neo4 and dividers

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

ilikemovie

Member
Apr 7, 2001
120
0
0
I was trying different memory setting this afternoon. I figure out that if I change the 3 memroy setting in the bios, it allows me to to change the diverder without any problem. They are tras, trcd and trp. I had them set to 10-6-6 and my devider started to work. I have the bios 1.1 as well. scoobydoo, i guess i have a bad chip too. I cannot get my chip pass 2.2 stable. Dimitris, play with the 3 mem setting and see whether you can get your chip higher. If yes, congratulation. I am stuck with 2200 currently. What memory do you guys use? I have a feeling that this MB (or any nforce4 MB, I dont know) is kind of picky with memory. I am using cosair value select pc3200. My current setup

FSB: 220
CPU: 2200 (10x)
RAM: 220
HTFreq: 5x
WTWidth: 16 x 16
 

dimitris

Member
Jan 22, 2005
34
0
0
i will try when i return home .. right now i am sipping coffee and writing on my pda.

BUT i doubt setting these latencies so high (6-6-10) will help. the whole point of dividers is to pump up the fsb while keeping mem relatively down so it can have good latencies...

right now i am at 295 fsb 180 divider 265 memory 2.5-3-3-8

i am using crucial ballistix pc4000 2x512mb
 
Feb 5, 2005
43
0
0
Originally posted by: ilikemovie
I was trying different memory setting this afternoon. I figure out that if I change the 3 memroy setting in the bios, it allows me to to change the diverder without any problem. They are tras, trcd and trp. I had them set to 10-6-6 and my devider started to work. I have the bios 1.1 as well. scoobydoo, i guess i have a bad chip too. I cannot get my chip pass 2.2 stable. Dimitris, play with the 3 mem setting and see whether you can get your chip higher. If yes, congratulation. I am stuck with 2200 currently. What memory do you guys use? I have a feeling that this MB (or any nforce4 MB, I dont know) is kind of picky with memory. I am using cosair value select pc3200. My current setup

FSB: 220
CPU: 2200 (10x)
RAM: 220
HTFreq: 5x
WTWidth: 16 x 16

I think that the reason why you can't get past 2200 is because you have left your HTFreq at 5x. You do realise that the HTT (FSB x HTFreq) has to be around 1000MHz otherwise the system becomes unstable don't you? Why don't you try lowering your HTFreq to 4x and upping the FSB a bit more?

I apologise if you know this already!

 

ilikemovie

Member
Apr 7, 2001
120
0
0
scoobydoo

Thank you so much for your tip ((FSB x HTFreq) has to be around 1000MHz). That really help me a lot. After I read your post, I took out my calculator and started to so calculation. Now, I could not be happier with my setup. Thank you very much !!!!!!!

my setup
FSB: 330
CPU: 2640 (8x)
RAM: 231
HTFreq: 3x
WTWidth: 16 x 16
Divider: 140Mhz/0.7

Again, thank you scoobydoo and thank you everybody. Dimitris, good luck !!
 

dimitris

Member
Jan 22, 2005
34
0
0
nop on me divider 166 (which i want) does not work with any mem timings...

@ilivemovie : nice could you tell me if you just set up these settings and it did work .. or did you have to go up (in the bios) in small increments ?

also, does cpu-z recognize 140 divider ? i have 180 and cpu-z does not recognize it .. memory frequency box is shaded ...
 
Feb 5, 2005
43
0
0
Just a quick update:

I'm now up to 2430Mhz, (270 x 9), running 4x HT frequency (1070MHz HTT). It's been running Prime (In place large settings, max heat, etc) for about 9 hours so far and it hasn't conked out yet. Voltage has been whacked up to 1.664v (although is fluctuating, the lowest goes to is 1.632v). My memory is at 220MHz at this setting (166MHz I think) and at 2.85v (7-3-3 2.5). Northbridge voltage upped slightly to 1.55v (don't think that this makes much of a difference). Max load temperature is 51C. (Watercooled).

I've pushed the chip as far as nearly 2700, but isn't very stable at those settings. If only my 3x HT freq multiplier worked!!

Edit: Forgot to say that I'm running beta bios 1.21
 

dimitris

Member
Jan 22, 2005
34
0
0
try S&M for max load temp and great mem checking ... found me some errors even memtest86 and prime didn't ...




p.s. : this app GREATLY loads cpu .. keep an eye on temps

p.s. 2 : with this app and at 2655Mhz my 3000+ has max temp of 47 degress (air)
 

arswihart

Senior member
Jul 16, 2001
541
0
0
ilikemovie - good job straightening things out, told ya those memory settings could help, and I forgot to mention about lowering the htt multiplier

dimitris - too bad those other dividers aren't working, but as ilikemovie said, try relaxing your memory timings (and tras should be 10/11 for best performance anyways), and see if that helps, also try dropping your htt multiplier if you can. The most important thing is 1t, the other timings can usually be relaxed so that the memory freq. can be raised more, then you can do some benchmarks to compare performance (low latency vs. high bandwidth). The htt doesn't HAVE to be near 1000, a lot of people have found equivalent performance, all else equal, when switching the htt from, say 255 x 4 to 255 x3.
 

dimitris

Member
Jan 22, 2005
34
0
0
i know about HTT.. about tRAS i know it was 10 optimal at nForce3 but I've set it at 8 according to anand's article of nForce4 (DFI). seems 8 is optimal for nForce4.

i'm doing benches right now to see low latency vs high bandwidth. mem passes through memtest ok but crashes at S&M @264Mhz using 180 divider. I really don't want to lower my cpu (1:1 divider)..... and my htt is at 3x (3x294 which i believe is 882)...

anyway .. maybe it's my mobo
 

dimitris

Member
Jan 22, 2005
34
0
0
well i've set my settings at 3-4-4-10 166 divide ron a longshot and system booted .. when in windows benched and then pressed restart .. system did not boot and stuck again at processor initialization forcing me to bios reset

setting 3-4-4-10 again and 166 divider does not boot .. again lockup and bios reset. .. must be mobo's fault .. all other explanations are non-logical
 

ilikemovie

Member
Apr 7, 2001
120
0
0
Hi dimitris,
After I set my mem setting to tras=10, trcd=6 and trp=6, i was able to use divider 120/0.6, 133/0.66, 140/0.7, 150/0.75, 166/0.83 and 180/0.9. I did not know how the mem setting could affect the devider but it worked for me. Thus, I am going to stick with 10-6-6. Try 10-6-6 and see wheter it works for you. If not, play with your mem. That was just my experience (may be I did not know what I was doing, hehehe). Try to keep your mem around 250Msh (just like what arswihart mentioned). After I Figure out my divider was working, I increased my FSB=260, FTFreq=4, CPUMult=10, MEMDev=0.83. Then I tried FSB=330, FTFreq=3, CPUMult=8, MEMDev=0.66. I did go up (in the bios) in big increments. Cpu-z does not recognize 140 devider. My memory frequency box is shaded as well. On, one more thing, try to change your Cas Latency to 2.5. Neo4 does not like
my value select to run at Cas Latency 3.

scoobydoo
Dont you think your CPUVolt is a little too high?
 

dimitris

Member
Jan 22, 2005
34
0
0
i understand what you are saying but i now play @

294FSB 180 Divider (mem @264Mhz) with 2.5-3-3-8

all i am doing is changing the f$@!% divider .. so instead of 264 it will go to 25x or 24x .... that has nothing to do with increasing timings (more to do with decreasing them but i am not doing that)

i am going nuts with my board...

i think 3-6-6-10 which you use is kind of high especially for the two 6s. they should hurt your performance quite a bit.
btw what are you using for Cas Latency ?
 

ilikemovie

Member
Apr 7, 2001
120
0
0
try to change your Cas Latency to 2.5. Neo4 does not like my value select to run at Cas Latency 3. When I set my cas to 3, Board would not boot at all. I dont know why. I am using cas 2.5. Try to relax you mem setting a bit. Once you get your devider to work, then try to optimize your mem, bit by bit
 
Feb 5, 2005
43
0
0
I know that my volts are a little bit high, but it's running at max 55C under S&M (& I'm using watercooling), it seems to be stable at 2430, although I am hoping for more. It seems so illogical (and frustrating) that 4x (HT Freq) will work up to 295MHz, but 3x won't work at all! I think that it might work at some ridiculous low FSB, but that's pretty useless...
 

dimitris

Member
Jan 22, 2005
34
0
0
further testing :

I set up everything at stock ..

then set HTT 3x and FSB 210 (166 divider) OK !
FSB 220 OK !
FSB 230 OK !
FSB 240 OK !
FSB 250 no boot !
reset
FSB 241 no boot !
reset
fsb 240 ok !

seems divider 166 works only with FSB <=240 (tried with more than 15 different configs of mem timings and HTT 3x, 4x)

RMA ?
 

ilikemovie

Member
Apr 7, 2001
120
0
0
dimitris,
I am pretty sure that devider 166 works > 240 because I tried 260 fsb with devider 166 and it worked. Did you change your cas to 2.5?
 

dimitris

Member
Jan 22, 2005
34
0
0
yes cas is always 2,5

i am not saying that it doesn't work for everybody

maybe i have a faulty mobo
 
Feb 5, 2005
43
0
0
Another update:

I'm now at, 2501MHz@1.710v (!) (although it does undervolt by ~0.5v)
I'm running at a FSB of 278MHz, HT Freq. is at 4x, memory divider set at 166MHz (RAM speed ~230MHz), 8-3-3 CAS 2.5 and at max 2.85v
Northbridge voltage I've left alone as I have a Zalman Passive NBJ45 (blue) on there right now and even at default volts seems to get almost too hot to touch after a heavy session.

I haven't had the chance to give it 24 hours of Prime95 just yet, but I've given it a good session of S&M (v1.5.1) and that seemed to max out the CPU temp at 59C (ambient room temp is 26C). I know that 59C is a little high, but to be honest, I doubt that anything will ever push the CPU to that much heat. S&M seems to do someting else to achieve such a high temp. I'll leave it on overnight running Prime95 to hopefully test full stability.

I actually reset my board back to defaults to test out some stuff. In my case, HT Freq. 1x, 1.5x, 2x, 2.5x all work up to 330MHz. I'd settle for this, but unfortunately, the only memory divider that works for me at such a high speed is 100MHz (0.5)! Which sucks! I've also found that 3x HT Freq works only as far as 257MHz (which I can't figure out why, esp as 4x works all the way up to 295MHz) - even if I push the max amount of volts through the CPU (tried all the volt ranges inbetween too). The only mem dividers that I can seem to use at the higher speeds (>250MHz FSB) are either 100MHz or 166MHz. All the others don't work! Maybe it might just be my RAM?

I think that I'm calling it a day for now (at least until MSI release a new bios revision), I'm at 2.5GHz, which is what I wanted, good luck to everyone else...

Oh yeah, before I forget, I alos posted this in the other thread, but have any of you managed to successfully boot from a USB key/stick? I have my Sandisk Cruzer Micro (which I know is bootable), but none of the bios options (USB-FDD, USB-Zip, etc) (or 'Boot Other Device') seem to allow me to boot it - but it detects that that stick is there on bootup. Even going through all the boot options on pressing F11 on startup, yields no luck.
 

arswihart

Senior member
Jul 16, 2001
541
0
0
dimitris, you obviously are doing things differently all the time and randomly try things, change things, come back with magical results, I think you just need to learn more about how to OC correctly, I don't think there's anything wrong with your f'in board
 

dimitris

Member
Jan 22, 2005
34
0
0
yes you are right, i apologize...

in the meanwhile try to figure this out jack'...

mem is officially rated 250Mhz 2.5-4-4-8

This works : FSB 250, Divider 200, timings 2.5-4-4-8, effective ram 250Mhz
This works : FSB 278, Divider 180, timings 2.5-4-4-8, effective ram 250Mhz
This is unbootable : FSB 300, Divider 166, timings 2.5-4-4-8, effective ram 250Mhz

all were done with HTT 3x and CPU 7x...

oh, and please try to understand that lowering mem frequency DOES NOT require relaxing timings.... upping frequency DOES !

if you read previous postings you will see there is nothing random with what i try .. maybe you're too clever to get it ... and it appears there is a problem with my f'in board but i can live with it @

FSB 290,
CPU 9x (2610)
HTT 3x (870)
Mem Divider 180 2.5-3-3-8 effective 261Mhz.

so i will stay here even though i know that if i had a 166 divider working i would get 100 Mhz more out of my cpu

ciao
 
sale-70-410-exam    | Exam-200-125-pdf    | we-sale-70-410-exam    | hot-sale-70-410-exam    | Latest-exam-700-603-Dumps    | Dumps-98-363-exams-date    | Certs-200-125-date    | Dumps-300-075-exams-date    | hot-sale-book-C8010-726-book    | Hot-Sale-200-310-Exam    | Exam-Description-200-310-dumps?    | hot-sale-book-200-125-book    | Latest-Updated-300-209-Exam    | Dumps-210-260-exams-date    | Download-200-125-Exam-PDF    | Exam-Description-300-101-dumps    | Certs-300-101-date    | Hot-Sale-300-075-Exam    | Latest-exam-200-125-Dumps    | Exam-Description-200-125-dumps    | Latest-Updated-300-075-Exam    | hot-sale-book-210-260-book    | Dumps-200-901-exams-date    | Certs-200-901-date    | Latest-exam-1Z0-062-Dumps    | Hot-Sale-1Z0-062-Exam    | Certs-CSSLP-date    | 100%-Pass-70-383-Exams    | Latest-JN0-360-real-exam-questions    | 100%-Pass-4A0-100-Real-Exam-Questions    | Dumps-300-135-exams-date    | Passed-200-105-Tech-Exams    | Latest-Updated-200-310-Exam    | Download-300-070-Exam-PDF    | Hot-Sale-JN0-360-Exam    | 100%-Pass-JN0-360-Exams    | 100%-Pass-JN0-360-Real-Exam-Questions    | Dumps-JN0-360-exams-date    | Exam-Description-1Z0-876-dumps    | Latest-exam-1Z0-876-Dumps    | Dumps-HPE0-Y53-exams-date    | 2017-Latest-HPE0-Y53-Exam    | 100%-Pass-HPE0-Y53-Real-Exam-Questions    | Pass-4A0-100-Exam    | Latest-4A0-100-Questions    | Dumps-98-365-exams-date    | 2017-Latest-98-365-Exam    | 100%-Pass-VCS-254-Exams    | 2017-Latest-VCS-273-Exam    | Dumps-200-355-exams-date    | 2017-Latest-300-320-Exam    | Pass-300-101-Exam    | 100%-Pass-300-115-Exams    |
http://www.portvapes.co.uk/    | http://www.portvapes.co.uk/    |