tripple_x

Distinguished
Aug 22, 2003
11
0
18,510
Hi,

I recently set-up a P4 3Ghz 800FSB system, my question is about the speed of the FSB, I ran a 3D Mark test and the end results stated my FSB was running at 200.

How can I change this from 200 to 800, or is the 3D Mark result wrong? Is there a tool that I can use to see the system performance that reports things like the FSB speed?

P4 3Ghz 800fsb CPU
Abit IS7- i865PE Springdale 800fsb DDDR Mobo
512Mb PC3200 (PC400) DDR Memory
Creative 64Mb GeForce 2 Pro
Windows XP Pro

I have Duplicated this message on the CPU board.

Thanks in advance for any advice!
 

Prof133

Distinguished
Apr 20, 2003
1,329
0
19,280
<A HREF="http://forumz.tomshardware.com/hardware/modules.php?name=Forums&file=viewtopic&p=521622#521622" target="_new">Source</A>:
Intel uses a QDR bus, the numbers are inflated by 4x their actual frequencies to show “performance equivalence”. QDR transfers four bits per path per cycle instead of one.
 

tripple_x

Distinguished
Aug 22, 2003
11
0
18,510
I really should have checked the FAQ.. sorry about that,

So just to clarify 200mhz FSB is correct as its 200x400 = 800fsb?.. sounds right to me.

thanks for the help! :)
 

Crashman

Polypheme
Former Staff
Take your AGP cards as an example. AGP2x was part of the original AGP 1.0 standard. It sent data 2x per clock cycle. AGP4x sent data 4x per clock cycle.

AMD started selling Athlons with a Double Data Rate (DDR) bus, which is esentially the same technology that makes AGP2x..2x! And they lied by calling the bus "200MHz" instead of 100MHz or DDR200.

So successfull was this lie that when Intel released a Quad Data Rate (4x) CPU bus, they HAD to call it 400MHz, even though it was really 100MHz in clock cycles. This eventually lead to the 800 bus, which operates at 200MHz in clock cycles.

Now, it's been argued that this isn't really lying nor giving "performance equivalents", but simply using Mega Transfers Per Second as "MHz" rather than Mega Cycles Per Second of the clock generator. Although MHz means "Mega Cycles Per Second" the argument goes that since data flow is basically a cycle, MHz is an adequate and accurate lable for these inflated numbers.

<font color=blue>Only a place as big as the internet could be home to a hero as big as Crashman!</font color=blue>
<font color=red>Only a place as big as the internet could be home to an ego as large as Crashman's!</font color=red>