that seems like a really high pwm temperature. how's your case airflow? have you checked your heatsinks on your motherboard to make sure that they're making good contact?
Originally posted by: Peav
Originally posted by: Muadib
You might be on to something with the pushpins. I have the Xigmatek HDT-S1283, and it uses those plastic pushpins. I might order that bolt on kit, but if you say it's normal for quads, maybe I won't bother. I really don't want to rip everything out again.
My temps under load topped out at 60C with a room temp of 75F. Core temp says my VID is 1.2125v. I have the vcore set to 1.2725 in the bios. I also have my memory voltage set to 1.90V, with the timings set really low for now.
Another concern I had was the pwm temp. Speedfan had it at 70C under load. Is that normal for the IP35-E?
My PWM goes up to around 70C under load. My Overclock is the same as yours but my Vcore is 1.35V VID is 1.3.
I have noticed the cores on my quad have different temps under load. Cores 0 and 1 around 63C at load but cores 2 and 3 are usually around 3-5C lower. I think its normal and your temps all seem within acceptable limits.
I use a Xp120 with a YS-Tech 120MM 5 Blade fan @ 7v on it.
I have no control over the room temp. I stay with my gf, and she pays the heat bill. 75F is what she likes. It drives me crazy, especially at night. Thankfully, summer will be here soon. She keeps it pretty cool then, a nice 70F. I know, I don't get it either.Originally posted by: MadScientist
That's a great VID for a Q6600 and according to Frosty Tech's review the Xigmatek HDT-S1283 is one hell of a heatsink. So why are your CPU and pwm temps high? Are you running your HS's fan at max, 2000RPM? Your room temp's a bit toasty at 75F. Poor case air flow? At 1.1275 Vcore in bios on full load my Q6600 is running in the low 50s and my pwm temp is in the 50s. My room temps are cooler than yours, but even on medium fan settings my Antec 900 case moves a lot of air. With your VID, HS, a slightly lower room temp, and good case air flow your Q6600 should have no problem doing 3.6 Ghz at full load temps in the 60's.Originally posted by: Muadib
You might be on to something with the pushpins. I have the Xigmatek HDT-S1283, and it uses those plastic pushpins. I might order that bolt on kit, but if you say it's normal for quads, maybe I won't bother. I really don't want to rip everything out again.Originally posted by: MadScientist
It's not uncommon to have one or more cores running at different temps on a Q6600. I have one core that's 7C lower than the other three. It could be that your IHS is not flat, lapping?; could be your HS mounting, try remounting or if you have those POS pushpins replace them with a Thermalright Bolt-Thru Kit; or as in my case it's just the way the IHS sits on the cores and there's nothing you can do about it.Originally posted by: Muadib
I have a problem with the temps on my Q6600. The first two cores are 5c hotter than the other two. Any idea what would cause this? I'm using both speedfan & core temp for the measurement.
I'm currently running at 3060 MHz, 340 x 9, with the memory at 1:1 5-5-5-15. I fail prime95 if I clock it any higher, and I'm afraid to add anymore voltage to the cpu, until I can figure out why I have such a variance in the core temps. Any ideas?
What's the VID, per Core Temp, of your Q6600? Mine is 1.2375V. What's your load temps, Small FFTs, at 3.06 Ghz?
My Q6600 needs 1.2975 bios, 1.248V idle, 1.208V load (CPUZ) @ 3.40Ghz, 12 hrs. Prime95 (small FFts) stable, 32C idle, 52C load with a Coolermaster Hyper TX2 HSF and an Antec 900 case, room temp 66-68F (winter).
It will do 3.6Ghz at 1.3575V bios, 1.312V idle, 1.272V load (CPUZ), 34C idle, 57C load stable, but for now I'm running it at 3.4Ghz.
Q6600 GO stepping @3.4 Ghz (9 x 378), 1.2975V bios
Gigabyte Geforce 8800GT GV-NX88T512HP
4 x 1GB Crucial Ballistix PC6400 @946Mhz, 1:1.25 ratio, 5-4-4-12-2T, 2.1V, MCH 1.25V
Vista Ultimate 64, SP1
2 x Maxtor 300GB SATA HDs
Samsung SH-S203B SATA DVD Burner
Corsair 620W PSU
Cooler Master Hyper TX2 HS/fan
Antec 900 case
My temps under load topped out at 60C with a room temp of 75F. Core temp says my VID is 1.2125v. I have the vcore set to 1.2725 in the bios. I also have my memory voltage set to 1.90V, with the timings set really low for now.
Another concern I had was the pwm temp. Speedfan had it at 70C under load. Is that normal for the IP35-E?
No, your overclock it better than mine. I get errors in Prime95 at 9x390. Your Vcore is higher too. I want to try a Vcore of 1.35, but I'm afraid of raising my PWM even higher. Your PWM temp doesn't concern you?Originally posted by: Peav
My PWM goes up to around 70C under load. My Overclock is the same as yours but my Vcore is 1.35V VID is 1.3.
I have noticed the cores on my quad have different temps under load. Cores 0 and 1 around 63C at load but cores 2 and 3 are usually around 3-5C lower. I think its normal and your temps all seem within acceptable limits.
I use a Xp120 with a YS-Tech 120MM 5 Blade fan @ 7v on it.
Which heatsink is for the PWM? I would try getting a little fan for it, but I don't know where to put it.Originally posted by: bryanW1995
that seems like a really high pwm temperature. how's your case airflow? have you checked your heatsinks on your motherboard to make sure that they're making good contact?
Everyone has different OC methods. Re-read the OC sticky I like to take the isolate and then consolidate approach to overclocking. You can either start with your ram or CPU. I usually start with my CPU. In the bios, for now, disable C1E and EIST. You can always enable them later and see if your system is stable.Originally posted by: METALWORK
I have yet to attempt OC with my system and I would like to know if Vcore or fsb should be pushed first? I read the oc tips sticky and there the fsb is raised first, vs. in SR's original review under oc tips the vcore for a quad was first raised to 1.4v and then incremental increases made in fsb.
Do you have the Fan EQ Control enabled in the bios?Originally posted by: Muadib
Back on topic, I have the fan on my HS set to run at max, but it won't go faster than 1360 RPM. Maybe I should look into replacing it. I think my cpu temps are on par with yours though, if you factor in the room temp difference.
No, I turned it off, because I only have the cpu fan connected to the board, and I wanted it to run at 100%. Should I have it on?Originally posted by: MadScientist
Do you have the Fan EQ Control enabled in the bios?Originally posted by: Muadib
Back on topic, I have the fan on my HS set to run at max, but it won't go faster than 1360 RPM. Maybe I should look into replacing it. I think my cpu temps are on par with yours though, if you factor in the room temp difference.
Acccording to Frosty Tech's review: "The Xigmatek HDT-S1283 120mm PWM fan spins at 1000-2200RPM and draws power from a standard 4-pin PWM motherboard fan header. It is backwards compatible with 3-pin connectors." Do you have it connected to the 4 pin CPUFan1 header or the SYSFan1 3 pin header on the MB? Connected to either header with Fan EQ Control disabled your fan should be running at 100%, close to 2200RPM. Where are you reading the fan RPMs from, SpeedFan? Are the fans set to 100% in SpeedFan?Originally posted by: Muadib
No, I turned it off, because I only have the cpu fan connected to the board, and I wanted it to run at 100%. Should I have it on?Originally posted by: MadScientist
Do you have the Fan EQ Control enabled in the bios?Originally posted by: Muadib
Back on topic, I have the fan on my HS set to run at max, but it won't go faster than 1360 RPM. Maybe I should look into replacing it. I think my cpu temps are on par with yours though, if you factor in the room temp difference.
I have it connected to the 4 pin CPU Fan1 header. It reads 1361RPM in both the bios, & speedfan. I have all fans set to 100% in speedfan.Originally posted by: MadScientist
Acccording to Frosty Tech's review: "The Xigmatek HDT-S1283 120mm PWM fan spins at 1000-2200RPM and draws power from a standard 4-pin PWM motherboard fan header. It is backwards compatible with 3-pin connectors." Do you have it connected to the 4 pin CPUFan1 header or the SYSFan1 3 pin header on the MB? Connected to either header with Fan EQ Control disabled your fan should be running at 100%, close to 2200RPM. Where are you reading the fan RPMs from, SpeedFan? Are the fans set to 100% in SpeedFan?Originally posted by: Muadib
No, I turned it off, because I only have the cpu fan connected to the board, and I wanted it to run at 100%. Should I have it on?Originally posted by: MadScientist
Do you have the Fan EQ Control enabled in the bios?Originally posted by: Muadib
Back on topic, I have the fan on my HS set to run at max, but it won't go faster than 1360 RPM. Maybe I should look into replacing it. I think my cpu temps are on par with yours though, if you factor in the room temp difference.
My bad for trusting Frosty Tech. I just checked Xigmatek's website and the specs for the fan are 800-1500RPM, Link. Fan specs are only appproximations, so at 1361RPM your fan is probably running at 100%.Originally posted by: Muadib
I have it connected to the 4 pin CPU Fan1 header. It reads 1361RPM in both the bios, & speedfan. I have all fans set to 100% in speedfan.Originally posted by: MadScientist
Acccording to Frosty Tech's review: "The Xigmatek HDT-S1283 120mm PWM fan spins at 1000-2200RPM and draws power from a standard 4-pin PWM motherboard fan header. It is backwards compatible with 3-pin connectors." Do you have it connected to the 4 pin CPUFan1 header or the SYSFan1 3 pin header on the MB? Connected to either header with Fan EQ Control disabled your fan should be running at 100%, close to 2200RPM. Where are you reading the fan RPMs from, SpeedFan? Are the fans set to 100% in SpeedFan?Originally posted by: Muadib
No, I turned it off, because I only have the cpu fan connected to the board, and I wanted it to run at 100%. Should I have it on?Originally posted by: MadScientist
Do you have the Fan EQ Control enabled in the bios?Originally posted by: Muadib
Back on topic, I have the fan on my HS set to run at max, but it won't go faster than 1360 RPM. Maybe I should look into replacing it. I think my cpu temps are on par with yours though, if you factor in the room temp difference.
I have my case closed right now. Is that the blue heatsink?Originally posted by: MadScientist
My bad for trusting Frosty Tech. I just checked Xigmatek's website and the specs for the fan are 800-1500RPM, Link. Fan specs are only appproximations, so at 1361RPM your fan is probably running at 100%.Originally posted by: Muadib
I have it connected to the 4 pin CPU Fan1 header. It reads 1361RPM in both the bios, & speedfan. I have all fans set to 100% in speedfan.Originally posted by: MadScientist
Acccording to Frosty Tech's review: "The Xigmatek HDT-S1283 120mm PWM fan spins at 1000-2200RPM and draws power from a standard 4-pin PWM motherboard fan header. It is backwards compatible with 3-pin connectors." Do you have it connected to the 4 pin CPUFan1 header or the SYSFan1 3 pin header on the MB? Connected to either header with Fan EQ Control disabled your fan should be running at 100%, close to 2200RPM. Where are you reading the fan RPMs from, SpeedFan? Are the fans set to 100% in SpeedFan?Originally posted by: Muadib
No, I turned it off, because I only have the cpu fan connected to the board, and I wanted it to run at 100%. Should I have it on?Originally posted by: MadScientist
Do you have the Fan EQ Control enabled in the bios?Originally posted by: Muadib
Back on topic, I have the fan on my HS set to run at max, but it won't go faster than 1360 RPM. Maybe I should look into replacing it. I think my cpu temps are on par with yours though, if you factor in the room temp difference.
Not enough airflow getting through the Xigmatek HS to cool the heatsink on the MB just left of the CPU could be the reason why your pwm temp is high.
Yeah, the blue HS right below the 8pin ATX12V1 connector.Originally posted by: Muadib
I have my case closed right now. Is that the blue heatsink?
Not if you have an A: drive. I guess you don't though. Since you are having problems with the link you posted, go here, and use method #2. You need a pc running XP though.Originally posted by: owbert
a little help with updating the bios? i am lost.
how did everyone else update their bios? this way: http://www.abit.com.tw/page/uk/download/guide.php#7 or this way: http://www.xtremesystems.org/f...2397392&postcount=2023
i have the pro version and e version and would like to update both of them to version 16.
would i be unable to flash using abit's guide with a thumbdrive? if i were to use ace's way i am lost as to why i have to edit the .bat files.
is it this hard to update a bios
Thanks! I'll see if I can find a fan for it sometime this week.Originally posted by: MadScientist
Yeah, the blue HS right below the 8pin ATX12V1 connector.Originally posted by: Muadib
I have my case closed right now. Is that the blue heatsink?
If you have big kahunas you can flash your bios in Windows using Abit's Flash Menu, be sure to follow Master of None's caveats on Abit's forum. Link. I've done it sucessfully a number of times. Also exit out of all running applications, i.e., AV software, before attempting.Originally posted by: owbert
a little help with updating the bios? i am lost.
how did everyone else update their bios? this way: http://www.abit.com.tw/page/uk/download/guide.php#7 or this way: http://www.xtremesystems.org/f...2397392&postcount=2023
i have the pro version and e version and would like to update both of them to version 16.
would i be unable to flash using abit's guide with a thumbdrive? if i were to use ace's way i am lost as to why i have to edit the .bat files.
is it this hard to update a bios
Originally posted by: MadScientist
Everyone has different OC methods. Re-read the OC sticky I like to take the isolate and then consolidate approach to overclocking. You can either start with your ram or CPU. I usually start with my CPU. In the bios, for now, disable C1E and EIST. You can always enable them later and see if your system is stable.Originally posted by: METALWORK
I have yet to attempt OC with my system and I would like to know if Vcore or fsb should be pushed first? I read the oc tips sticky and there the fsb is raised first, vs. in SR's original review under oc tips the vcore for a quad was first raised to 1.4v and then incremental increases made in fsb.
1. Set your ram at loose timings, 5-5-5-15 or 18-2T, 1.8V, or whatever SPD in the bios sets it at, and at a 1:1 divider. Set the CPU to it's default setting and vcore. For a Q6600 GO with this board it's 272 FSB x 9 = 2.448 Ghz at and your VID . Run Memtest86+ to make sure your ram is stable (no errors) at these settings. If Memtest86+ errors raise the vdimm. This takes your ram out of the equation.
2. Keep the ram at 1:1 divider and start raising the FSB in 5 to 10 increments. Run Prime95 V25.6 Small FFts at each setting until it errors. This will usually happen within the first 15 minutes. Small FFts puts maximum stress on your CPU and very little on the ram. Raise the vcore and test again until it passes. Monitor your CPU temps with CoreTemp. Repeat until you reach an overclock and temps you are comfortable with. Run Prime95 for 6-8 hours to make sure it's stable. If not stable raise the voltage and repeat.
3. To find out what my ram is capable of I set my CPU to a 266 FSB and a 6 multiplier. This takes my CPU out of the equation. I start with my ram's rated timings, in your case, 5-5-5-15-2T, 1.8V I'm assuming. Start raising the FSB by 5 to 10 increments and run Memtest86+ test 5 until it errors. You may have to increase the ram divider or CPU multiplier. Try to keep the CPU at or below stock setting. Increase the vdimm and repeat. With your ram I would not go any higher than 2.0 vdimm. This gives you a ram profile at these settings. Loosen the timings up to 4-4-4-12-2T and repeat. When you find ram settings you are comfortable with run Memtest86+ test 5 for at least 1 hour and all tests for at least 10 passes. With 4 sticks of ram you may have to bump your MCH voltage up a notch or two.
There is a debate on whether higher mem freqs, tighter timings, or whole integer dividers, i.e, 1:1, 1:2 are better for real life applications.
4. Now it's time to consolidate. Set your CPU and ram to known stable settings. Run Prime95 V25.6 Blend setting for at least 8 hours. The Blend setting will stress both your CPU and ram. You may find that some ram settings that were stable in Memtest86+ fail Prime95 blend and you will have to adjust your ram settings. MemTest86+ is better than Prime95 at finding bad RAM but it isn't as good as Prime95 at stressing the CPU/RAM interface. Monitor your idle and load temps with CoreTemp.
Originally posted by: METALWORK
Originally posted by: MadScientist
Everyone has different OC methods. Re-read the OC sticky I like to take the isolate and then consolidate approach to overclocking. You can either start with your ram or CPU. I usually start with my CPU. In the bios, for now, disable C1E and EIST. You can always enable them later and see if your system is stable.Originally posted by: METALWORK
I have yet to attempt OC with my system and I would like to know if Vcore or fsb should be pushed first? I read the oc tips sticky and there the fsb is raised first, vs. in SR's original review under oc tips the vcore for a quad was first raised to 1.4v and then incremental increases made in fsb.
1. Set your ram at loose timings, 5-5-5-15 or 18-2T, 1.8V, or whatever SPD in the bios sets it at, and at a 1:1 divider. Set the CPU to it's default setting and vcore. For a Q6600 GO with this board it's 272 FSB x 9 = 2.448 Ghz at and your VID . Run Memtest86+ to make sure your ram is stable (no errors) at these settings. If Memtest86+ errors raise the vdimm. This takes your ram out of the equation.
2. Keep the ram at 1:1 divider and start raising the FSB in 5 to 10 increments. Run Prime95 V25.6 Small FFts at each setting until it errors. This will usually happen within the first 15 minutes. Small FFts puts maximum stress on your CPU and very little on the ram. Raise the vcore and test again until it passes. Monitor your CPU temps with CoreTemp. Repeat until you reach an overclock and temps you are comfortable with. Run Prime95 for 6-8 hours to make sure it's stable. If not stable raise the voltage and repeat.
3. To find out what my ram is capable of I set my CPU to a 266 FSB and a 6 multiplier. This takes my CPU out of the equation. I start with my ram's rated timings, in your case, 5-5-5-15-2T, 1.8V I'm assuming. Start raising the FSB by 5 to 10 increments and run Memtest86+ test 5 until it errors. You may have to increase the ram divider or CPU multiplier. Try to keep the CPU at or below stock setting. Increase the vdimm and repeat. With your ram I would not go any higher than 2.0 vdimm. This gives you a ram profile at these settings. Loosen the timings up to 4-4-4-12-2T and repeat. When you find ram settings you are comfortable with run Memtest86+ test 5 for at least 1 hour and all tests for at least 10 passes. With 4 sticks of ram you may have to bump your MCH voltage up a notch or two.
There is a debate on whether higher mem freqs, tighter timings, or whole integer dividers, i.e, 1:1, 1:2 are better for real life applications.
4. Now it's time to consolidate. Set your CPU and ram to known stable settings. Run Prime95 V25.6 Blend setting for at least 8 hours. The Blend setting will stress both your CPU and ram. You may find that some ram settings that were stable in Memtest86+ fail Prime95 blend and you will have to adjust your ram settings. MemTest86+ is better than Prime95 at finding bad RAM but it isn't as good as Prime95 at stressing the CPU/RAM interface. Monitor your idle and load temps with CoreTemp.
Thanks for the info, very helpful, especially running FFts to check stability along the way.
I've decided that I need another fan for the NB before moving along but I was stable at 350 x 9 with 1.345vcore. On another note yesterday I installed SP1 and now it seems that
EIST is not functioning. I am back at default settings for now and my mult. doesn't step down to 6 as it did the last time I checked. Could this be an issue with SP1 and ver 15 bios? I noticed that something was installed for ICH9, a message about such popped up.
It also seems that my idle temps are higher now, though I'm not sure, not too bad at 36-45C. I haven't messed with windows power settings, at default as always. Just something to be annoyed about I guess. Thanks again, digging my new build.
http://forum.abit-usa.com/showthread.php?t=124207Originally posted by: owbert
i have the pro version and e version and would like to update both of them to version 16.
would i be unable to flash using abit's guide with a thumbdrive?
We need more info. From your other thread you are running OCCT, CPU or Mix setting under load? What CPU HS/F and case are you using? 50C under load is is pretty good for 1.3V. My Q6600 runs in the low 50's under load at 1.2975V bios by CoreTemp. Try running Prime95 V25.6 on Small FFts setting and report back what load temps CoreTemp reports.Originally posted by: daggermeerkat
My Abit pro shows very different temperatures in coretemp and uGuru. Which one do I go with? It shows 50 under load in coretemp and 63 in uGuru? Any ideas? This is @ 3.15Ghz @ stock voltage of 1.3V