Is 100 Degrees Hot For a CPU?

[Kieeran]

Hi guys, so recently, I have been getting a low FPS in the majority of my games, especially CSGO which seems to hammer CPUs. I began wondering why I was getting such low frames and started doing some monitoring of my CPU temperature. I noticed that my temperature turned to 100 degrees when playing CSGO, could this be a reason for why I am getting such a low fps? If so, how would I go about keeping the temperature low again, because I have only had this problem recently.

 

[JackNaylorPE]

Yes....throttling kicks in in the high 90s for Intel CPUs.

 

[dextermat]

if you keep it up, you will toast your chip

 

[sivthasan_2001]

I would recommend you getting a better cooler or re-applying the thermal paste on the CPU ; but not too much (a pea-sized amount on the middle). 100 degrees is very hot and if you keep it up, your CPU would get toasted!! Your CPU shouldn't get any hotter than 80 Degrees, at 100% Load (in my view). That is probably why you are achieving such a low FPS.

 

[babernet_1]

Yes, that is a dangerous temperature and will cause your CPU to throttle down. Get a better CPU cooler.

 

[sz0ty0l4]

Yes its clearly too high. you should downclock, or get a better cooling system. did you experience shut downs? would help if you could list specifications.
It can happen that your cooling system resigned. you could check if the fans are working while running heavy games, just take off the side of the case and take a look

 

[The Original Ralph]

i ran mine, unknowingly, at 99-100C when editing videos, for about 3 months, and i'm talking files that took anywhere from 1 - 4 hours to render. Reason i say unknowingly, the ASUS performance monitoring utility never exceeded 67C. It wasn't till i ran RealTemp, then intel's XTU (Extreme tuning Utility) that i realized what temps i'd been hitting. Assuming you're running an intel cpu, it has built in safety limits that throttle it, in terms of current, power and thermal throttling, so those basically saved mine from self-destructing.

Point is, 3 months of operating at those temps has not affected my cpu, i7-4790 (locked) - it may have shortened it's life expectency but performance hasn't been effected.

is your cpu intel?

 

[IamTimTech]

You should also look in your task manager to see what is eating CPU Cycles, malware will cause your CPU to work extra hard. Aside from that it's the basics. Clean out your PC with a can of compressed air and a paint brush for tough dust bunnies. Reapply your CPU cooler with fresh thermal paste and make sure all of your fans still work properly.

 

[JackNaylorPE]

i ran mine, unknowingly, at 99-100C when editing videos, for about 3 months, and i'm talking files that took anywhere from 1 - 4 hours to render. Reason i say unknowingly, the ASUS performance monitoring utility never exceeded 67C. It wasn't till i ran RealTemp, then intel's XTU (Extreme tuning Utility) that i realized what temps i'd been hitting. Assuming you're running an intel cpu, it has built in safety limits that throttle it, in terms of current, power and thermal throttling, so those basically saved mine from self-destructing.

Point is, 3 months of operating at those temps has not affected my cpu, i7-4790 (locked) - it may have shortened it's life expectency but performance hasn't been effected.

is your cpu intel?

Luckily you weren't overclocked..... at those elevated temps, with the higher voltages the OC'd CPU might have suffered deleterious effects.

If you want an all inclusive tell you everything, graph everything utility, download HWinFO64
http://www.hwinfo.com/download.php

Scroll down here and check out the screenshots

http://www.hwinfo.com/

 

[Kieeran]

I decided to take off the side panel and hoover any dust that was visible. I then decided to reinstall the heat sink just incase. I then booted my PC up with the side panel left off and launched CSGO. I am now only getting temperatures of 40 degrees and am getting a very high FPS.

I was thinking about constantly leaving the side panel off permanently, would that be safe?

 

[JackNaylorPE]

Put it back on and see if the side panel or dust cleaning did the trick.

To determine if you have adequate case cooling, I recommend taking a desk fan and having it blow in thru the open side panel. If this reduced temperature to any significant level, you need some case ventilation.

http://www.newegg.com/Product/Product.aspx?Item=N82E16835709023
http://www.newegg.com/Product/Product.aspx?Item=N82E16835709032

 

[The Original Ralph]

i ran mine, unknowingly, at 99-100C when editing videos, for about 3 months, and i'm talking files that took anywhere from 1 - 4 hours to render. Reason i say unknowingly, the ASUS performance monitoring utility never exceeded 67C. It wasn't till i ran RealTemp, then intel's XTU (Extreme tuning Utility) that i realized what temps i'd been hitting. Assuming you're running an intel cpu, it has built in safety limits that throttle it, in terms of current, power and thermal throttling, so those basically saved mine from self-destructing.

Point is, 3 months of operating at those temps has not affected my cpu, i7-4790 (locked) - it may have shortened it's life expectency but performance hasn't been effected.

is your cpu intel?

Luckily you weren't overclocked..... at those elevated temps, with the higher voltages the OC'd CPU might have suffered deleterious effects.

If you want an all inclusive tell you everything, graph everything utility, download HWinFO64
http://www.hwinfo.com/download.php

Scroll down here and check out the screenshots

http://www.hwinfo.com/

appreciate - obviously a lot more info than the intel XTU provides, but i like it cause it shows me throttling in real time - here's a screen shot. The three throttling limits are all at "0" in this shot, but if it's throttling any one of those items, it reports it. But one of the advantages to me, is that i can adjust the green line chart from 3 minutes to 1 hour, and then move the cursor along the peaks and see what values it hit at those points, whether it was throttling, what memory utilization was, core voltage etc, etc. That and i can adjust voltage & current from the utility without having to reboot into BIOS.

and if anyone is interested, the XTU utility can be downloaded at Intel XTU download

As to OCing, i had and have the Turbo boost enabled - the i7-4790 is 3.6 GHz and 4.0 with turbo boost.

btw, tks on the USP assist - i did as you suggested and did much more research, and exactly as you reported, APC seems to be disappointing to a lot of folks, ended up going with a cyberpower unit

 

[The Original Ralph]

Put it back on and see if the side panel or dust cleaning did the trick.

To determine if you have adequate case cooling, I recommend taking a desk fan and having it blow in thru the open side panel. If this reduced temperature to any significant level, you need some case ventilation.

http://www.newegg.com/Product/Product.aspx?Item=N82E16835709023
http://www.newegg.com/Product/Product.aspx?Item=N82E16835709032

what JackNaylorPE said - the final bit of temp adjustment on mine, to bring temps down to where i wanted them, came when i added three 60mm fans blowing directly onto the CPU and MB - i cut openings in the side case cover and fabricated a mount from fiberglass, but a thin polymer sheet would have worked as well - my cpu lives in a SFF case, and i'm actually thinking of upgrading to a larger silent case - the sound of a hoover vacuum cycling up doesn't do a great deal for my enjoyment level.

 

[JackNaylorPE]

appreciate - obviously a lot more info than the intel XTU provides, but i like it cause it shows me throttling in real time - here's a screen shot. The three throttling limits are all at "0" in this shot, but if it's throttling any one of those items, it reports it. But one of the advantages to me, is that i can adjust the green line chart from 3 minutes to 1 hour, and then move the cursor along the peaks and see what values it hit at those points, whether it was throttling, what memory utilization was, core voltage etc, etc. That and i can adjust voltage & current from the utility without having to reboot into BIOS.

btw, tks on the USP assist - i did as you suggested and did much more research, and exactly as you reported, APC seems to be disappointing to a lot of folks, ended up going with a cyberpower unit

You can do the same thing and more in HWiNFO. Open it and look at the list of 100+ items being monitored. Right click on any of the listed items to initiate a graph. You could perhaps for example set up a 3 x 4 grid and graph:

Core 0 Voltage - Core 1 Voltage - Core 2 Voltage - Core 3 Voltage
Core 0 Temp - Core 1 Temp - Core 2 Temp - Core 3 Temp
Core 0 Throttling - Core 1 Throttling - Core 2 Throttling - Core 3 Throttling

The thing is tho Intel XTU doesn't put a very strong load on ya CPU. I'm not suggesting ya use P95 as it's load is entirely unrealistic and therefore a waste if T & E IMO, but RoG Real Bench for example, will use actual programs (image editing, video encoding, etc) to put realistic program loads on your CPU. You can expect as high as a +6C difference with RoG RB as compared to Intel XTU. And it's a multitasking load rather than just a continual hammering of the same thing over and over again.

With regard to real time BIOS adjustments, you can also do this, and more reliably, with the MoBo manufacturer's MoBo utility. Sometimes a change can cause an instability in and of itself. Don't get me wrong, I do use Intel XTU, but as a final check sorta thing. After a 2 hour run under Rog Real Bench, I'll run XTU overnight "just to make sure'.

 

[The Original Ralph]

i'd already downloaded it and was going to check it out later when everyone's gone to zzz land

as to using the mfgr's mobo utilities, for whatever reason, it was the asus AI Suite 3 that was partially giving me high temps, per intel's recommendation i un-installed it and temps dropped a few degrees - apparently it's creating a conflict with the BIOS - that fact plus the fact that it never read over 67C bothered me - it would show 67C while RealTemp, XTU and one other (CPUID irrc) were showing 96-100C

thanks - i will check out HWiNFO

 

[JackNaylorPE]

i'd already downloaded it and was going to check it out later when everyone's gone to zzz land

as to using the mfgr's mobo utilities, for whatever reason, it was the asus AI Suite 3 that was partially giving me high temps, per intel's recommendation i un-installed it and temps dropped a few degrees - apparently it's creating a conflict with the BIOS - that fact plus the fact that it never read over 67C bothered me - it would show 67C while RealTemp, XTU and one other (CPUID irrc) were showing 96-100C

thanks - i will check out HWiNFO

The two utilities are reading two different things. My CPU package temp as read by AI Suite never breaks 41C; My core temps never break 72C. Both are absolutely correct. HWiNFO64 will show you both.

AI Suite reads CPU Package temp which is a conglomeration of several different external temp sensors. Asus uses this for various control functions. CPU Core Temp is an entirely different thing and measures individual core temps internal to the CPU.

Uninstalling AISuite is a bad thing in that it removes one of the best features of Asus Boards and that is FanXpert. You can cause various issues if you have AI Suite's monitoring window and other monitoring utilities at the same time. Not using FanXpert to control case and CPUs fans is a sure way to limit the breadth and flexibility of your system cooling.

 

[The Orig Ralph]

i'd already downloaded it and was going to check it out later when everyone's gone to zzz land

as to using the mfgr's mobo utilities, for whatever reason, it was the asus AI Suite 3 that was partially giving me high temps, per intel's recommendation i un-installed it and temps dropped a few degrees - apparently it's creating a conflict with the BIOS - that fact plus the fact that it never read over 67C bothered me - it would show 67C while RealTemp, XTU and one other (CPUID irrc) were showing 96-100C

thanks - i will check out HWiNFO

The two utilities are reading two different things. My CPU package temp as read by AI Suite never breaks 41C; My core temps never break 72C. Both are absolutely correct. HWiNFO64 will show you both.

AI Suite reads CPU Package temp which is a conglomeration of several different external temp sensors. Asus uses this for various control functions. CPU Core Temp is an entirely different thing and measures individual core temps internal to the CPU.

Uninstalling AISuite is a bad thing in that it removes one of the best features of Asus Boards and that is FanXpert. You can cause various issues if you have AI Suite's monitoring window and other monitoring utilities at the same time. Not using FanXpert to control case and CPUs fans is a sure way to limit the breadth and flexibility of your system cooling.

i believe i understand what you're saying, but i'll clarify - a) i couldn't understand why the AI Suite would conflict with other reporting or monitoring utilities, but wanted to see if that was the case - i dis-abled AI Suite by exiting it in the taskbar, then "ending the process" in task manager, idle temps dropped 3-4C, whether monitoring in XTU or RealTemp, and that drop remained after i un-installed it altogether.

When i read the manual on the AI Suite, and it's been a few months now, irrc the temp reported is what amounts to the cpu's case temp - when i would compare AI performane utility's temp with the other temps in XTU & RealTemp, they were within 5-6 degrees of each other, and as they're reporting core temps, that is the difference that intel indicates to expect. But the Asus reported temp would hit a ceiling of 67C (somewhere in the high 50s it's temp would start to equal the temps XTU and RealTemp were reporting for core temps) - Why would Asus ignore, in it's temp reporting, core temps hitting 99-100C?

The suggestion to un-install it was from the intel reps in that overheating thread re the i7-4790k & i5-4690k CPUs, and it was a generic suggestion solution, not directed at me specific - i only tried it in desperation and did see some positive result.

Then when i cleared the CMOS (per intel's suggestion) and temps dropped considerably more, approx 6-7C, i wanted to see if that cleared the issue between AI Suite as i couldn't understand a mobo mfgr's utility mis-reporting like it was and causing a rise in temps. The same situation returned, where MAX temp AI Suite would show was 67C - and oddly it remains there after a stress test, even though temps have dropped.

The aspect of the asus monitoring utility that really alarms me though is that it would ignore those high core temps but the situation i'm seeing may be unique to my mobo, i don't know, i do know what i'm seeing though

the other day i re-installed it for a 2nd time with same results. As far as fan control, i'm happy with the settings in the UEFI BIOS - i've adjusted them manually to where i want them. Between the un-installing Ai Suite, clearing CMOS, new cooler and additional case fans, i've got my temps at idle in the low 30s, and editing video files in the low to mid 60s, and XTU stress test in the low to mid 70s with occasional spikes to 79C

There was another anomaly with the Asus suite showing in the taskbar but not running. Someone in the intel thread had noticed this but with asus in the taskbar, adjusting watt limits for turbo boost power max or turbo boost short power max in XTU, you had to use a number that was 1/8th of what you wanted, somehow a multiplier of 8X had been factored into the bios - ie, if you wanted turbo boost power max of 88W, you had to select 11W, and you'd find it at 88 in the BIOS.

i'm fairly certain there was a conflict between my bios (and it's the latest vs) and the asus suite

 

[JackNaylorPE]

i believe i understand what you're saying, but i'll clarify - a) i couldn't understand why the AI Suite would conflict with other reporting or monitoring utilities, but wanted to see if that was the case - i dis-abled AI Suite by exiting it in the taskbar, then "ending the process" in task manager, idle temps dropped 3-4C, whether monitoring in XTU or RealTemp, and that drop remained after i un-installed it altogether.

When i read the manual on the AI Suite, and it's been a few months now, irrc the temp reported is what amounts to the cpu's case temp - when i would compare AI performane utility's temp with the other temps in XTU & RealTemp, they were within 5-6 degrees of each other, and as they're reporting core temps, that is the difference that intel indicates to expect. But the Asus reported temp would hit a ceiling of 67C (somewhere in the high 50s it's temp would start to equal the temps XTU and RealTemp were reporting for core temps) - Why would Asus ignore, in it's temp reporting, core temps hitting 99-100C?

The suggestion to un-install it was from the intel reps in that overheating thread re the i7-4790k & i5-4690k CPUs, and it was a generic suggestion solution, not directed at me specific - i only tried it in desperation and did see some positive result.

Then when i cleared the CMOS (per intel's suggestion) and temps dropped considerably more, approx 6-7C, i wanted to see if that cleared the issue between AI Suite as i couldn't understand a mobo mfgr's utility mis-reporting like it was and causing a rise in temps. The same situation returned, where MAX temp AI Suite would show was 67C - and oddly it remains there after a stress test, even though temps have dropped.

The aspect of the asus monitoring utility that really alarms me though is that it would ignore those high core temps but the situation i'm seeing may be unique to my mobo, i don't know, i do know what i'm seeing though

the other day i re-installed it for a 2nd time with same results. As far as fan control, i'm happy with the settings in the UEFI BIOS - i've adjusted them manually to where i want them. Between the un-installing Ai Suite, clearing CMOS, new cooler and additional case fans, i've got my temps at idle in the low 30s, and editing video files in the low to mid 60s, and XTU stress test in the low to mid 70s with occasional spikes to 79C

There was another anomaly with the Asus suite showing in the taskbar but not running. Someone in the intel thread had noticed this but with asus in the taskbar, adjusting watt limits for turbo boost power max or turbo boost short power max in XTU, you had to use a number that was 1/8th of what you wanted, somehow a multiplier of 8X had been factored into the bios - ie, if you wanted turbo boost power max of 88W, you had to select 11W, and you'd find it at 88 in the BIOS.

i'm fairly certain there was a conflict between my bios (and it's the latest vs) and the asus suite

Yes, there is an issue with AI Suite where it can be discombobulated and has to be uninstalled and reinstalled. There is a utility on RoG Forums designed explicitly to do that. I find that numerous utilities and programs leave what I call ghost images in Task Manager. Among them I can recall Firefox, HWiNFO64, Powerdesk, etc. I surmise this happens when you have a program or utility that has various modules running and when you close them, one of the modules isn't finished doing something it wants to so and remains running. If you restart, it won't restart and you can wind up with 3 or 4 instances in TM.

The thing to recognize is that AI Suite is an suite of utilities, and you're speaking of just one of them.

AI Charger
USB Boost
EZ Update
Network Control
BIOS Flashback
System Info
USB Charger
WiFi Go
WiFi Engine
4-Way Optimization (Auto OC)
- TPU is the BIOS adjustment function thru windows
- EPU is energy Control
- FanXpert auto calibrates all your fans and allows you to set up speed control curves for each channel.

If nothing else, I would set up FanXpert and Flashback. It's not that Ai Suite is ignoring Core Temps, it's just that as a MoBo utility, it exercises no control over core temps.. This is built into the CPU. It has been reported that this behavior has been observed sometimes when running multiple monitoring programs at the same time. You can also shutdown the hw monitor service but leave the fan controller service running to maintain fan control.

If you had problems with AI Suite, uninstall using the utility on the Asus Forums .... and reinstalling is the recommended procedure.

https://rog.asus.com/forum/showthread.php?36977-How-to-Install-Ai-Suite-3!-%28For-People-Who-Cannot-install-the-new-AI-Suite-3%29&highlight=uninstall+cleaner
http://rog.asus.com/forum/showthread.php?34503-AI-Suite-III-Cleaner

You can set the fan curves up in AI suite, then disable Ai suite III from starting up at login, the Asus fan service, that runs as a Windows 7 service, will apply the settings set in AI Suite at boot and not the BIOS settings As long as you don't stop the fan controller service.

I have never used AI Suite to do any BIOS settings adjustments simply because the very act of changing these settings under load can crash the system. It's good for establishing a starting point but all final fine tuning should be done directly in the BIOS.

Again, Intel XTU is "watered down" a bit so as not to put too much of a consistent a load on the CPU. While it's good for a long term stability check since it's cycling prevents and serious temps, you will be able to generate much higher temps doing say video encoding. In addition, with all of the extension sets used in modern CPUs, it doesn't test all of them, leaving you as yet uncertain as to whether your system is still stable when such conditions occur. And the fact that you sometimes get crashes after 4+ hour run is frustrating.

What I do is, say you have established a stable foundation at 4.3 GHz and you're now ready to try 4.4 .... run RoG Bench multitasking test .... upping ya voltages and making adjustments to the "usual suspects" along the way..... if you pass that 2 minute test, you are very close. Run the full benchmark test and if ya pass that, run a 2 hour stress test.

I you can pass an 8 hour Intel XTU tests you are probably very close but not quite sure of a stable OC under all real usage conditions. But after passing that I find that you will often fail under RoG Bench because not only does it not have cyclic loading which puts full stress on the CPU only for short durations, it also is a 'same thing over and over and over again type of loading. RB throws thing at the CPU in varying order so as to present a true muti-tasking environment. I didn't have to make big changes as a result of the RB testing but I did wind up having to make minor changes to either VCCIN, Core or Cache voltages to get past the stress test.

 

[The Original Ralph]

thanks for the ROG link - i'd already tried using the ai suite cleaner and kept getting error msgs - i'd even posted in the asus motherboard thread on this forum and am still waiting for raja's answer, but i found it in that ROG thread

i think we've taken this thread sideways waaaaay too far - apologies to the OP