• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

running nforce2 async memory

M

MTB2Live Live4Comps

[H]ard|Gawd
Joined
Dec 27, 2002
Messages
1,968
hello guys
i know that running any athlonxp system with the RAM faster than the chip is pointless and sometimes harmful, but what about running the chip faster FSB?
aka using like a 6:5 memory divider? for example, my system right now is 215FSB with 357MHz memory (i should hope the memory can handle 430 at more relaxed timings but regardless, i'm curious)
 
I have found that AMD based systems do not thrive as much on FSB as Intel systems. Instead I have found that AMD systems LOVE tightly timed ram. So my suggestion to you is to keep your ram runnig sync (even if this is lower then the rated speed of the ram) and lower the timings at the 'underclocked' bus. I have low latency ram at 2-2-2-5 (or 2-2-2-11) and I have to say that when I bought this low latency ram I noticed SIGNIFIGANT gains over my old 2-3-3-7 ram.
 
worse performance than running it at 1:1, AMD's prefer tighter timings :)

(it's been said, so i wont get repetitive here)
 
Originally posted by diredesire
worse performance than running it at 1:1, AMD's prefer tighter timings :)

(it's been said, so i wont get repetitive here)
Click to expand...

My testing has showed otherwise on my rig. ????

I run the cpu at 200fsb and the ram at 166fsb. If I lower the fsb on the cpu to match and raise the multiplier, I get lower benches because the cpu is running slightly less ghz...I do not have all of the multipliers half steps so I can't dial it in any closer to 2.4ghz.
 
No what they mean is if you run at say 200/200 FSB/Mem you will get 'x' performance. If you were to then run 200/233 FSB/Mem, even though you have a higher memory clock, you may get worse performance.

This may have changed in the latest revisions so someone out there correct me, but the first nForce II boards ran the best on synchronous buses.
 
Originally posted by reflex33
No what they mean is if you run at say 200/200 FSB/Mem you will get 'x' performance. If you were to then run 200/233 FSB/Mem, even though you have a higher memory clock, you may get worse performance.

This may have changed in the latest revisions so someone out there correct me, but the first nForce II boards ran the best on synchronous buses.
Click to expand...
i'm curious about running single channel, 212/177 FSB/Mem (at 2-2-2-5) verse running 212/212 FSB/Mem (at 2.5-3-3-7) right now i'm running stable on the first choice there, but haven't really run many benchmarks to test it (linux synthetic benchmarks are rather - uh - missing :) )
 
MTB2Live,Live4Comps like I stated in my first post (which diredesire seemed to have completely disregarded :rolleyes: ) you'll prolly find better results with the tighter timings. But hell, its worth testing yourself, may as well try the looser timing settings and let us all know what you find in benchies
 
Originally posted by HiTech-Hate
MTB2Live,Live4Comps like I stated in my first post (which diredesire seemed to have completely disregarded :rolleyes: ) you'll prolly find better results with the tighter timings. But hell, its worth testing yourself, may as well try the looser timing settings and let us all know what you find in benchies
Click to expand...

I dont see how i disregarded your post, i agreed with it... hell, i acknowledged it. i was the 3rd post and i said "it's been said"... Man i feel like everyone's out for me today...
 
I have a fic AU13 also a nforce2 board.

So can all nforce2 boards change their ram and fsb speeds?

My ram is stuck at 133mhz cpu at 333fsb.
 
Check out my sig... i match exactly what u are talking about. nforce2 board, overclock. in my tweaking i have found(as u can see in my sig) that while yes as stated in other replies these boards will run excellent with tighter say 2-2-2-5 timings i cannot even o/c the fsb 1 mhz. when i relaxed it to the settings i have now and at a 1:1 ratio fsb/ram, i was able to acheive a 12 mhz fsb jump and subsequent benchmarks have proven higher with these settings. maybe try the settings i have at 1:1 for yourself and see what happens. GL.

awdark....

it's a bios function. what settings do have for fsb, clock multiplier, ram timings, etc. ? what speed ram do u have 133 is pc2100 check it out an mess around i guess.
 
Originally posted by MTB2Live,Live4Comps
i'm curious about running single channel, 212/177 FSB/Mem (at 2-2-2-5) verse running 212/212 FSB/Mem (at 2.5-3-3-7) right now i'm running stable on the first choice there, but haven't really run many benchmarks to test it (linux synthetic benchmarks are rather - uh - missing :) )
Click to expand...

I would sugest lowering the first setting a few MHz to get it stable rather than running asynchronous. Like 210 or 209 sychronous would be a lot better than 212/177.
 
You must log in or register to reply here.
Back
Top