*** The Official MSI GS65 Stealth Owners and Discussions Lounge ***

Discussion in 'MSI Reviews & Owners' Lounges' started by Skylake_, Apr 3, 2018.

  1. sk3tch

    sk3tch Notebook Deity

    Reputations:
    97
    Messages:
    821
    Likes Received:
    177
    Trophy Points:
    56
    The system comes from you guys configured with "Auto" fans at default. I was playing ARK: Survival Evolved, not extreme stress testing - normal usage on a gaming laptop is to play games?

    Anyway, results are better using AIDA64, which yes, is a stress test but I do not have time to play ARK right now. I just wish you guys outlined which fan profile you were using to get your thermal numbers. Seems to be a fair thing to represent/disclose. I have outlined I use "Auto" all along (in addition to other information).

    [​IMG]
     
  2. Donald@HIDevolution

    Donald@HIDevolution Company Representative

    Reputations:
    11,229
    Messages:
    8,932
    Likes Received:
    4,661
    Trophy Points:
    431
    Yes, it is delivered with Auto Fans, but the fan settings are user controlled. I already stated when we stress test, which includes gaming and benchmarks as will as AIDA64, we always set the fans to max, which any user should do as well. Auto Fans will not react to temp changes fast enough when pushing it with full load. Anytime you feel the temps are too high, kick the fans into gear at max.

    As I said before, this is common knowledge, and at a minimum, common sense. Temps too high, but fans are not set to max, well, that is why those settings are there....use them.
     
    raz8020, Papusan and JeanLegi like this.
  3. sk3tch

    sk3tch Notebook Deity

    Reputations:
    97
    Messages:
    821
    Likes Received:
    177
    Trophy Points:
    56
    Good to know. I never had to game at max fans on my Razer Blade v5 and that wasn’t repasted. I’ll make sure to point out that your numbers are “wind tunnel” edition noise profile so potential buyers know - since you guys don’t seem to want to state it. :)
     
    hmscott likes this.
  4. Mr. Fox

    Mr. Fox Undefiled BGA-Hating Elitist

    Reputations:
    23,727
    Messages:
    32,538
    Likes Received:
    46,439
    Trophy Points:
    931
    Only a moron would have anything other than nice things to say about you, or your posts, Brother @Falkentyne, so don't worry about any losers that don't like your posts. I think voodoo is a bad thing to be involved with, but I do believe in miracles.
     
  5. Donald@HIDevolution

    Donald@HIDevolution Company Representative

    Reputations:
    11,229
    Messages:
    8,932
    Likes Received:
    4,661
    Trophy Points:
    431
    Come on now...I have already stated so multiple times. This is not some hidden secret the we "...don't seem to want to state it."

    Why would anyone doing max stress testing run their fans at anything but max? We use Ultimate settings for all gaming tests as well. No cheating, no shortcuts, just full out max testing.

    I understand this may be new information for you, but try being fair and take the information in, without trying to suggest we are not reporting accurately.

    Still waiting to see your results when tested properly with max fans.
     
  6. bloodhawk

    bloodhawk Derailer of threads.

    Reputations:
    2,907
    Messages:
    5,846
    Likes Received:
    8,535
    Trophy Points:
    681
    It could be a lot of factors, specially ambient temperatures and how confined of a space the laptop is in. Is the back raised up high enough for good air flow etc etc.

    I would rule these out first and then start thinking about the paste job being dodgy.

    Also high end laptops, do need tweaking/fan profile changing on the user end. Because one profile/setup will never work for everyone. Every user has their own tolerance level.

    If someone wants a out of the box plug and play experience, better buy a mac book.

    The Razer Blade's have very non aggressive fan profiling because the fans sound obnoxiously high pitched so they rather have the laptop throttle than have people complain about noise.
    Most important part however is, 6 cores running at a higher clock speed vs 4 cores running at a low clock speed.
    I really fail to see the problem here?

    You really need to switch to max fans.

    Yeah happened with one of my 6850k's where 2 cores were running 15C hotter than the other 4. NO matter that i tried, the difference was always there, 2 replacements later, the third chip only had a differential of 2-3C at most.

    Even on the 8700k in my desktop, there is a core differential of about 2-4C.

    Seems to be pretty normal these days.
     
    Last edited: May 16, 2018
  7. Johnksss

    Johnksss Notebook Nobel Laureate

    Reputations:
    9,687
    Messages:
    18,238
    Likes Received:
    8,768
    Trophy Points:
    931
    Too late. He already compared it to a Razor. LOL
     
  8. Falkentyne

    Falkentyne Notebook Prophet

    Reputations:
    5,610
    Messages:
    4,728
    Likes Received:
    6,364
    Trophy Points:
    581
    Don't use AVX / FMA3 testing. This brings out absurd loads and can overinflate ANY tiny deficiency in a heatsink including parts cooling VRM's and the EC can even shut your taptop off too.

    Open LOCAL.txt in prime95 folder.
    Add:

    CPUSupportsAVX=0
    CPUSupportsFMA3=0

    at the top.
    If you want to disable the FMA3 power virus (FMA3 is AVX2, btw, fused multiply and add), you can set AVX to 1 and FMA3 to 0 but only do this for OCD testing purposes.
     
    Donald@HIDevolution likes this.
  9. Falkentyne

    Falkentyne Notebook Prophet

    Reputations:
    5,610
    Messages:
    4,728
    Likes Received:
    6,364
    Trophy Points:
    581
    I have super obsessive OCD so I've probably done more repastes of the same CPU than anyone who was not doing it as a paid job.
    I've seen the behavior of LM.
    The "issue" of LM deciding to have sex with itself happens if there is any pressure imbalance when pushing a heatsink down just to get the screws in the hole, even if you do a perfect tiny partial turn criss cross pattern perfectly. The LM will have already partially collected slightly more in another area, resulting in a temp imbalance.

    This can be circumvented by using a VERY THIN layer of LM, but doing this can be BAD on direct die to copper because of the partial absorption issue. I noticed this on my first ever attempt at using LM. Applied a razer thin layer. In one week there was none on the CPU =). Gotta use a little extra LM than 'necessary' and also vigorously paint the heatsink too. Getting a flat heatsink (instead of convex) and fully even complete downwards pressure=vomit orgasm of even core temps.

    I tried the same experiment on a 290X. Did super thin even layer, 1 week later...card crashed hard in PUBG. Reached 95C. Opened it...there was zero LM on the GPU :) Just a hardened faint film. It was all hardened on the copper surface. Had to apply more to allow contact over time. Of course the almost nonexistent mounting pressure of the anorexic X-bracket and cancer scrawny 1mm springs didnt help anything and totally inflated the problem to unrealistic levels only AMD's awesome design team was capable of.

    LGA under IHS would never have this problem. Solution: use more to compensate on BGA, and break out them foam dams and Mr Fox creative barriers.
    Obviously, much higher static pressure would solve all these problems. But you can only get that on LGA.
     
    Donald@HIDevolution and Mr. Fox like this.
  10. Telos

    Telos Newbie

    Reputations:
    0
    Messages:
    4
    Likes Received:
    4
    Trophy Points:
    6
    Can anyone share their fan settings on Dragon center? I have mine currently set like this https://imgur.com/8ZFzJAi Shared a link for some reason I can't get the insert image to work.
     
    Last edited: May 16, 2018
    hmscott likes this.
Loading...

Share This Page