Switch to Linear ModeSwitch to Hybrid ModeSwitch to Threaded Mode
Printer Friendly View | Email this page | Register Now to start posting!
keat Registered User


Join Date: Jan 2006
Posts: 37
Trade rep: 0 (0%)
Infractions: 0/0 (0)
High Intel DG965RY board temperature! keat Jun 5th, 07, 08:48 PM #1

I using this intel desktop utilities and check my computer's temperature.
cpu temperature: 43c
Motherboard temperature: 48c & 50c (why got 2 i also not sure)
I/O controller hub (ICH) temperature: 79C! (WHY so High!)
Memory controller hub (MCH): 56C
chassis Inlet fan: 1016rpm
chassis outlet fan: 700rpm

I dont know why the temperature of the ICH on the motherboard is so high sia
guys if u are using this board or 965 chipset, may i check if the temperature is also so high?

Thanks


sg.png
Chirag Registered User


Join Date: Apr 2007
Location: India
Posts: 10
Trade rep: 0 (0%)
Infractions: 0/0 (0)
Chirag Jun 8th, 07, 03:20 PM #2
Well yes, the tempearture is pretty normal for I/O Channel. I guess it is because of the amount of Data that constantly needs to be handled in this channel. Since all the processes basically depends upon Inputs and Outputs, we can expect this channel to remain busy. Well i wouldn't say i'm certain about it, but i find it to be the reason...don't really think there could be any other reason. ICH is meant for that purpose right?

So, relax...and...keep it "COOL"
in.png
keat Registered User


Join Date: Jan 2006
Posts: 37
Trade rep: 0 (0%)
Infractions: 0/0 (0)
keat Jun 10th, 07, 07:33 PM #3
oic, hopefully it is normal, anyway, i tried to adjust the internal cables so that they dont block the airflow on the ICH, the temperature managed to drop to 76C while other chips maintain @ the same temperature..
sg.png
C.Jackson
C.Jackson's Avatar
Intel's weakest link


Join Date: Dec 2006
Location: Florida
Posts: 188
Trade rep: 0 (0%)
Infractions: 0/0 (0)
C.Jackson Jun 11th, 07, 04:05 AM #4
Mine used to run as high as 100C under load before I got it fixed (bios issues) so it can get pretty warm before causing any damage. As a matter of a fact it now runs at around 42C in a 18C room, the rear facing the A/C and less than a foot of room between them.
The G965 boards have thermal issues that have yet (From what I've seen) to be resolved by Intel; the bios do not allow fan oscillation to increase with heat output. Although there are no official fixes, QST is the main cause of this. Quiet System Technology, is the limiting factor on sound produced by your fans but what Intel did not take into consideration is that your motherboard, processor, ect. need sufficient cooling to remain functioning (Hell of a thing to forget). The only fix is to disable QST and this will render your fans running at 100% at all times. This way you'll go thru alot of fans over time and you'll have the sound of a 747 in your computer room but at least you won't worry about your system overheating.
I do not suggest you do this but if you must (Have constant/fear of overheating) you can download Intel's developer bios from my previous thread, make sure these are compatible with your board and if they aren't do not attempt to use them. Please follow all of the instructions in the thread because this is a very touchy subject. I really do wish Intel would get their act together because an overclocker/gamer's worst nightmare is a hot computer.
us.png
Scooter Guest


Posts: n/a
Trade rep: (%)
Infractions: / ()
Who told you this was too hot? Scooter Jul 8th, 07, 03:44 AM #5
According to the info on Intel's web site, the temperature of the ICH is being measured right on the die, not in some external sensor. Thus, I don't see anything necessarily being wrong with temperatures in the 70's or even 80's. Call me calavier, but I trust Intel to understand and manage their own devices' temperatures correctly...
us.png
C.Jackson
C.Jackson's Avatar
Intel's weakest link


Join Date: Dec 2006
Location: Florida
Posts: 188
Trade rep: 0 (0%)
Infractions: 0/0 (0)
C.Jackson Jul 11th, 07, 07:24 AM #6
Quote:
Originally Posted by Scooter
According to the info on Intel's web site, the temperature of the ICH is being measured right on the die, not in some external sensor. Thus, I don't see anything necessarily being wrong with temperatures in the 70's or even 80's. Call me calavier, but I trust Intel to understand and manage their own devices' temperatures correctly...
Actually to be honest if you'd look into it a bit more you'd find that almost all G965 boards have cooling problems. Constant reboots and shut downs because of overheating are very common due to Intel QST (I explained some in my previous post). A system, I agree, should be fairly quiet so the user can use it comfortably but QST is compensating way too much for noise that it forgets those fans are actually cooling the computer. Intel will probably never fix the problem because the boards aren't really new and if you look there have been numerous complaints on the Intel forums so short of looking into a better case with better airflow, active cooling, better thermal paste or some other means your pretty much stuck with a hot board.
On another note, I just saved a bunch of money on my car insurance by switching to Geico .
us.png
matt12 Guest


Posts: n/a
Trade rep: (%)
Infractions: / ()
matt12 Jul 16th, 07, 12:05 AM #7
I have the DG965RY and experience the same issue that the cpu cooling fan does not increase in speed as the processor increases in temperature; the cpu fan is always at minimum rpm no matter what the temperature is - cpu temp. has reached a max of 60C. I thought there was something wrong with the motherboard. Are you saying it is normal behavior for the fan to stay at minimum rpm no matter what the temperature is?
us.png
C.Jackson
C.Jackson's Avatar
Intel's weakest link


Join Date: Dec 2006
Location: Florida
Posts: 188
Trade rep: 0 (0%)
Infractions: 0/0 (0)
C.Jackson Jul 16th, 07, 02:07 AM #8
Many others are having the same problem as you can see here. It's a problem that Intel will niether address let alone fix as this is a problem that has been present since the release of the boards.
us.png
Soon
Soon's Avatar
Pokemon Trainer


Join Date: Nov 2004
Location: Tampines
Posts: 7,297
Trade rep: 47 (100%)
Infractions: 0/0 (0)
Soon Jul 18th, 07, 04:52 PM #9
I have the same issue. Using on-board X3000. Appearance that boards gone out a few times with purple, green etc single screens then it restarts. Very irritating. Was pairing with Intel 925 D processor here.
sg.png
DavidC1 Registered User


Join Date: Feb 2007
Posts: 187
Trade rep: 0 (0%)
Infractions: 0/0 (0)
DavidC1 Jul 18th, 07, 05:35 PM #10
My temps with DG965WH:

Just 4 windows open

Processor temperature: 30C
Motherboard temperature: 37C
Motherboard temperature: 36C
I/O Controller Hub(ICH) temperature: 62C
Memory Controller Hub(MCH): 46C

looks like I am well off .

I have QST on, and I use the stock heatsink, and the fan is running at 877RPM to cool the CPU to 30C. I think they give two temperature measurements for motherboard since there are two locations for temperature measurement(for accuracy??). I also have a outlet fan.
ca.png
Soon
Soon's Avatar
Pokemon Trainer


Join Date: Nov 2004
Location: Tampines
Posts: 7,297
Trade rep: 47 (100%)
Infractions: 0/0 (0)
Soon Jul 19th, 07, 07:15 PM #11
Quote:
Originally Posted by DavidC1
My temps with DG965WH:

Just 4 windows open

Processor temperature: 30C
Motherboard temperature: 37C
Motherboard temperature: 36C
I/O Controller Hub(ICH) temperature: 62C
Memory Controller Hub(MCH): 46C

looks like I am well off .

I have QST on, and I use the stock heatsink, and the fan is running at 877RPM to cool the CPU to 30C. I think they give two temperature measurements for motherboard since there are two locations for temperature measurement(for accuracy??). I also have a outlet fan.
My office's ICH just went 82*C and kaput away. Crashed and restarts... then crash and restarts again....
sg.png
Chirag Registered User


Join Date: Apr 2007
Location: India
Posts: 10
Trade rep: 0 (0%)
Infractions: 0/0 (0)
Chirag Jul 21st, 07, 09:37 PM #12
Quote:
Originally Posted by Scooter
According to the info on Intel's web site, the temperature of the ICH is being measured right on the die, not in some external sensor. Thus, I don't see anything necessarily being wrong with temperatures in the 70's or even 80's. Call me calavier, but I trust Intel to understand and manage their own devices' temperatures correctly...
I find Scooter's logic more acceptable. ICH is actually being measured right on the die. CPU fan is responsible for CPU Temperature and ICH doesnot come under CPU fan.

However, C. Jackson do have a good point about high CPU temperature. But I think I/O Controller doesnot depend for it's cooling upon CPU fan.
in.png
keat Registered User


Join Date: Jan 2006
Posts: 37
Trade rep: 0 (0%)
Infractions: 0/0 (0)
keat Jul 22nd, 07, 07:06 PM #13
yes, i also trust intel on temperature management, but if i dont remember wrong, it actually state a maximum temperature of 85°C. But having a ICH that operate at a region of 76°C actually leave less than 10°C as a buffer zone. Some more this is temperature when the CPU is running with EIST kicks in. What if the CPU is running at 100% load for a prolong period of time, then temperature inside the casing will rise considerably and the ICH temperature will hit the 85°C limit easily.
sg.png
Chirag Registered User


Join Date: Apr 2007
Location: India
Posts: 10
Trade rep: 0 (0%)
Infractions: 0/0 (0)
Chirag Jul 23rd, 07, 07:46 PM #14
Quote:
Originally Posted by keat
yes, i also trust intel on temperature management, but if i dont remember wrong, it actually state a maximum temperature of 85°C. But having a ICH that operate at a region of 76°C actually leave less than 10°C as a buffer zone. Some more this is temperature when the CPU is running with EIST kicks in. What if the CPU is running at 100% load for a prolong period of time, then temperature inside the casing will rise considerably and the ICH temperature will hit the 85°C limit easily.
Under such a "SYSTEM", where the Temperature inside the casing has increased... The CPU Fan speed won't AID. Now it is upto REAR Casing FAN or any other that you have fixed (SIDE OR FRONT in some casings i have seen), provide better cooling. The CPU Fan uses INFLOW of air to cool the heatsink. Releasing HOT air. It is now needed to let this air to escape the casing. So, Faster CPU fan speed will result in higher collision of air with heatsink, releasing more heat...thus increasing the amount of heat inside casing. Which will raise casing temperature.

But Consider, lower fan speed and higher CPU temperature. Casing temperature won't rise in proportion to CPU temperature. However, CPU temperature will go very high but not CASING TEMPERATURE. So, which won't result in any effect on ICH temperature.

Now again, If CPU fan speed is high and CPU under load for long time. Will increase casing temperature. Thus, increasing ICH temperature. BUT...I don't remember exactly, according to a law of physics...diffussion occurs from higher to lower concentration. So, Atmospheric Temperature is Let's say 20°C and ICH Temperature is 76°C, which is the better attraction for the heat to diffuse? I feel Atmosphere. Oh and one more point "If the concentration is high (here ICH temperature) it will require more energy (heat) to have some change in it's current state (heat level)." Thus, the point is...for any considerable change to occur in ICH temperature...very high difference is required inside casing, which would definitely not let CPU function.

And one more thing, the Air pressure inside CASING will reduce. Atmosphere will have high pressure, thus inflow of air can be expected through any little crack available in the casing and heat will escape in proportion. But in case of fast CPU fan, which would result into more casing heat...the proportion of heat escape won't be equal to heat generated...thus over heating.

I'm not too sure about my theory, but...i feel the pervious point of keat is a bit irrelevant. Please, shed some light on it if any of you can...

Thank you!
in.png
Last edited by Chirag; Jul 23rd, 07 at 07:54 PM..
keat Registered User


Join Date: Jan 2006
Posts: 37
Trade rep: 0 (0%)
Infractions: 0/0 (0)
keat Jul 23rd, 07, 08:14 PM #15
Hmm this is an interesting topic.

Okay, let say when the CPU under load for a period of time, the CPU temperature will rise, then casing temperature will rise too. You are correct to say that under the law of physics, diffusion occurs and high temperature from ICH will still move on to the immediate atmosphere temperature, which is the casing temperature.

However when CPU is under load, the temperature in CPU and casing rises accordingly, the fan controller didnt work as it should by increase the front and rear fan speed according to the temperature, then the air inside the casing will not able to remove the more heat at the same speed. Thus casing temperature will increase too.

Now we go back to the law of physics u mentioned. I assume the 76°C of ICH is the equilibrium temperature maintain at the ICH after diffusion occurs during the period when the CPU is not under load. But when the CPU is under load, casing temperature will increase, then the equilibrium temperature at the ICH after diffusion will increase too.
sg.png
Thread Tools Display Modes
Linear Mode Linear Mode