*** Official Clevo PB5xExx(-G) Owner's Lounge ***

Discussion in 'Sager/Clevo Reviews & Owners' Lounges' started by sicily428, Jan 7, 2019.

  1. Meaker@Sager

    Meaker@Sager Company Representative

    Reputations:
    8,381
    Messages:
    52,852
    Likes Received:
    15,203
    Trophy Points:
    931
    Have you re-installed the ccc?
     
  2. TheUberMedic

    TheUberMedic Notebook Consultant

    Reputations:
    9
    Messages:
    250
    Likes Received:
    73
    Trophy Points:
    41
    FN + 3 should change the profiles (I think)

    As for the PL1 kicking in too early, I'm having the same issue. It seems to only affect benchmarking stuff. In other tasks (like games) it'll keep the boosts.
     
  3. accordion

    accordion Notebook Enthusiast

    Reputations:
    0
    Messages:
    43
    Likes Received:
    2
    Trophy Points:
    16
    Will try reinstalling CCC, the LED keyboard and fan control settings do work. FN+3 doesn't work for me, FN+1 does boost the fan.

    Have you tried disabling the PL via throttlestop? The TPL settings work once that box is checked. I was getting consistent 55W without throttling which produced a 2900+ R20 score. How far into Cinebench does throttling start for you currently? No matter how much I undervolt the timing is about the same. I'm also wondering if it's worth it to spend more time on this as the scores are already good. Will try playing some games to see.

    The two screws behind the keyboard are by the left and right sides and marked "M.2.". Otherwise taking it apart was straightforward. I didn't put the screws back and the keyboard feels fine. I didn't wanna risk it with a double sided stick so went with a 2.5" MX500. There seems to be more than enough clearance for one to fit. The stick on the left is a WD Blue.

    IMG_20190629_190438.jpg IMG_20190629_190556.jpg IMG_20190629_190731.jpg

    Edit:
    How bad is my backlight bleed?
    IMG_20190629_214825_1.jpg
     
    Last edited: Jun 29, 2019
  4. accordion

    accordion Notebook Enthusiast

    Reputations:
    0
    Messages:
    43
    Likes Received:
    2
    Trophy Points:
    16
  5. Ashlander

    Ashlander Notebook Consultant

    Reputations:
    14
    Messages:
    286
    Likes Received:
    128
    Trophy Points:
    56
    That light bleed looks pretty excessive. Mine is nowhere near that bad.
     
  6. accordion

    accordion Notebook Enthusiast

    Reputations:
    0
    Messages:
    43
    Likes Received:
    2
    Trophy Points:
    16
    I was able to get CCC to work but my benchmarks were no better than before. Even though Cinebench R15 didn't lead to throttling the extra power didn't result in higher clock speeds, same for R20. 3DMarks results were noticeably worse even on Performance setting although over a very small sample.

    Without CCC I was able to get 18205 Firestrike and 8242 Timespy which was sixth best for this GPU/CPU combo. GPU max temp was around 65C so I feel like more can be done, I'm just not sure how. There seems to be very little difference between flattening the curve at 900mV vs 800mV. The highest scores were with 850mV as shown below. The 600 Mhz memory overclock was picked because other posters mentioned the number earlier, haven't tested other values.

    It looks like Gentech did a great job with the LM and thermal pads. I have never seen the CPU go above 82C after almost two straight days of benchmark and stress tests. High 60s/low 70s are the norm during stress tests.

    Edit: Changed the temps because I didn't actually have HWInfo open. Also, it seems that my PL1 is 55W despite having uninstalled CCC because that was the last setting. Thus all of the benchmarks were with the higher PLs.

    explorer_MNfZKVOdmc.png explorer_HlvdtWLHIr.png explorer_1g2eUt9kdI.png Cinebench_3PB74Bbd3j.jpg SOTTR_Kca7qZC39n.jpg
     
    Last edited: Jun 30, 2019
    4W4K3 likes this.
  7. Meaker@Sager

    Meaker@Sager Company Representative

    Reputations:
    8,381
    Messages:
    52,852
    Likes Received:
    15,203
    Trophy Points:
    931
    It can be really hard to say with a photo as cameras will dramatically alter the contrast.
     
  8. ZaiArgylle

    ZaiArgylle Notebook Geek

    Reputations:
    2
    Messages:
    76
    Likes Received:
    22
    Trophy Points:
    16
    Went with a Liquid Metal repaste and my temps have dropped drastically (~20-10c less). Still, there's one particular core with a +6c difference from the others. I feel I did a good job controlling even spread while repasting, so could there be any other reason behind this differential?.

    Screenshot of said temps while running TS benchmark
     

    Attached Files:

  9. Meaker@Sager

    Meaker@Sager Company Representative

    Reputations:
    8,381
    Messages:
    52,852
    Likes Received:
    15,203
    Trophy Points:
    931
    It could be either the liquid metal contact there or the sensor or just a hotter running core or temperature sensor calibration difference in the die. Lots of reasons basically.
     
  10. JEAMN

    JEAMN Notebook Geek

    Reputations:
    5
    Messages:
    86
    Likes Received:
    30
    Trophy Points:
    26
    Dang! That's a nice Fire Strike score! I've managed a similar Time Spy (on a 17"), but my best Fire Strike is only around 17500.
     
Loading...

Share This Page