My Intel 3.0c and Abit IC7 won't do 240 fsb...WHY????

[superHARD]

Well I have a Zalman 7000, AS3'ed hsf and nb, tried a Antec 350 and Antec 430, I have GEiL 4200 3-8-4-4 ram and I can not run a 3dMark01. It freezes at same test (first loby). Any one have any ideas why?

And yes I have the newest bios on the mobo.

Any help or mobo settings would be great!!!!!!!!!!!!!!!!!!!!!!

 

[Duvie]

Mayb estupid but do you have stability issues with any other application??? I mean 3dmark is not the toughest test you can put the cpu through....

Remove and reinstall or maybe redownload...maybe corrupt download and loading of app...

Secondly use prime95 or superPI and run 32mb and see if system does fine..


Also how do we know it isn't the vid card or drivers of the vid card...

 

[superHARD]

Bios Voltage:

Cpu Core: 1.61
DDR: 2.84
DDR VTT 1.39
AGP VDDQ: 1.58
3.3: 3.27
5: 4.99
12: 11.85

In hardware doctor after prime 95 for 10 min:

Cpu Core: 1.61
DDR: 2.85
DDR VTT 1.39
AGP VDDQ: 1.6
3.3: 3.26
5: 5.03
12: 11.74

 

[superHARD]

Prime 95 errors off to at 240 but at stock voltage and 230 fsb it works perfect!

 

[butch84]

looks like your chip just wont get there without more voltage, or maybe even at all. I hate to say it, but somebody has to: overclocking isnt garanteed!!

good luck dude

 

[superHARD]

I know but it works at stock voltage up to 235 but at 240 even after upping the voltage to 1.7 it still won't pass prime95 or 3dMark!

 

[Duvie]

The voltages look nice!! how are you getting +2.8v??? I may see 2.82v when set to 2.8v bios but underload it is 2.78v and my 3.3v which it is set from is a solid 3.32 even under load...


May be a chip limit.....The only time I have seen chips where added vcore did not result in any benefits was something else limiting the oc or a damaged cpu. Damaged cpus from heat or xcessive vcore (p4's) that I have seen or experienced seem to be reluctant to take any vcore much above default...Seems to possibly have some of those characteristics....



If it runs 3.0c at default and you are not happy then just sell it and try to do better with another one...

Otherwise 15x235 = 3.53ghz is awesome and just be happy...I think many is this range are startig to se the limitations possibly or our motherboards for power regulating and hence why the prescotts and EE are supposed to be different and why not all boards are being gauranteed to run them...

 

[superHARD]

I also get Prime errors at 235...and yes at any voltage I give to it!

I can't get the ram to post at 5:4 but it will at 3:2...why?

 

[Technonut]

The voltages look nice!! how are you getting +2.8v??? I may see 2.82v when set to 2.8v bios but underload it is 2.78v and my 3.3v which it is set from is a solid 3.32 even under load...

My IC7-G is like this also. Running P95 x 2, the VDIMM never dips below 2.80v, and floats between 2.80-2.83V.

superHARD: Do you have the last two GAT settings disabled, and the rest on Auto?

 

[Duvie]

Find your absolutely 100% stable (prime, superPI, 3damrk, and whatever encoding app) at default vcore first...I assumed you were stable at 240fsb from the other thread but that appears not to be...

Once you find our default stable then list that....I usually only see maybe 8-10fsb from low multiplier chips like my 2.6 and 2.4 but you have a higher 15x muliplier so I would think from default to say 1.65v actual you may only get 7 more fsb or 100mhz more....That is why most now don't seem to go as high with vcore cause the benfits are outweighed by little performance, more heat and more potentially instability....

1.58v to 1.6v may be the best and just quit cause you may only get 2more fsb with even jumping to 1.65v....

 

[superHARD]

In my other thread I never said it worked at 240. The fact is it doesn't pass 3dmard at 240. Or for that matter it won't pass at 235.


My absolutely 100% stable fsb is 232 at defualt voltages. After bumping fsb I still can't hit 235 or 240 fsb. This is why I am confused.

 

[zShowtimez]

Whats there to be confused about? If your chip wont hit higher than 232, and youve eliminated other limiting factors, then its not going any higher. ~3500MHz isnt something to be upset about either. If you must have the best, then I suggest you sell that and buy another and give it another shot.

 

[superHARD]

Originally posted by: zShowtimez
Whats there to be confused about? If your chip wont hit higher than 232, and youve eliminated other limiting factors, then its not going any higher. ~3500MHz isnt something to be upset about either. If you must have the best, then I suggest you sell that and buy another and give it another shot.

Do you understand it when I say it runs at stock voltage then at a certian fsb no matter how much higher past stock voltage it won't work.

Don't trust me read this from Duvie

"Obviously he is unstable but the fact is why...It is clear and common sense that cpu do not hit a wall like that and all of sudden not work no matter the vcore....Even if he was at the limit of the default vcore most p4c chips will run another 5-10fsb at least running it up to as high as 1.65v actual....

In the past my 3dmarks could run on a slightly unstable system but not on an extremely flaky system which his would have to be to act like this..... "

 

[Duvie]

And overall it is very strange...

My system...

12x288=3456 I can run 1.53v actual vcore stable....

12x292=3504 I need 1.58v actual to run 100% stable ( a lot of testing done here and rock solid )

12X295=3540 I need 1.63v actual to run 100% stable

12x296=3552 I need 1.65v actual to run stable....( not that much testing here other then the normal stress testing regiment)

The moral here is from default to 1.65v the highest I will go with the p4c chip only can yield me 8 more fsb or about 100mhz....

The other thing is I could run 3dmark2k1se and UT2003demo at 1.56v actual at 292fsb though not 100% stable....

I could also run 3dmark2k1se and UT2003demo at 3.6ghz w/ 1.65v actual though I don't think 1.7v would have been prime stable at that speed. It also ran superpi 1mb and 2mb fine...


Now if we translate this to superhard then we could say he should be able to 6-7fsb since his multplier is higher with up to 1.65v of actual vcore...The fact only 3fsb more he an't do anything no matter the vcore is strange....

Only way to confirm this is too get another chip and see if it behaves similarly...

 

[Jeff7181]

Use some logic...

RAM is running in spec, so that's not the problem.
No matter how much voltage you give the CPU it's stability doesn't increase, that should tell you the CPU is not the problem.
What's left? The motherboard/chipset.