OK...here is the situation....
2.4c@3.5 (292fsb) on IC7 (bios version 14&18)
Swiftech MCX4000 with AS2 ans 92mm Panaflo (58cfm) 3 case fans...
1.65v bios set(1.6v actual)
Bios reports 52c after sitting for several minutes at idle....I go into windows and let it sit and it will settle down to the mid 40's...What is up with that. I am running bios version 18 but noticed similar discrepancies with bios version 14.
So 45-46c idle and then I start 2x prime95...I will only get about 3-4 minutes into it on one and maybe 13-15minutes on the other...rounding error...Load temp will peak with case on and window closed at 66-67c....63-64 with blowing way but same end results of errors...
If I open window and let some of that 40-45f weather come in then I see no greater then 59.5-60c with sucking and now only 56.50 with blowing results????? Last night I went the whole night no prime error, and right now I am agin 3hours into it with no errors...Weird huh???
I used same AS2 but with the consistency that I mentined..."bottom of the barrel" stuff....on the NB chipset.
temps on the PWM are also down with window cracked and case open so it may be that as well. System temp is alos down a few degrees right now loading...
Bios 14 is best for me to get my geils to their maximum cpabilities but prim95 will crash fast. I upgraded to bios 18 and same sped and vcore went 2 hours before I stoppe it. That was at 3.38ghz but vcore was default still and temps loaded with system closed was only 60-60.5c loaded....
Bios 14 problem happens around 250 with this board and I think similar to what I saw with my 2.6c in there though I couldn't definitely state that cause I was clos to the limit of the chip.
Downside is that with the bios 18 I can not run 5:4 with my geil pc3500 even at 280fsb with 2.8v and cas 3,8,4,4 when it previously ran memtested and prime stable at 466ddr for 8 hours each...No matter vcore, vdimm, or vagp will get it to boot....Another oddity
Next thing is the vddq voltage is nice and high (I guess what high is for this board) 1.38-1.39v when have the agp set to 1.6v and the cpu is around 3.4ghz or less....When I bump it to 3.5ghz it runs 1.3-1.33 range...I then up the vagp and it goes back to 1.39 range...You know anything about this reading???
Voltage rails are still tight to good and show no difference in testig above so I have ruled them out excpet for the VDDQ oddity. Naturally the Abit Ic7 undervolts a lot of things on this board and that is the one case where the bios and the windows environment report the same thing.
vcore = 1.65 bios = 1.60idle 1.58-1.57 load
vdimm=2.8v bios = 2.78-2.8 idle same as load
vagp=165v bios = 1.63idle 1.62v load
vagp=1.6v bios 1.58v idle and 1.57 load
+3.3v = 3.32idle 3.28load
+5v= 5.08idle 5.04load
+12v=11.94vidle 11.8load
Theories:
1) I have a cpu or chipset heating issue that I need to resolve. I think my AS2 is bad as it jas run out and the consistency of the stuff in the bottom of the tube looked either runny or a bit more tacky then normal AS2 I have epxerienced over the last 2 years...Opps maybe that is a bit long for the shelf life!!!
I will be apply new AS5 as soon as it arrives.
I know the hsf is applied right as I got same results with alpha and swiftech on my 2.6c....again they had high cpu and pwm temps.
2) bios issue of reported temp in bios and monitoring program. I amy actually be running higher then reported by 3 different programs. results is when I see 63-64 I could likely be in the 70's and hence the sumout error and rounding errors...
2.4c@3.5 (292fsb) on IC7 (bios version 14&18)
Swiftech MCX4000 with AS2 ans 92mm Panaflo (58cfm) 3 case fans...
1.65v bios set(1.6v actual)
Bios reports 52c after sitting for several minutes at idle....I go into windows and let it sit and it will settle down to the mid 40's...What is up with that. I am running bios version 18 but noticed similar discrepancies with bios version 14.
So 45-46c idle and then I start 2x prime95...I will only get about 3-4 minutes into it on one and maybe 13-15minutes on the other...rounding error...Load temp will peak with case on and window closed at 66-67c....63-64 with blowing way but same end results of errors...
If I open window and let some of that 40-45f weather come in then I see no greater then 59.5-60c with sucking and now only 56.50 with blowing results????? Last night I went the whole night no prime error, and right now I am agin 3hours into it with no errors...Weird huh???
I used same AS2 but with the consistency that I mentined..."bottom of the barrel" stuff....on the NB chipset.
temps on the PWM are also down with window cracked and case open so it may be that as well. System temp is alos down a few degrees right now loading...
Bios 14 is best for me to get my geils to their maximum cpabilities but prim95 will crash fast. I upgraded to bios 18 and same sped and vcore went 2 hours before I stoppe it. That was at 3.38ghz but vcore was default still and temps loaded with system closed was only 60-60.5c loaded....
Bios 14 problem happens around 250 with this board and I think similar to what I saw with my 2.6c in there though I couldn't definitely state that cause I was clos to the limit of the chip.
Downside is that with the bios 18 I can not run 5:4 with my geil pc3500 even at 280fsb with 2.8v and cas 3,8,4,4 when it previously ran memtested and prime stable at 466ddr for 8 hours each...No matter vcore, vdimm, or vagp will get it to boot....Another oddity
Next thing is the vddq voltage is nice and high (I guess what high is for this board) 1.38-1.39v when have the agp set to 1.6v and the cpu is around 3.4ghz or less....When I bump it to 3.5ghz it runs 1.3-1.33 range...I then up the vagp and it goes back to 1.39 range...You know anything about this reading???
Voltage rails are still tight to good and show no difference in testig above so I have ruled them out excpet for the VDDQ oddity. Naturally the Abit Ic7 undervolts a lot of things on this board and that is the one case where the bios and the windows environment report the same thing.
vcore = 1.65 bios = 1.60idle 1.58-1.57 load
vdimm=2.8v bios = 2.78-2.8 idle same as load
vagp=165v bios = 1.63idle 1.62v load
vagp=1.6v bios 1.58v idle and 1.57 load
+3.3v = 3.32idle 3.28load
+5v= 5.08idle 5.04load
+12v=11.94vidle 11.8load
Theories:
1) I have a cpu or chipset heating issue that I need to resolve. I think my AS2 is bad as it jas run out and the consistency of the stuff in the bottom of the tube looked either runny or a bit more tacky then normal AS2 I have epxerienced over the last 2 years...Opps maybe that is a bit long for the shelf life!!!
I will be apply new AS5 as soon as it arrives.
I know the hsf is applied right as I got same results with alpha and swiftech on my 2.6c....again they had high cpu and pwm temps.
2) bios issue of reported temp in bios and monitoring program. I amy actually be running higher then reported by 3 different programs. results is when I see 63-64 I could likely be in the 70's and hence the sumout error and rounding errors...