Page 8 of 17 FirstFirst ... 6 7 8 9 10 ... LastLast
Results 71 to 80 of 166
  1. #71
    New ROGer Array
    Join Date
    Aug 2018
    Reputation
    10
    Posts
    4

    Quote Originally Posted by MyRyzenWorld View Post
    Funny thing though, if I don't drop in MPT gpu tdc and Power Limit to 100-115w zone,
    And keep it stock 130w, and in crate I keep platform sppt around 170-190w, then in firestrike I see 103c peaks gpu Hotspot and it sits around 95-99c fluctuations.

    Now what I figure today, with mpt limit 100w on gpu for TDC and Power Limit and crate set to 180w platform and 80w apu, my firestrike hotspot peaks at 91c and BF2042 (game) peaks 94c ... so game peaks hotspot gpu above 3dmark firestrike? How come?
    well running the apu that high just dumps more heat into the SHARED vapor chamber which leads to higher gpu temps. drop your apu power... to 30-40w for 2042. even at 80w setting it kind of tops out around 50w in 2042 anyways after it settles. i have been messing around with 2042 today at various sppt settings in crate. the game is not cpu bound so letting the cpu run full tilt is just dumping excess heat for very little gain if any. i understand that we bought a high performance laptop and not running it at full power seems weird but you cannot complain about very high temperatures when you run the darn thing at near full power. play with the sppt get both the apu and platform as low as you can until you actually see a fps drop. SOME games do benefit from the extra cpu power but many do not. The games that greatly benefit generally are the ones that hit the cpu for near 100% like csgo or older games like gta 5. Spend some time tuning the settings for each game you play you might be surprised how low you can go on power and still not lose performance in many games or if you do its a good compromise.

    now i forgot to say i slightly modded my bottom panel to allow less restricted airflow. not wide open just doubled the vent openings over the cpu and gpu fans. which drastically changed my temperatures at a given fan speed. cpu and gpu clock higher at lower power settings since they run cooler. i run my fans 50% max so that should tell you how restrictive that bottom panel is... I dont recommend it as bye bye warranty and results in a warmer keyboard deck and slightly warmer ssd. so my machine performance/temperatures cant compare to anyone elses so my comparison is kind of useless lol. ill upload a pic of my bottom panel if you request it.
    Last edited by cstow88; 11-13-2021 at 01:45 AM.

  2. #72
    New ROGer Array
    Join Date
    Sep 2021
    Reputation
    40
    Posts
    87

    Quote Originally Posted by cstow88 View Post
    well running the apu that high just dumps more heat into the SHARED vapor chamber which leads to higher gpu temps. drop your apu power... to 30-40w for 2042. even at 80w setting it kind of tops out around 50w in 2042 anyways after it settles. i have been messing around with 2042 today at various sppt settings in crate. the game is not cpu bound so letting the cpu run full tilt is just dumping excess heat for very little gain if any. i understand that we bought a high performance laptop and not running it at full power seems weird but you cannot complain about very high temperatures when you run the darn thing at near full power. play with the sppt get both the apu and platform as low as you can until you actually see a fps drop. SOME games do benefit from the extra cpu power but many do not. The games that greatly benefit generally are the ones that hit the cpu for near 100% like csgo or older games like gta 5. Spend some time tuning the settings for each game you play you might be surprised how low you can go on power and still not lose performance in many games or if you do its a good compromise.

    now i forgot to say i slightly modded my bottom panel to allow less restricted airflow. not wide open just doubled the vent openings over the cpu and gpu fans. which drastically changed my temperatures at a given fan speed. cpu and gpu clock higher at lower power settings since they run cooler. i run my fans 50% max so that should tell you how restrictive that bottom panel is... I dont recommend it as bye bye warranty and results in a warmer keyboard deck and slightly warmer ssd. so my machine performance/temperatures cant compare to anyone elses so my comparison is kind of useless lol. ill upload a pic of my bottom panel if you request it.

    Thanks for all the info, mind sharing photos of your modding? My unit anyways not under warranty, as I ran repasting and also done SPi to update vbios

  3. #73
    New ROGer Array
    Join Date
    Sep 2021
    Reputation
    40
    Posts
    87

    Quote Originally Posted by cstow88 View Post
    the hotspot temperature reading is not a bug. If you look at the total system draw for when its hitting 100c hotspot in benchmarks you will notice its only hitting that in the short boost burst then it falls to well below 100c for the rest of the test. If you have any experience with desktop amd gpus you will also notice the hotspot being 20-30c higher than the core thats just how it is.. The gpu in this thing is a desktop 6700xt and the fact it can even hit max 2.5ghz at sustained load in benchmarks while just touching 100c on the hotspot is incredible to be honest. the desktop part also tends to sit at 100c hotspot for the cards without a 4 slot cooler. the hotspot temp is in spec on this laptop nothing to worry about just dont run benchmarks 24/7 then you wont ever see 100+c hotspot unless running 190w platform sppt.... adjust sppt settings for gaming and it will be sitting mid 80s-low 90s. most seem to forget benchmarks doesnt equal real world use or in the case gaming temperatures.
    Btw yourr reporter Temps and tdps...what fan curve speeds ?

  4. #74
    New ROGer Array
    Join Date
    Nov 2021
    Reputation
    10
    Posts
    1

    Hello everyone, new to this forum and my first post so apologies in advance for any mistakes. I'm also primarily a desktop user so not much experience with laptops. I've been looking for potential solutions to the random reboot problem though found nothing that exactly matches my circumstances as yet. Therefore, I'd like to share my experience with the random reboots and my solution, whilst also seeing if any other forum members can replicate it.


    NOTE: please read my full post before jumping down my throat saying my solutions are useless as my issues seem to be a bit different, albeit very similar, to those posted in this thread and those reported in wider social media in general. Hence, my workaround may not work for some people.


    Important information:
    • My VBIOS is XXXXX77321 and my driver version is 21.10.2 as I have found it is the most stable, though as others have posted in this thread, YMMV.
    • The latest RX6800M driver currently on the ASUS website (for now it is the one from October) has been rather problem-ridden in my case, constantly corrupting itself after successive reboots (normal and otherwise). I did not realise at first, but after several reboots, I suffered a lot of FPS drops and general instability whilst the performance metrics overlay for GPU2 (the RX6800M) would stop working or just had missing or erroneous values.
    • I have since upgraded my RAM to 2x8gb 3200 CL20 kit, running at 3000 MHz CL17 using an alternate XMP profile. The memory is single-rank x8 (1Rx8). I also installed 2x 1TB WD Black SN750 SSDs, on one of which I installed a clean copy of Windows 10 Pro (did not use the recovery partition on the original drive) only manually reinstalling all of ASUS's recommended drivers and Armoury Crate.


    To the core issue:
    For me, I quickly realised that the random reboots only occurred: 1. after closing a game specifically, not just any high 3D-load applications, even with benchmarks I was initially unable to reproduce the reboot; and 2. after a few minutes of using my browser, specifically, any chromium based browser with HW acceleration enabled, I have tested Vivaldi, Brave and, of course, Chrome (all latest versions as at 15/11/21) but was unable to reproduce the reboots on Firefox. The issue was that the AMD software (at least for me) recognises Chromium browsers when hardware acceleration is enabled as a game and runs them on the dGPU not the Vega iGPU. Disabling HW acceleration on Chromium-based browsers stopped the reboot issue entirely for me.

    When reproducing the reboots using those browsers (again with HW acceleration checked) I notice large power spikes in CPU and GPU power consumption, whereupon the system locks up, and then it reboots while Windows event viewer states a code 41 error. It is important to note that, so long as HW acceleration checked on these browsers, even when running in the background (i.e. you have the setting to allow them to run in the background after closing enabled) I could reproduce those reboots. With this in mind, I managed to reproduce the issue outside of games, using Stata MP and Autodesk running dummy workloads simultaneously (to stress CPU and GPU). After they finished, I opened my Chromium-based browser with HW acceleration enabled and after a few minutes, promptly had a reboot. This was how I was able to narrow down the cause to Chromium based browsers as I use FF while I'm working and Brave while I'm gaming, hence, why I was unable to initially replicate the reboot.

    Can someone else reading here with a G15 AE please attempt to reproduce the reboots with a HW accelerated chromium browser and my solution? Please report back on your findings.

    Given that my problems have since stopped, and because I have had similar issues on my Vega 64, Radeon VII and 5700XT (only at launch with the 5700XT, tested 4 different PSUs and 3 different MoBo chipsets and platforms) I am inclined to blame AMD's drivers, though I may be wrong. In the case of the desktop GPUs, it wasn't a total system crash but a black screen that did need a manual reboot to rectify. So perhaps its the fault of the driver and vBIOS? But again, my problems seem rather unique compared to what everyone else has been posting so I don't know. I'd also be curious to know of anyone with the desktop versions of Big Navi have similar issues (black screens that need manual reboots to fix or rebooting issues in general). My desktop has a 6800 XT and I did experience some random reboot-inducing black screens, but only for a few updates after launch.

    Finally, can someone who has not had any reboot issues post your vBIOS version and RX6800M driver version please? My son's G15 AE has had zero reboot issues (even with HW-acceleration enabled on his browser, he uses Chrome), or any of my issues with corrupting drivers and he is running XXXX77321 and has had no issues to date with the out-of-box drivers and 21.10.2.

    To sum up:
    • After disabling HW acceleration on my Chromium-based browser, the reboot issue stopped entirely for me. Again, this solution may not be applicable for everyone, read the above post in its entirety. I have not had the random reboots for 6-weeks and counting as at 17/11/2021.
    • Furthermore, updating to the latest (recommended or stable) driver from AMD (at the time of posting is 21.10.2), any general system instabilities went away. However, YMMV depending on your vBIOS version as explored earlier in the thread.

    However, given that the reboot issues I have seen posted in wider social media and this thread are somewhat different to mine, it may be another issue entirely and my fixes may not work at all.

    Disparities in Temps and Performance:
    I only adjusted my fan curves and the platform sPPT in Armoury Crate (APU sPPT at 100w but platform sPPT at 175w).
    • After gaming for 2 hours at my jobsite dormitory with the A/C off ( ambient temps of ~35C, tested TW:WHII, TW:3K, Cyberpunk 2077, Doom (2016), RDR2 and War Thunder at 1440P). Provided HWinfo is accurate, my CPU package peaked at at 92.2C whilst my GPU edge runs at 79.5C with a hotspot of 94C and junction of 81.1C as per HWinfo.
    • At my home which is in a much colder area (ambient temps of ~1-15C earlier in the year), using the default Turbo profile my CPU would be sub-90C while gaming. (I will report on more detailed temps once I get back in a week or so, though it's summer here in the South so temps won't be that much different).

    With the 21.10.2 driver the RX6800M was able to maintain a SCLK of between ~2388-2466 MHz and power consumption between 135-145w, as per the AMD performance metric software and HWinfo (though this did go down to as low as 115w at times in War Thunder where I FPS limited to 100 for stability). With the default Turbo profile and the platform sPPT at 190w manually, I saw the GPU spike as high as 155w but for no more than 20 seconds or so, this is where I saw the highest hotspot temps in HWinfo, peaking at 97.1C in CP2077, whilst the same occurred during a TimeSpy run with 190W manually set. These tests were done in a very hot environment also (~33C ambient at the time).

    EDIT: adding in more tangible performance numbers. With the new RAM I am getting 12,003 on 3DMark TimeSpy and 28,488 on FireStrike, did not benchmark prior to upgrading unfortunately and tests were conducted without an external display. CB R23 returned a score of 14,661 on multicore. Game performance is a little redundant to quote as people will use different settings. This was with the aforementioned manually set power limits: APU sPPT at 100w and Platform sPPT at 175w.

    EDIT 2: I would have used RTSS for my performance metrics logging, however, AB does not support the RX6800M, only nVidia mobile GPUs and then only unofficially.
    Last edited by Khabaopie; 11-17-2021 at 05:35 PM.

  5. #75
    Moderator Array
    Join Date
    Feb 2019
    Reputation
    15
    Posts
    920

    Dear all,
    We have some cases now under RMA.
    It is expected that replacing the new motherboard is going to fix it.
    They are being carefully monitored.
    Thank you.
    Last edited by Blake@ROG; 11-18-2021 at 02:38 AM.

  6. #76
    New ROGer Array
    Join Date
    Sep 2021
    Reputation
    40
    Posts
    87

    Quote Originally Posted by Blake@ROG View Post
    Dear all,
    We have some cases now under RMA.
    It is expected that replacing the new motherboard is going to fix it.
    They are being carefully monitored.
    Thank you.
    Blake,

    As people reported that even with replaced motherboard it still reboots, before you start calling all the owners for RMA, please check with the bios and software team, to confirm that these issue can't be fixed via bios and firmware as it seems as some sort of power mis management issue, might be smart shift configuration issue or something around it.

    Crash happens not during game or during 3d render but minute or two after you close the game or the render, as switching from performance to idle fails somewhere.

    Thank you.

    IMPORTANT UPDATE:
    Those who used MPT tweak, if you changed both Power Limit and TDC, you may attempt to change TDC back to 136W and keep only Power Limit at the lower side (110W as example). We could observe that TDC limit not properly supported by the drivers/vBios as results, some games would crash/reboot at start up (as happened to @timthebull). Changing only Power Limit kept the reboot fix in tact, even if TDC set to default 136W (as we confirmed with @timthebull and @krizzy090).


    IMPORTANT UPDATE #2:
    Per Discord users that test and trial workarounds, MPT/Power changes may not be needed to temporary stop the reboots, see page 9/10 of this forum thread for more information about the new workaround that community members found using GPU-Z and ASUS GPU Tweaker 3.

    IMPORTANT DISCLAIMER: ANYTING, EVERYTHING WE (COMMUNITY) TEST AND TRIAL SHOULD NOT BE CONSIDERED AS A FINAL/OFFICIAL SOLUTION, AND MAY NOT WORK FOR EVERYONE. THEREFORE WE SHARE ALL WE FIND WITH ASUS, SO ASUS/AMD COULD WORK ON AN OFFICIAL SOLUTION/FIX BASED ON WHAT WE TEST AND TRIAL. ALL OUR CLAIMS BASED ON OUR PERSONAL TESTING AND THOUGHTS/OPINIONS, NOT ENGINEERING FACTS, AS WE DID NOT DEVELOP THE HARDWARE OR THE SOFTWARE AND WE CANNOT BE SURE WHAT HAPPENING INSIDE.
    Last edited by MyRyzenWorld; 11-27-2021 at 04:38 PM.

  7. #77
    Moderator Array
    Join Date
    Feb 2019
    Reputation
    15
    Posts
    920

    Quote Originally Posted by MyRyzenWorld View Post
    Blake,

    As people reported that even with replaced motherboard it still reboots, before you start calling all the owners for RMA, please check with the bios and software team, to confirm that these issue can't be fixed via bios and firmware as it seems as some sort of power mis management issue, might be smart shift configuration issue or something around it.

    Crash happens not during game or during 3d render but minute or two after you close the game or the render, as switching from performance to idle fails somewhere.

    Thank you.
    Hi there,
    those customer go to RMA themselves.
    I really appreciate them since they provide the laptop to us to physically check on it.
    And we are doing new things.
    If any other customer are in urgent need, they can always contact me and we are arranging proper service for them.

    I personally thank you for providing information and information sharing would be enough for you.
    I would not interfere how you are going to discuss online but that does mean I have to follow everything you said.
    Please do understand. Thank you very much.
    Last edited by Blake@ROG; 11-19-2021 at 03:43 AM.

  8. #78
    New ROGer Array
    Join Date
    Sep 2021
    Reputation
    40
    Posts
    87

    Quote Originally Posted by Blake@ROG View Post
    Hi there,
    those customer go to RMA themselves.
    I really appreciate them since they provide the laptop to us to physically check on it.
    And we are doing new things.
    If any other customer are in urgent need, they can always contact me and we are arranging proper service for them.

    I personally thank you for providing information and information sharing would be enough for you.
    I would not interfere how you are going to discuss online but that does mean I have to follow everything you said.
    Please do understand. Thank you very much.

    Thats fine, your company, your ways, I wanted help you guys help the users, but final call is yours.

    Good Luck,

    Thank you.
    Last edited by MyRyzenWorld; 11-19-2021 at 08:42 AM.

  9. #79
    New ROGer Array
    Join Date
    Nov 2021
    Reputation
    10
    Posts
    3

    Quote Originally Posted by Blake@ROG View Post
    Hi there,
    those customer go to RMA themselves.
    I really appreciate them since they provide the laptop to us to physically check on it.
    And we are doing new things.
    If any other customer are in urgent need, they can always contact me and we are arranging proper service for them.

    I personally thank you for providing information and information sharing would be enough for you.
    I would not interfere how you are going to discuss online but that does mean I have to follow everything you said.
    Please do understand. Thank you very much.
    Hi, Blake.

    Arleady sended(i wait 2 weeks) and wrote about it here. It's not helped problem just stay, rma-service say's "We replaced system of cooling", that's all. What I can with this? I'm just desperate now....


    upd. Now, have day one stable with no reboot after blender or games.

    I do 2 step's for this :

    1. Update vbios at 079208 version.
    2. Lowered power limit at gpu from 130 to 110 watt, and rewrite SPPT to my graphics card.


    I would like to emphasize that I did all this with third-party programs.

    In stock my laptot - is unstable.
    Last edited by gooodorsage; 11-19-2021 at 01:12 PM.

  10. #80
    New ROGer Array
    Join Date
    Sep 2021
    Reputation
    10
    Posts
    2

    Quote Originally Posted by Blake@ROG View Post
    Hi there,
    those customer go to RMA themselves.
    I really appreciate them since they provide the laptop to us to physically check on it.
    And we are doing new things.
    If any other customer are in urgent need, they can always contact me and we are arranging proper service for them.

    I personally thank you for providing information and information sharing would be enough for you.
    I would not interfere how you are going to discuss online but that does mean I have to follow everything you said.
    Please do understand. Thank you very much.
    Is this fix official ? Will there be official announcement or notice globally about this ? Because in my country, the authorized service center do not have a clue how to fix it. They inspect the product and does not jave any clue what is broken. They also did not offer any solution, refund or device replacement.

    It is hard to work with this kind of laptop.
    Last edited by ishengard; 11-19-2021 at 11:37 AM.

Page 8 of 17 FirstFirst ... 6 7 8 9 10 ... LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •