cdurkinz Posted January 16, 2016 Share Posted January 16, 2016 Hi, I was just wondering if anyone else noticed this. The CPU temp that the Sensor Panel shows for my 5820k just seems high to me, as my CPU is watercooled. Other programs are showing a lower temperature as well though, which you can see in the picture: So I was wondering if anyone else has seen this? I was about to re apply my thermal paste thinking I may have done it wrong or something but now I'm not sure whether the Sensor Panel may just be wrong or could Speccy be wrong? Thanks! Quote Link to comment Share on other sites More sharing options...
Fiery Posted January 17, 2016 Share Posted January 17, 2016 Modern processors have multiple temperature diodes. There's a good chance Speccy reads a different sensor and shows it as "CPU". It could be the CPU temperature as measured by the motherboard sensor chip, one of the CPU Core sensors, the CPU Package sensor, etc. Make sure to check all CPU-related temperatures on the Computer / Sensor page in AIDA64, and compare the values against HWMonitor and the factory monitoring software came with your motherboard (e.g. Asus AI Suite for Asus motherboards). Quote Link to comment Share on other sites More sharing options...
cdurkinz Posted January 17, 2016 Author Share Posted January 17, 2016 Thank you, I checked it out and I'm not sure exactly what speccy might be using, maybe just Core 1? lol. Aida64 matches HWMonitor. Here is a pic inside Computer / Sensor: It's weird because all of the cores are nice and cool so I guess I don't get where the "CPU" and "CPU Package" sensors are? I really don't like such a big difference in idle temps. But a +/- 10°C difference from the cores to where ever the other sensors are just seems crazy. They can't be too far away, right? I'm almost positive that some of the temperature might be due to it being so cold outside and having the heat up right now in the house. But that only goes so far, I mean it isn't 40°C in my house lol. My rig is nowhere near a vent but I know ambient temps add to it. I don't know, I built this rig a few months ago, it's been fine up until recently. All of a sudden my watercooled CPU and GPU temps at idle are in the upper 30's or even 40's when they used to be in the mid 20's. Doesn't make any sense, and I just recently thoroughly dusted the radiators and everything else too. I even checked them to make sure and they are spotless right now. It's very weird. I'm starting to wonder if the Arctic Silver 5 TIM isn't working so well, weird thing is that it's supposed to work BETTER after it cures lol. So frustrating. Quote Link to comment Share on other sites More sharing options...
Fiery Posted January 17, 2016 Share Posted January 17, 2016 High variation between various CPU readings is absolutely normal, and you shouldn't worry about it that much. Quote Link to comment Share on other sites More sharing options...
cdurkinz Posted January 24, 2016 Author Share Posted January 24, 2016 So I have an update, sort of. When I booted my PC today the AIDA64 sensor manager, HWMonitor, and Corsair Link were reporting that my CPU was 120°C. So needless to say I about freaked out. I shut the PC down immediately and removed the water block off of the CPU and re-applied the thermal paste, even though the paste on there looked perfect. I booted the PC again, and still it's reporting 120°C. So I'm thinking something weird is going on here, I mean the cores were at 29°, 30°, 31, etc. Super low. So I shutdown the PC again and boot into the BIOS, the Bios's hardware monitoring said the motherboard, cpu, etc were all fine at 30-34°C. So I remembered when I first built the PC a few months ago that some of the monitor's weren't reporting correctly because of the chipset being newer or something? Apparently at the time at least, x99 was newer and it just didn't work with some of the monitors or something? Anyways, so I shutdown the PC again, switched the onboard BIOS switch back to the stock Bios and tried to reboot. Well Windows 10 didn't want to work with an older Bios apparently because it kept trying to repair my OS and wouldn't boot. So I shutdown one last time, switched back to the most recent BIOS, and cleared my CMOS and rebooted. PC rebooted fine, Windows 10 loaded up nice and quick, and all of the temp monitor's now show the CPU at 34°C. I just wanted to give a heads up, I'm not sure what the issue is here but it sounds like something to do with my motherboard, BIOS, or something on my end. I figure at least someone else might see this who is having the same issue and get some relief. My motherboard is an MSI X99S SLI Plus just by the way. Anyways, thank you for the help admin. Quote Link to comment Share on other sites More sharing options...
cdurkinz Posted January 24, 2016 Author Share Posted January 24, 2016 This is driving me crazy it's doing it again. I have the side of my case off, I know the pump is running, if I touch the CPU's block it is not even warm to the touch, neither is the tubing, but it is supposedly 120°C. Everything is acting normally, I cannot for the life of me understand what is going on. Speccy says my CPU is 27°C though I know this may not be real, but Aida 64 and HWmonitor says 120°C even though the cores are in the 20's. I was starting to think it's somehow on my end, now I am once again unsure. Please help me out somebody. I'm losing my mind over this: and then there's Corsair Link: Quote Link to comment Share on other sites More sharing options...
Fiery Posted January 25, 2016 Share Posted January 25, 2016 It seems to be an intermittent hardware failure of the Intel DTS (Digital Thermal Sensor) subsystem. We've seen such issues before, although it's very rare to happen. Quote Link to comment Share on other sites More sharing options...
cdurkinz Posted January 25, 2016 Author Share Posted January 25, 2016 It seems to be an intermittent hardware failure of the Intel DTS (Digital Thermal Sensor) subsystem. We've seen such issues before, although it's very rare to happen. I take it that this means it's basically a failure in how the sensor reads the temperature? I guess what I mean is this something I need to be very worried about? Or are the temperatures of my cores, my motherboard, and everything else good enough to go by? I mean I've been contemplating buying a new motherboard all day today, and x99 motherboards aren't cheap to put it nicely, but that's how much this is driving me nuts. It's just so weird that it's been fine for months, and then today it happens on boot. I was able to get it to work again for a while, but then after rebooting it went back to this high reading and I have not been able to replicate the "fix" that I think I luckily pulled off earlier. I have also read of at least two other guys with the same motherboard, different haswell-e's who had the exact same problem. We are talking to the exact temperature (120°C). Quote Link to comment Share on other sites More sharing options...
Fiery Posted January 25, 2016 Share Posted January 25, 2016 I'm not a CPU engineer, but to me it seems to be the fault of the CPU, where the DTS logic is located. But just because of that issue, I wouldn't go out and replace any of my hardware. There're several other temperature sensors in the computer that you can watch to detect an overheating situation. But, if you're considering replacing your CPU or motherboard, then I'd still wait a few months, since the successor to your CPU, Broadwell-E is right around the corner. With the new LGA2011v3 CPU generation I'm pretty sure motherboard vendors prepare a new fleet of X99 motherboards as well. Quote Link to comment Share on other sites More sharing options...
Fiery Posted January 25, 2016 Share Posted January 25, 2016 I'm sorry, I thought the temperature is 120+ Celsius on the DTS rails, which are listed as "CPU #1 / Core #n" temperature by AIDA64. It seems culprit is a temperature measured by the motherboard sensor chip. Have you tried checking the temperatures using MSI's own monitoring software? Quote Link to comment Share on other sites More sharing options...
cdurkinz Posted January 27, 2016 Author Share Posted January 27, 2016 I'm sorry, I thought the temperature is 120+ Celsius on the DTS rails, which are listed as "CPU #1 / Core #n" temperature by AIDA64. It seems culprit is a temperature measured by the motherboard sensor chip. Have you tried checking the temperatures using MSI's own monitoring software? Sorry I haven't checked back in so long, I've been real busy. Yes it's strange, when I boot into the Bios and select the Bios's built-in HW Monitoring, it shows the CPU in the 30-35°C range and the motherboard is the same. Those are the only temperatures it gives me, it won't break it down per core or anything unfortunately. So it really just adds to the strangeness of it all. Unless you are talking about a different MSI monitoring software? Quote Link to comment Share on other sites More sharing options...
Fiery Posted January 27, 2016 Share Posted January 27, 2016 I meant MSI Command Center. Doesn't it have a monitoring module? http://download.msi.com/uti_exe/command_center.zip Quote Link to comment Share on other sites More sharing options...
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.