Great chip there shogo :) Better than mine. I'm fine with 82c loaded because I'm almost positive it's wrong. There's no way a CPU can run fully loaded at 82C for 2 and a half hours when the heatpipes on my typhoon are not even warm.
Great chip there shogo :) Better than mine. I'm fine with 82c loaded because I'm almost positive it's wrong. There's no way a CPU can run fully loaded at 82C for 2 and a half hours when the heatpipes on my typhoon are not even warm.
hehe, that would leave me with a very queasy feeling in my gut everytime i started my computer:DQuote:
Originally Posted by afireinside
edit, very immature, but just noticed this also marks my 69th post :p:
Double post here :P
But thank you. Yes its a great chip. I think coretemp is still right but because it takes temp directly from the on die diode, the real temp that we are used to see is lower than that, thats maybe why the heatpipes arent really hot. My previous 4400 X2 was loading at 55c and the heatpipes were really at 55c, pretty warm to the touch. but maybe the temp on die was something like 85c.... who knows..
But then again 4ghz is cool but not cool if i cant 3D bench with it. I would like to break 11k in 3dmark 06 with my GX2. Im at 10554... 450 mark needed, i dont think 200mhz more can give me 500 mark but ill be closer. More fvsb is needed maybe but i dont know where to stop.... and also Supertim0r is harassing me to put 2.6vdim for some 3-3-3-8 action i feel like im gonna kill something today. OS corrupted at 4140mhz :(
I ran my 5300 ballistix at 2.8v for maybe 10 hours and they're fine, you could give 2.6 a shot without worrying.
My real temp is about 55c when gaming since orthos is dual load AND orthos pushes temps about 10c higher than dual 32m.
hey AFI, what vcore are you giving to your chip @ 3.75 ?
1.5v ;)
3.75 max ?
Haven't tested higher, maybe 3800 stable with 1.5 :) I'll try more over night.
Upped vfsb to 1.5, vmch still to 1.7, vcore to 1.6, all air, Now its Aquamark Benchable
http://www.oliviercouture.com/PC/Benches/AM3_4ghz.JPG
Pretty stable, no random reboot, but idles at 50c so it may load over 85 so thermal throttling kicked in during 3dmark cpu test1 ... Maybe time for balcony benching..
Now went back to my sweetspot (3800mhz, and trying to make 3860mhz 24/7 stable)
http://www.oliviercouture.com/PC/Ben...in_3860mhz.JPG
Im now at 2h 20min and still loading. Could be enough to call it stable but maybe not 8h stable...
After that it would be the time to play with memory. im keeping 1:1 for now but maybe 4:5 (1070mhz) is doable with 4-4-4-10 2.35v for everyday use.....
Thanks for the info on the revision changes.
For anyone who needs the latest everest grab it here.
http://www.lavalys.com/beta/everestu...pwpo3mzmtx.zip
I think I corrupted my OS on stock. Woo woo for me.
Anyone have a list of their "ideal" safe voltages for any increases?
Why everybody want to mess with everest? Coretemp and IDU are doing the same thing and doesnt mess your board eheh.Quote:
Originally Posted by Teratism
For "ideal" voltage it always depends on your temps, there is no maximum voltage, only maximum temperature... A rule of thumb would be not to go higher than 70-75c coretemp during orthos if you want a good overclock if you are mostly just gaming and browsing.... If you do f@h and other stuff that loads both core 100% for a long period of time, maybe 65c max in coretemp is more appropriate.
It also depends on your cpu stepping. My L631B is really more hot than other Cpu for same voltage but can do more with same voltage too... Thats why i wont tell you : max vcore is 1.45v .... Max vcore would be the max it can while having good temps 100% load...
Even with good temps won't having a high voltage going through it be able to damage the circuitry?
I guess people want to mess with everest to get a "second opinion" and it is easier if it is all in a single package.
My board is stable @ 3.4ghz @ 1.47x vcore @ 377fsb w/ load temps maxing @ 57c.
Ill try some other timing, but I think my ram is holding me back.
People with Phase that runs 24/7 can run 1.65 - 1.7vcore... I dont know maybe youre right about vcore is bad for the pcb or something but how bad could it be?Quote:
Originally Posted by Teratism
I keep the faith in temps...
If anyone buys one of the soon to be released Q6600s and drops it in this board, post up some results!
Temp does have a lot to do with it, but you also have to worry about the current that you are causing as well. You may be able to dissipate all the heat in the world if you have phase but your transistors are also taking care all the energy that the current is creating by switching on and off. This is the point where you get in to the realm where maybey you won't fry your chip by melting connections, but you are fusing transistors through current load. It takes a bit longer to notice but it's killing things as well.Quote:
Originally Posted by shogo_ca
what am i blabbering about in my half drunk state? basically anyone who over-volts their chip shortens the lifetime. uh.. so if you want to hold on to your CPU for 2-3 years be careful of ending up with anything that puts the temp up to the 60's and 70's. But really, who here plans on playing with one chip for 2-3 years:p:
max fsb is 390ish with quad coreQuote:
Originally Posted by malficar
3.5ghz max with Q6600...with actual bios (this could change with a newer bios)
anyway, 3.4-3.5ghz air a lot on air ;)
So we see the P5B DX to XBX2, now how about Commando - XBX2 if you were running a X3060 with highend DDR2? and only wanna run about 400x9 DDR1000-1100 what board would do it better, and be stable?
I'm having a hard time deiciding, ive abandoned hope of the RD600 so now its come down between the new 965 board, or a 975X board :rolleyes:
Could someone summarise 2333 vs 2395? Timings are better but are people hitting fsb walls quicker?
2395 would randomly not post and force me to boot in configuration mode and set everything back, even at 100% stable settings. 2333 has been flawless. 440 on 1066 strap. I'll try for 450 again with 1.5vfsb tomorrow instead of 1.425 but I think I need a fan directly over chipset HSF and vmch mod.
I suggest a 50mm fan over the NB heatsink with some long screws. Worked great for me.
I'll change today and see if I can go any higher than 3.9 on my QX6700 :D
Quote:
Originally Posted by Omastar
Hey Omastar have you used a P5B DX before? If so how does it compare to XBX2, I need a new board and right now its between XBX2 and Commando.
Thanks for the feedback.Quote:
Originally Posted by Supertim0r
I am reading through the thread but can't seem to really find any information on those 2 letter codes that appear on the monitor during boot up. I saw that it may be a reference to bios errors but how do we find out or where to look to check which ones? I also noticed no one answered palase about the unknown PCI device.
On a side issue I am having a BS time with getting my board stable even at stock. I can't tell if it is a board, biod, ram, OS, or what issue. I get Blue screens of death after a few minutes of anything intensive, be it gaming, rars, stress testing etc. All programs report my temps at mid to high 30's and low to mid 40's the air coming out of the case isn't hot. The BSODs have come from ntkrpamp.exe, klif.sys, tcpip.sys, memory corruption, one from bit defender, one from zone alarm, 2 from 2 different ati dlls. I can't remember the rest right now but anyone have an opinion on this?