• 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.

A64 memory divider

There is no fsb, so you can't be dividing that, and from my understanding htt is just the system reference clock so you can't be dividing that either. I know the formula to determine memory speed is CPU speed/(CPU Multiplier/RAM Divider). I am just trying to figure out what you are running in ratio to. Because according to the formula its not to anything which means that there should be a 0% performance penalty by running a divider on an A64 but I know that's not true for there is a small performance penalty. Does my question make any sense?
 
the divider is essentially the number that divides your clockspeed to get the memory speed

for example, i'm running at 2.6 ghz and use a 13 divider, my memory runs at 200mhz

the hard part is figuring out what the divider will be

I use the 7:10 ratio

10/7=1.42
my cpu multiplier is 9x
9x1.42=12.8

it always rounds up so you get 13

thats how you calculate the divider, their are problems with half multipliers but as long as you know the ratio's (ie 1:2, 2:3, 3:4, 4:5, 7:10) you can figure out what dividers are available
 
and the reason dividers don't impact performance significantly is that the k8 isn't bandwith starved, look at the memory matrix and you'll see that low latency is better than high bandwith
 
Back
Top