The Notebook Review forums were hosted by TechTarget, who shut down them down on January 31, 2022. This static read-only archive was pulled by NBR forum users between January 20 and January 31, 2022, in an effort to make sure that the valuable technical information that had been posted on the forums is preserved. For current discussions, many NBR forum users moved over to NotebookTalk.net after the shutdown.
Problems? See this thread at archive.org.

    Thermal Testing Results Database (17r4, 15r3, 13r3)

    Discussion in '2015+ Alienware 13 / 15 / 17' started by Snapple37, Sep 3, 2017.

  1. Snapple37

    Snapple37 Notebook Guru

    Reputations:
    9
    Messages:
    53
    Likes Received:
    18
    Trophy Points:
    16
    Hey guys (and gals, I'm sure you're out there),

    This is an attempt to collect thermal performance of our machines so that new and current owners can make informed decisions on whether to re-paste, re-balance, repair, or return our infamously overheating laptops. This database is only as good as the information we enter, and I do hope that as many people as possible will take the time to enter this information.

    Disclaimer: Stress testing a system with poor thermal performance can be detrimental to your system. If you do not have warranty, stop the test anytime you feel uncomfortable (i.e. 90c+ core temps).

    This test is designed to be simple and easily replicated by anyone. While it can push the system close to its thermal limits, a good re-paste and re-pad should allow any laptop to run them safely, especially while lifted or on a laptop cooler.

    BIOS Settings: Reset to default. Enable "Performance CPU" & "Performance Fan". NO OC. (resetting to default is important if you've been using XTU for tweaking).
    Software Settings: vCore offset -0.100 or whatever is stable with your CPU
    Power Plan: Windows default, Balanced (what most people use)
    Test: 10-15 minute Aida64 - CPU and GPU tests only. Do not check FPU, it gives unrealistic temps and is not representative of real life performance.
    Flat or Lifted: Flat on a hard surface desk/table. Lifted back at least 1", or on a laptop cooler (back can be lifted in the center by anything as long as it doesn't block the fans).

    *CPU Idle Temps comparison is highly dependent on ambient temps, but it could provide some valuable information so it is included as optional. It will take at least 5 minutes to show accurate idle temps.
    *Performance Fan in BIOS is required so that laptop doesn't have to ramp up at the beginning of the test. Most people will not leave this on during everyday use.

    https://docs.google.com/spreadsheets/d/1rsyeiZb_Z5ZNnK6ZW5JoH-m99fqaGQH0w1vfKgOP3ww/edit?usp=sharing
     
    Vasudev, ShotOfB12 and 0lok like this.
  2. Snapple37

    Snapple37 Notebook Guru

    Reputations:
    9
    Messages:
    53
    Likes Received:
    18
    Trophy Points:
    16
    Here are my results, details in the spreadsheet:

    Flat - CPU - 90/88/88/86, GPU - 82
    Lifted - CPU - 74/72/77/74, GPU - 68

    This again illustrates the massive difference in temps when more airflow is allowed, suggesting that Dell should really consider a flip out stand underneath the laptop (like the rear spoiler on a Porche!).

    Edit:

    Here's what this system is capable of, now that my thermal limits have been established. Overclocked all cores to 4.3ghz stable at 0.000 vCore offset. Overclocked GPU to +210/+370.

    On a back to back Firestrike + Timespy run, max temps reached 85/79/84/79 on CPU, 64 on GPU (on laptop cooler). Scores were 17,888 and 7111, respectively. http://www.3dmark.com/fs/13534957

    If you've seen my first posts on this board, my unit was one of the worst out of the box. It crashed running Prey at stock after 15 minutes and reaching 96c+ on cores. It throttled the base clock down to 2.6ghz when running CPU only stress tests, and could not handle any CPU and GPU tests simultaneously.
     
    Last edited: Sep 3, 2017
    ShotOfB12 and 0lok like this.
  3. ShotOfB12

    ShotOfB12 Notebook Consultant

    Reputations:
    16
    Messages:
    159
    Likes Received:
    89
    Trophy Points:
    41
    1" lift
    85/85/84/84 GPU - 74

    Flat
    90/90/92/92 GPU - 82

    Ambient 32

    Should of probably tested at night instead of the hottest time of day when I don't even game but in this case I guess its better. I didn't notice until after the test but my core clocks were x39/x37/x37/x37. I don't think that 3.9ghz core really makes a difference though.
     
    Vasudev and Snapple37 like this.
  4. Snapple37

    Snapple37 Notebook Guru

    Reputations:
    9
    Messages:
    53
    Likes Received:
    18
    Trophy Points:
    16
    I think those are great temps for running them at almost 10c higher ambient temps and higher clocks!

    Although... I don't know how people live without AC, I'm a bit spoiled.
     
    Vasudev and ShotOfB12 like this.
  5. ShotOfB12

    ShotOfB12 Notebook Consultant

    Reputations:
    16
    Messages:
    159
    Likes Received:
    89
    Trophy Points:
    41
    Living in Hawaii can be a *****. I'm down to my underwear praying for a nice breeze to come through. I would love AC but none of my windows are compatible unless heavily modded and I tried a few portable ones that just didn't work well. Plus my electricity bill is already high so it's better to make due I guess.
     
    Vasudev likes this.