OCing E8500.. STUCK

Toggle sidebar Toggle sidebar
A

alphatronix

Distinguished
Oct 13, 2008
15
0
18,510
:heink: I have a 750i mbo for starters.. AC7 Cooler... ATM, its sitting at 3.91 ghz.. multi x9.5 FSB 412(around there) P95 stable for 24 hours. Only thing I had to raise was Vcore, to about 1.33(CPU-Z- even though bios states 1.63 :sweat: )

Once I hit 421x9.5 i keep hitting an error on P95. Hardware failure, rounding .5 should be .4, some crap like that between 18 and 40 minutes within the start of the test.. I have raised every volt i have in bios... CPU FBS volt, MeM volt, Nforce volt, Nf200 volt... At one point I raised ALL of them at once way above normal conditions and still it would read error.. I dont know why...



:heink: Very Confused :pfff:
 
B

br3nd064

Distinguished
Aug 20, 2008
807
0
18,990
Woah, do the pencil mod. 1.63 is WAY above the max voltage recommendation. You shouldn't need that much voltage anyway, but start with the volt mod.
 
A

alphatronix

Distinguished
Oct 13, 2008
15
0
18,510
These are my temps and CPUz

temp.JPG

cpu.JPG



Do they look right? I have it stable 24 hours on P95.
 
A

alphatronix

Distinguished
Oct 13, 2008
15
0
18,510
Forgot to mention, under full load, CPU temp goes to 69c... Stable. maybe 70c here and there.. Core temp stays no higher then 32c under full load... IDK if thats good or not. My first OCing in years.
 
A

anartik

Distinguished
Jul 14, 2008
56
0
18,630
I went up in 4 steps... first just bumped to 400 fsb with no other changes required. Then went up to 440, 450 and 460. Each of those required significant CPU voltage increases. You really should not need to change any voltages other than CPU, RAM if required(I used a 5/4 divider so only slightly overclocked) and maybe NB (don't know your board). If Prime is failing its usually from overheating, inadequate voltage on droop or possibly memory issues. Prime small unit will consistantly push mine close to tjunction at 100c and fail 1 core (using lapped chip and lapped Thermalright ultra 120 X). I really don't see prime95 stability as the holy grail of OC'ing anymore. Nothing I do on the computer comes close to stressing the system that much. I can run any benchmark suite, compress large amounts of video, compile or play games w/o issue. I have played COD4 as much as 5 hours straight w/o issue. From your realtemp pic I would suspect that maybe one of the sensors is not working right. It could also be the way your cooler is seated. The core difference on mine is 1C.

One thing I noticed... some people recommend turning off loadline calibration with 45nm CPU's. If you notice your voltage at idle you will see it drop off significantly under load (droop) i.e. prime. What that ends up meaning is you need higher voltage to start with. Once I enabled loadline calibration on my board things were much more stable. My voltage is a little high but is stable at 1.4875. Temps of course vary by room temp. Yesterday the room was down to 70F and the cores were idling in the low 40C's. The room can get up to 80F with all the heat pumped out of the case and then Idle is more like 48-50C.

Bottomline I would say you are being too conservative with CPU voltage.
 
B

br3nd064

Distinguished
Aug 20, 2008
807
0
18,990
1.35 seems way too high for 3.9ghz. Mine's running at 3.8ghz and 1.264v. See how much you can lower it without losing stability. Also, 70c is pretty hot. Are you using the stock cooler?
 
A

alphatronix

Distinguished
Oct 13, 2008
15
0
18,510
stock 3.16 ghz, CoreTemp tells me 100 load 54c and 51c..... Kinda high for stock using AC7? maybe i need good thermal paste. HMMM, dnt knw.
 
aeiouandxyz

aeiouandxyz

Distinguished
Apr 18, 2008
157
0
18,680
your temps are high because the more voltage you put into your cpu, the more heat its going to generate. core 2 cpus have a max thermal limit around 70 C so your temps are fine for now. your temps might also be high (but i dont think its the case) because your hsf is not seated properly over the cpu.
 
G

Guest

Guest
voltage does seem really high... i have a q9550 (worse overclocker... and a quad core)

thats at 3.85 ghz at 1.25 volts... like is 1.35 volts really necessary?

anyways you probably have a dud chip or a bad mobo... more than likely a bad mobo
 
B

br3nd064

Distinguished
Aug 20, 2008
807
0
18,990
I think you're confusing TjMax with load temps... You're saying the cpu temps get to 55-60 under load?

Those temps are so high because you're vcore is way too high. My e8500 OC'd to 3.8ghz @ 1.264v is idling at 36c/32c right now with the same heatsink.
 
B

br3nd064

Distinguished
Aug 20, 2008
807
0
18,990
If you think about it though, 60c under load at stock speeds and a decent cooler is pretty hot. I know mine never got that hot with the same cooler; I think mine maxed around 50c.
 
A

akandy

Distinguished
Apr 11, 2008
27
0
18,530
I FEEL YOUR PAIN!!
Just to let you know what I am doing with my E8500 on an EVGA 790i Ultra SLI.
I have mine at 4.16 GHz tested stable on Prime95 for 14 hours(I stopped it with no errors)

My settings are:

Core Voltage 1.40625 (CPUZ says 1.368 with no load and 1.336V under 100% load)
FSB is set to 1.35V ,Default setting was 1.1V
nforce SPP is set to AUTO witch is 1.4V ,Default was 1.3V
nforce MCP is set to AUTO witch is 1.5V, the same as the Default was

My temps under 100% load stay at about 65C
I picked this speed due to the fact that the voltage is right at the intel max recommended ,the temp is well under the max, and I get a round 1 GHz OC for style points : )

Just FYI I was able to get my E8500 to benchmark a 3DmarkVantage (performance setting)score of over 30,000 at 4.6 GHz on only air cooling!

Good luck with your rig man, hope this helps a little.


 
A

alphatronix

Distinguished
Oct 13, 2008
15
0
18,510
ok, this nis what im going to do. ima take a pic of all my utilities running, then take a pic of them all at 3.6, then take em all at 3.9 then take em all at 4 ghz... Take in account the voltage that I use for every stage will be the the most minimum without P95 giving me an error.. I want to know what you guys think... Maybe its my hsf, maybe its the mobo, idk but somthing doesnt feel right if ya know what im saying.. Give me some time, but hopfully all the pics will give a clear answer, i will also post all my bios pics within the voltage section for advice...


thank you...
 
A

alphatronix

Distinguished
Oct 13, 2008
15
0
18,510
SOMTHING TELLS ME MY HSF IS MESSED UP... Im running on stock defualt options and my cpu core temps on 0 load is 41c and 35c.. big difference in them 2 aND KINDA high for stock.. my AC7 is on the cpu correctly but maybe I need to take it off and put on some good thermal paste and not the stock that it comes with... What do you guys think? if its that, then i wont bother taking pics and all that stuff..
 
A

alphatronix

Distinguished
Oct 13, 2008
15
0
18,510
OK, before I decide to OC again, im not taking any chances. I just order Arctic Silver 5 Thermal Paste. Once that **** is on, im almost certain my temps will go down big time.. Then the only thing i will have left to blame is my chip set.....
 
B

br3nd064

Distinguished
Aug 20, 2008
807
0
18,990
The temperature difference is probably just due to a faulty chip. Lots of these 45nm cpu's report faulty temps. Also, don't expect a huge difference in the pre-applied paste (MX-1) and Arctic Silver 5. You might get a 1-2c drop, but nothing much.
 
murdoc

murdoc

Distinguished
Aug 5, 2008
356
0
18,780
I have similar issues. I'm using CoolMaster HyperTX2 and I have to push my E8500 to v1.33 to get 3.94 ghz. Anyone have any suggestions? I've tried to lower my vcore but then I started getting error on OCCT.

My mobo is an Asus P5Q motherboard.
 
murdoc

murdoc

Distinguished
Aug 5, 2008
356
0
18,780


can you tell me what cooler you are using? Would CoolMaster Hyper-TX2 do the job or do I need something like this?

 
You must log in or register to reply here.

TRENDING THREADS

Latest posts

Moderators online

Share this page

COMPANY
RESOURCES
FOLLOW
Top Bottom