• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

More multiplier vs More core speed

The Pentium Guy

Diamond Member
What am I aiming for here? Total speed (ex: 1.8 ghz) or more core speed.
Quick, arbitrary example:

200x10 vs
100x20 <-- haha. Just an example.

I'm guessing 200x10 would be better right? See what got me thinking was that they both amount to the same total speed.....

Thanks,
-TPG



 
Higher FSB is better, usually (assuming same overall clock speed like in your example), BUT if you can get a higher overall speed with lower FSB, then sometimes that can be better.
eg: 188x11 = 2070MHz
200x10 = 2000MHz.
With something that likes memory bandwidth, the 200x10 may perform better than the other, but if raw speed is more important, and bandwidth less so, then the 188x11 might be better.
Usually P4's like bandwidth, Athlon XP's like speed (bandwidth doesn't matter too much), and I don't know about Athlon 64's.
 
Clock speed is clock speed, doenst' matter how you reach it. However, depending on what multiplier you use, you different memory options will be available to you (ie, a divider of 1/2 will yeild a different memory speed at a FSB of 215 than at a FSB of 250). The best thing to do is make sure you keep yoru CPU at its highest speed while trying to achieve the highest stable memory speed (through the use of different dividers and different multiplier&FSB's).
 
The best thing to do is make sure you keep yoru CPU at its highest speed while trying to achieve the highest stable memory speed (through the use of different dividers and different multiplier&FSB's).
Gotcha then.

One other question. What exactly is the point of the Multipler. Why can't CPUs just function at, ex: 2.0 ghz?

Thanks.
-The Pentium Guy
 
Motherboards have this clock generator, from which other parts' clocks are derived in set ratios to the main clockgen. So an NF2 board whose FSB runs at 200 sets its AGP clock to 200 * 1/3 = 66, the CPU to 200 * 10 = 2000 (in your hypothetical example), and so on with the other parts.
 
A higher multiplier will allow you to achieve higher CPU core speed without overloading your RAM. A 10x multiplier will bump the CPU core speed by 10MHz for each 1MHz increase in FSB. A 15x multiplier will bump the CPU core speed by 15MHz for each 1MHz increase in FSB. The RAM speed in both cases will only go up 1MHz will 1:1 memory divider.

If both systems are running stable at the SAME CPU core speed, then the one with the higher mutiplier will be a tad slower due to the lower RAM speed. If both systems are running stable at the SAME RAM speed, then the one with the higher multiplier will be faster due to the faster CPU core speed. Again, we're using 1:1 memory divider.
 
Higher multiplier is better because it stresses the motherboard less. Remember that the hypertransport doesnt operate at the HTT frequency, it operates at HTTxmulti Mhz, it's not like Intel's that operates at 200MHz with quad data rate. So, unless you're also overclocking the ram, using a higher HTT will only stress the motherboard without ANY benefit at all.

About multipliers: The cpu derives it's clockspeed from the clock generator, other system devices use this same clock generator to keep their clock speeds synced.
 
Back
Top