with a
> multiplier of 2 or 66MHz and a multiplier of 1.5.
True, it CAN be, but I'd feel sorry for the fool who bought such a
machine.
>** 66 Mhz may actually be 66.667 MHz, but don't assume so.
See above...
>As can be seen the best parts to be using are the 100, 133, 166 and 200,
>with the exception that at a mulitplier of 3 the CPU starves for memory.
Yes, it's been said that the P5/200 runs barely faster than a P5/166
in many benchmarks, because the bus is saturated.
-----------------------------------------------------------------------------
Michael L. VanLoon michaelv@MindBender.serv.net
--< Free your mind and your machine -- NetBSD free un*x >--
NetBSD working ports: 386+PC, Mac 68k, Amiga, Atari 68k, HP300, Sun3,
Sun4/4c/4m, DEC MIPS, DEC Alpha, PC532, VAX, MVME68k, arm32...
NetBSD ports in progress: PICA, others...
-----------------------------------------------------------------------------
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199610011803.LAA18291>