[CPU and GPU Temperatures] - Clevo P870DM2/P870DM3 (Sager NP9873/NP9872)

Greetings,

Please copy and paste the format below and post your temperatures as many are curious about the Skylake and Pascal (specifically the GTX 1080 chips) and how they do temp wise in these beasts.

* Feel free to tweak the post to your liking as the main point and purpose here is, "Organization." We're not noobs here...we get the gist.

-------------------------copy/paste below----------------------------

Machine Brand: Clevo P870DM3
GPU: 1080SLI

CPU Thermal Paste: Stock (Ex. ICD) or ______ Thermal Paste?
GPU Thermal Paste: Stock (Ex. ICD) or ______ Thermal Paste?

Ambient Temp: ##C

---> Run OCCT: (15 minutes minimum.)

CPU Stock Clocks:

• CPU Idle Temp: ##C
• CPU Max Temp: ##C
• Max Under Volt: ##mv (Core and Cache)

CPU Over-Clocked:

• CPU Idle Temp: ##C @ ####MHz
• CPU Max Temp: ##C @ ####MHz
• Volt +/-: ##mv (Core and Cache)

GPU Stock Clocks:

• GPU [1] Max Temp: ##C
• GPU [2] Max Temp: ##C or N/A

---> Run Cinebench R15, Fire Strike and especially Unigine Heaven:

GPU Over-Clocked:

• GPU [1] Max Temp: ##C @ ####MHz
• GPU [2] Max Temp: ##C @ ####MHz or N/A

Links / Screenshots / Scores - Benchmark Scores/Validation:

• HWiNFO64: (Post screen shots showing temps.)
• FireStrike:
• TimeSpy:
• Cinebench R15:
• OCCT: (30 minutes minimum.)
• Gaming: (Game, Settings, Duration...)

Notes:
Please provide any additional info. that you would like to share. Ex. "I've tried CLLU / GC (Liquid Metal), but the heat sinks seem to be warped so I switched to traditional paste"...etc...

------------------------------------------------------------------------------

Organization is key and I hope this helps others who are seeking this info.

* If there are any additional categories or info that you would like to see added, please let me know and I'll update the OP.

Cheers

 

2

 

3

 

Example: Below are some of the data from HWWingNut.

Machine Brand: Clevo P870DM3
GPU: 1080SLI

CPU Thermal Paste: ICD
GPU Thermal Paste: ICD

Stock Clocks:
CPU Idle Temp: ##C
CPU Max Temp: 100C
Max Under Volt: ##mv (Core and Cache)

Over-Clocked:
CPU Idle Temp: ##C @ ####MHz
CPU Max Temp: ##C @ ####MHz
Volt +/-: ##mv (Core and Cache)

Stock Clocks:
GPU 1 Max Temp: 90C
GPU 2 Max Temp: 83C

Over-Clocked:
GPU Max Temp: ##C @ ####MHz

Links / Screenshots / Scores - Benchmark Scores/Validation:
3DMark FireStrike: Score (15868) / 18533 (Graphics)
Cinebench R15:
OCCT:

Notes:

 

Very strange, no one wants to share his results, not good at all.

 

Yea I hear ya. I hope people are not too ashamed to post them. We know the temps are rubbish on the Clevo's, but that's no reason not to post as there have been reports of a lot of tamed temps after a repaste...

I will be posting some data likely this week for the 775.

Stay tuned.

 

Not enough information to make it relevant really. Max temps under what load? Idle temperatures without ambient (for load too).

Fair temperature comparisons need a consistent (both during the test and between tests) load for about 30-60 minutes to let it even out with a rough idea of room temperature. Average clocks and general performance metrics over this time help too.

It takes time and effort to do properly so don't be surprised if people don't jump all over it.

To call owners out as being ashamed is silly.

 

I disagree.

It doesn't take rocket science to record max temps. Perhaps you missed it in the OP where I've listed the three apps to use that are specific for taxing the CPU and GPU.

Ambient temps? - If your ambient temperature is more than your CPU and GPU then one must be on a different planet closer to the Sun. On Earth, we have a small margin of temp ranges where humans live, so putting that much emphasis on not recording max temps is a bit extreme.

You don't need a science lab, especially when the component that you're testing has a thermal kelvin value that is above the absolute in ambient temperature...ie much greater than the ambient temperature.

Also to correct you...I've said that, "I hope that people are not ashamed." ..."hope" and "are," are indeed very different.

Don't worry, people are still going to buy DTR's regardless of the, "out of the box," high temps. That's not going to stop us and for the enthusiasts we look at it as a fun challenge eager to tackle.

Cheers

::iunlock::

 

It makes a noticeable difference whether your in a 15C room or a 30C room. I've had to be in both for real usage situations so I have absolutely no idea what your talking about here. Some of our users are in Canada and others in deserts. That will clearly make a difference in temps. My old GT60 was affected to the point where a nice cool room would be fine, but a hot room would throttle it. Considering max temps are usually around 80-100C in all normal circumstances a room being 15C hotter than another matters a lot. Your closer to the sun reference makes no sense. Perhaps if the average computer was hitting 10,000C than it wouldn't matter what temp your room is.

I also think what Meaker is saying (correctly) is that you aren't being specific enough for max temps. Something like "What are your max temps after running Firestrike for an hour" or something would make the results significant because then everyone is reporting the exact same thing. Firestrike will have a different max temp than say Furmark (plz don't run Furmark anyone reading this). Accurate data has to leave as little as possible to chance and you weren't specific enough for what max temps are from.

 

My max temps in a single run of firestrike, cinebench or OCCT (for a couple of minutes) are going to look amazing Especially if I take it into work and do it with the air conditioning on max.

My max temperature with heaven and OCCT running in the background for 30 minutes in my hot enclosed room at home will be quite different.

 

Yes with that much of a difference in room temps everything will be warmer, hotter and sweat more...of course.

The point that I'm making here is that most of the temps that are reported in general are usually from users that have similar ambient temps within a close margin of each other.

For those of you who are confused, you're blowing this way out of proportion. It's really not that hard or technical. I'm actually surprised it has gotten this far off to where I'm questioning if the OP is even being read thoroughly.

If you read it, you'll see the specific apps being mentioned along with the specific benches to be tested.

You're really making it sound like I should be stating the obvious here on how to run these benches? I'm getting confused here.

Max temps...that's all. ..simple.


::iunlock::

 

Most definitely. An OCCT ran for a couple of minutes would be insufficient to be considered as a valid stability test.

::iunlock::

 

... But you cant assume that! Users here all the time talk about how they lack AC! And you are the one missing the point about max temps not us... you have 3 tests, you just want people to run them and report the max temps from them? A firestrike wont hit a max temp, but running it continously could which you dont specify.

Hell, Im one of those people who has that much of a temp difference.

The reason this stuff is being brought up is because we read the OP and its ambiguous, and we want good results too.

 

Dear friends, I think we are all in the same boat, we want to get to the safety,
Collecting as much as we can from the data, analys it and extract the necessary information from it, to make the operation of our devices as best as we can.
Our objective must be the same, we can all cooperate with each other. "Working as a Team" can solve a lot of problems for us and for others who need us to show them the way, we need each other this time of a shift in new technologies, new for all of us, need patience, teamwork and consolidated efforts.

 

Note that in the OP, it does clearly state the following,

" * Feel free to tweak the post to your liking as the main point and purpose here is, "Organization." We're not noobs here...we get the gist."

" * If there are any additional categories or info that you would like to see added, please let me know and I'll update the OP."

Therefore, where's the problem and why are you making this so hard? Honest question.

Suggestions would be more helpful than weightless attacks, because it's not me missing the point here, as proven above and in the OP.

Again, whether the OP is being read thoroughly or not is in question; actually if it's being understood or not? That's what I find ambiguous.

Instead of attacking, suggestions would be appreciated. Just trying to help keep things organized here.

::iunlock::

 

Not trying to attack you, we already made the suggestions. Just add max temp from the specific test combo you think is best.

 

I've added a few things and more is always welcome.

Cheers

::iunlock::

 

Ambient temperatures will make a hell of a difference.
My DM1 used to idle at mid 20's in San Francisco , here in LA it never idles under 35C . Unless the AC has been running for over an hour.
That's why you guys need an average run duration. To let the temps settle.

 

Absolutely.

::iunlock::

 

I thought I would post some results as no one else has yet. I haven't had much time so these are just preliminary, but it will give people interested something to work with. I have HID EVOC P870DM2, single gtx 1080 with a vapor chamber heat sink. I have delidded my 6700K and repasted with CLU. I ran out of CLU, so I have used ICD between the heatsink and heat spreader. I don't have stock results, but I think most are the same out of the box.

Here are the FN 1 fan results:

Ambient Temp: 20.4

CPU Over-Clocked:

• CPU Idle Temp: 29C @ 4200MHz
• CPU Max Temp: 62C @ 4200MHz
• Volt +/-: -210 mv Core
• Fan: 4200 RPM

Here are the automatic fan results:

Ambient Temp: 20.4

CPU Over-Clocked:

• CPU Idle Temp: 28C @ 4200MHz
• CPU Max Temp: 77C @ 4200MHz
• Volt +/-: -210 mv Core
• Fan: 2300 RPM

On the automatic fan run, I barely heard the fan which was great! I will do the GPU when I have time and people are interested.