8700k idle temps for watercooled R7 and R8 owners?

@22GREEN   At idle about 10 min ago

Looking forward to seeing your image and continuing this discussion. Unfortunately . . . 24 hours is insufficient time for the forum moderators to approve your photo.   

Interesting I could see it as soon as I posted. I guess it's a 24 hr. turn around. I would think it would be available today.

@22GREEN  Interesting I could see it as soon as I posted. I guess it's a 24 hr. turn around. I would think it would be available today.

Yes, you can always see your own photos prior to moderator approval, if you are logged in.

Photo is approved now, and the 27C CPU temp at idle seems to be the most popular.

I must say I am really happy with my R7 i've had it for little over 2 years. Just updated Windows 10 to 1903 and it installed flawlessly and no issues after a couple weeks. Created a backup image but disappointed we can't upgrade to an i9 and with the R9 out its looking bleak..

Dell Alienware Aurora R7
OS: Windows 10 64-bit
Motherboard: Alienware 0VDT73
CPU: Intel i7K 8700K Coffee Lake-S (6-Core/12-Thread, 12MB Cache up to 4.7GHz)
SSD: (OS) 512GB M.2 PCIe w/1TB 7200RPM SATA 6GB/s (Storage)
HDD: 1TB HP External Drive
RAM: 16GB Dual Channel DDR4 @2666MHz
Graphics: NVIDIA GeForce GTX 1080 Ti 11GB GDDR5X
' 4K LCD
Power Supply: 850W EPA Bronze PSU Liquid Cooled Chassis
Logitech z506 5.1 Surround / Arctis Wireless Pro headset
Logitech Wireless Keyboard K800 / Performance MX Mouse

You guys have great temps.  I'm going to have to look at power draw a little closer.

@22GREEN  - when you posted that screen shot with your CPU at 27 C, do you remember what your ambient air temps were?  Are you also at stock clocks or OC? If OC, what's vcore set to?

As for the IHS on my 8700k, there's a small spot near the center where the TIM doesn't appear to have been pushed out.  So either the IHS is slightly concave at that spot or the copper heatsink is.  Will have to take a closer look this weekend.

@amstel78

Thanks nothing special going on. My room temps I wish I knew at the time but for the most part its on the cooler side. I'm in my furnished basement - upstairs is central air and warmer on any given day with central air off. It always cooler down there.

No OC at all. Everything runs stock.

I've never opened my case lol i'm curious but i've seen plenty of pics to know whats in there I guess.

I was playing Wolfenstein II sometimes I check my temps right after heavy gaming but right now i'm playing Fallout 3 so I don't think my CPU/temps are really being tested but will check when I load up Wolfenstein: Cyberpilot

You guys have great temps.  I'm going to have to look at power draw a little closer.

My garage where the R8 is located can be frigid cold and I never went below 30C for CPU temp. It used to be 27C when the R8 was new and I noticed the CPU temp ticked up after I swapped out the stock fan for the ML120 Pro. Even pushing the fan to max speed, it's around 30C, idle, with balanced power profile so Vcore can dip to < 0.7V. Maybe I should swap back the stock fan and test it out. 

@GTS81   Maybe I should swap back the stock fan and test it out.

I read a very interesting article to which someone in the Community posted a link. The technical aspects of the article were far beyond my metallurgical understanding, but the gist of it was:

. . . for best component performance, thermal stability is a more important factor than the coolest possible temperature. Meaning (and I am making up these values) . . . your CPU will perform better (and last longer) if you can maintain the temperatures between, say 45C and 60C, than if you can idle at 25C, but temperatures will spike up to 80C, then back to 30C, then spike again to 70C.

I am not saying this is true, but this was the contention in the article.

Yeah, heat goes hand in hand with component degradation.  That is probably true for many components, not just the CPU.  For example, someone who uses their GPU for data mining, has it running 24/7, does not OC, and runs it constantly at 60*C, might expect a longer useful life than someone who OC's and is constantly stressing the the GPU to the max with wild temperature fluctuations.  Aside from external factors (e.g., airflow) and internal factors (e.g., voltages) such spiking is influenced too by the use to which you put your PC.  Like, for example, you're unlikely to spike it to 80*C playing minesweeper. 

Like, for example, you're unlikely to spike it to 80*C playing minesweeper. 

Where's my minesweeper?! How many years have Microsoft secretly yanked it from standard Windows installation?? Why did you have to remind me? Now I can't even find Solitaire! 

I don't know ... these days I only occasionally play it on my phone to kill time on the crapper.  Just tell google "minesweeper" and a blue "play" button will show up.  It's a little more colorful than the 90's version. 

Anyway, speaking of wild fluctuations, I let HWminitor run in the background for a bit today and just noticed the CPU IA core sensor registered a 2,920.74 watt power draw at some point??  I realize this is just an estimate, but ...   

Also, to keep on point, end of day @ 76*C room temp. 

I should probably get a UPS, but I thought the surge protector and EVGA PSU, would be the first lines of defense against power fluctuations before reaching the CPU. There weren't any outages or flickering lights that I noticed yesterday.  

Ha.. typing too quickly.  

Did I say 76C? Lol... If so, my bad.

I haven't had much time to mess with the R7 lately. I've been dialing in newly issued loads for my SASS. Testing and recording DOPE readings for 1000+ yards engagements is incredibly tedious and entirely dependent on the weather. The past few days have been good so I'm prepping and will hopefully get my re-cert over at the Westpoint range done by Monday.

As soon as I'm in front of the R7 though, I'll test by stabilizing room temperature to 73F, re-pasting with MX4 again, and ensuring IA voltage is stock. I'll record those temps then perhaps reinstall the OE fan and retest under similar conditions. 

It might be worthwhile testing under safe mode also with just minimum drivers and background tasks loaded to ensure a consistent baseline.