Legion Pro 5 16IAX10H - Sudden Freeze > Restart

[ivory seal]

Model: Legion Pro 5 16IAX10H / Legion Pro 5i

Describe the issue:

I just went through the web and suddenly the screen froze. It was a 3d model in web. Then the PC restarted two times and showing me a windows update? I am pretty sure that no update was planned around that time. I think it tried to initialize a update/fix a faulty driver.

I am running dGPU mode and also I noticed that the Nvidia symbols are not in the tray. However, I am not sure if they were there before. I noticed though, that LegionSpace still needed an update.

What have you tried already?:

I checked event logger so far.
I would wait a little bit more before I DDU because I am not sure if LegionSpace might have failed because it wasnt updated yet? Like a issue between iGPU and dGPU.

I would however plan to DDU intel and nvidia driver and install both from the site again if it happens again. But maybe someone had the same issue and knows the real root.

When did it start producing the issue:

Just today for the first time when opening the browser(firefox). Gaming worked really well so far the past weeks.

Did you install new drivers or software?:

When I got the laptop before christmas holiday, I tried to update the nvidia driver(591.59). I DDUd and went back to the Lenovo Nvidia VGA driver (577.05). I got the gray screen issue with the updated nvidia driver and thus decided to go back. But nothing else so far.

Can you provide images/videos to show us the issue:

Pictures are in german, sorry :( But I copied the content so one can translate it.

Picture 1+2 are related to point 1 and 2.

1. nvlddmkm say all the same thing at the time of the restart:
   "Die Beschreibung für die Ereignis-ID "153" aus der Quelle "nvlddmkm" wurde nicht gefunden. Entweder ist die Komponente, die dieses Ereignis auslöst, nicht auf dem lokalen Computer installiert, oder die Installation ist beschädigt. Sie können die Komponente auf dem lokalen Computer installieren oder reparieren.

Falls das Ereignis auf einem anderen Computer aufgetreten ist, mussten die Anzeigeinformationen mit dem Ereignis gespeichert werden.

Die folgenden Informationen wurden mit dem Ereignis gespeichert:

\Device\Video16
GpuRcReset TDR occurred on GPUID:200

Die Nachrichtenressource ist vorhanden, die Nachricht wurde in der Nachrichtentabelle jedoch nicht gefunden"

=====

2. The restart has a critical tag and had the following standard text:
   Das System wurde neu gestartet, ohne dass es zuvor ordnungsgemäß heruntergefahren wurde. Dieser Fehler kann auftreten, wenn das System nicht mehr reagiert hat oder abgestürzt ist oder die Stromzufuhr unerwartet unterbrochen wurde.

====

3. The next restart consists of these events:
   -"Überwachungsereignisse wurden vom Transport gelöscht. 0"

-"Fehler beim Laden des Treibers \Driver\WUDFRd.
Gerät: PCI\VEN_8086&DEV_AD03&SUBSYS_3D7317AA&REV_01\3&11583659&1&20
Status: 0xC0000365"

-"Fehler beim Laden des Treibers \Driver\WUDFRd.
Gerät: ROOT\SYSTEM\0001
Status: 0xC0000365"

-"Der Computer wurde nach einem schwerwiegenden Fehler neu gestartet. Der Fehlercode war: 0x00000116 (0xffffd90b97f653d0, 0xfffff805a410ae80, 0xffffffffc000009a, 0x0000000000000004). Ein volles Abbild wurde gespeichert in: C:\Windows\Minidump\010926-19703-01.dmp. Berichts-ID: 36e964b4-a80c-40f6-a608-d426b689dc24."

Just saw that there is also WATCHDOG dmp

Thanks in advance !!

Legion Pro 5 16IAX10H - Sudden Freeze > Restart

[rigid igloo]

I experienced this on 16IRX8h too. try to swap profiles between hybrid and discrete mode with restart, and stay on discrete mode. I think it is a windows problem.

[ivory seal]

Discrete mode = dGPU mode right?

Sounds good, I will try that later. Thank you! :)

But I am also very sure that it is some kind of driver/software issue

[true onyx]

Have you been able to solve this issue? I'm having the same problem

[bright juniper]

Yeah I think it's pretty common - I got it also, brand new laptop, error came during office work - no big workloads, BSOD, freeze...I did DDU and Lenovo drivers, now I'm in dGPU mode only, seems stable for couple of days.

[ivory seal]

I am currently abroad so I cant really try it until March :/

But I will probably ddu once more too but switch to hybrid before as Bungo did
Also install the nvidia AND intel driver from lenovo. Not only the nvidia one

[true onyx]

No problem! When you get back I suggest trying to disable hibernation with the command powercfg -h off, actually seems to have fixed it for me but i'm still testing

[bright juniper]

Well I got another crash today, two days without however...nlvddmkm, I'm running out of ideas and patience, it's quite frustrating...

From my point of view is obviously NVIDIA driver causing problems, and it's so random I can't tell when to expect it...

I saw the nlvddmkm error yesterday in event log also, but no crash.

[ivory seal]

The only other thing I could imagine is hardware acceleration in browser/discord etc. Thats also GPU related.

Because so far I only got it on firefox and discord, which make use of it by default

[ivory seal]

@true onyx I am pretty sure I turned hibernation off as soon as I got the laptop out of the box. Same with quick startup.

@bright juniper do you have hardware accel activated?
Office work = ms office programs? Idk if they make us of that though then. This is the next thing I would try then.

[bright juniper]

This I haven't checked yet, I will...I'm stable now, for two days...driver same 577.05 lenovo, discrete dGPU mode, and performance in LLT - also in windows power modes, best performance. I did also benchmarks, stress tests - everything was OK. Fingers crossed. 😄

I recommend LLT - lenovo legion toolkit, not the lenovo apps, it's really nice and useful.

[ivory seal]

Yes, keep me updated 🙂
So far it only happened on application where that feature is used as far as I know

-3D model in website
-I had a livemap opened for a game and made the map zoom more out by changing varibales(a little too much xD)
-Discord Screenshare

[bright juniper]

Okay note to myself and keep you posted. 🙂 Two days stable, no problem - regular office, work use. Today I had another crash - ddnvlmkm, 13, 153 and 14...i noticed it happened when i plugged in AC adaptor, last crash was the same...maybe this is the cause?

[bright juniper]

I was in outlook writing email, otherwise office work.

[ivory seal]

Office stuff doesnt use hardware accel right? At least I cant find anything in Outlook and Teams.

Next thing would be display settings > graphics and checking the GPU preference of some apps. Maybe its prefering the iGPU even though we are at dGPU mode, which might cause issues?

[bright juniper]

Aha I checked, yeah OK some apps have preference for integrated (ms store for example), it's "let windows decide"...haven't changed anything yet. Another option caught my eye - HAGS - is ON. Maybe I will try this first...

[true onyx]

I actually turned off hibernation, turned off PCIe Link State Power Management, and the crashes happened again, I don't know what else to try

It was stable for some days but now it isn't

[strange siren]

I also had this same issue

Again, stable for some days then randomly crashed with same error

Hibernation, link state, fast startup, etc are all disabled

Could it be windows HAGS or game mode somehow?

Definitely seems like a widespread specific issue

[pure elm]

Going through the same issue right now, The freezes are just too random. In my case the screen just goes black with just the backlit on, the keyboard freezes, and none of the peripherals work. Tried everything from DDU to Soft EC Reset but to no avail.
This does look like a widespread thing if I'm not wrong because I've never faced this before.

would like to see people throw some more light on this.

[bright juniper]

I think it's a widespread problem, but not affecting everybody, this I have concluded from internet research. I noticed in my case it can also happen after plugging in ac or hdmi external monitor. In my case it also happens every time at minimal workload - office programs...i have run benchmarks, tests, diagnostics - everything seems fine. 3 days now stable, i keep monitoring everything.

[pure elm]

Just let me know if you find any leads on this. This thing is atrocious

[dusty crag]

i think i solved it. Please try the following.

Choose a stable driver for your gpu. i installed 591.74. no issues other than HDR acting a bit weird, but not a huge issue to impact HDR performance.

next set up a undervolt for your GPU and straighten the curve at around 800 to 950mv. I did mine at 837mV. mine is 3080. if yours is 5070, then maybe you need higher voltage, i am not sure about that. so choose a stable voltage to flatten the curve

Disable Hardware GPU scaling in windows settings and restart laptop

also make sure you dont alt tab when in game. Like set full screen option oder borderless its fine. just dont alt tab.

Then play the game for an hour and see if it crashes. I did mine for more than an hour and not a single crash. usually it crashed every 10 mins or so,

this has been my error

[pure elm]

Alright, an update from my side as well - It's been a day and half since I've not encountered any crashes so far.
Things I did were -

1. Ran a complete windows update (that included updates to my windows version, bios, lenovo system, and some other stuff)

2. DDU'd both my AMD and NVIDIA GPU drivers with the OEM drivers from the legion support website (546.30 for the NVIDIA one, which is pretty old I'd agree).

3. One common pattern I noticed while I was facing the crashes before was that everytime it crashed, I was browsing using Chrome.
   After all the changes I've done, one thing I noticed again is Chrome once crashed on me but this time only the app crashed and not the whole system. Could be a widespread chrome thing I suspect.

[dusty crag]

i also played a lot without alt tabbing. Seems to me that alt tabbing was the issue? cos i always play with HDR on. Maybe thats what cause the gpu to lose signal and stop the game. Now no issues at all.

[weak oxide]

I have been getting nvlddmkm crashes every 5-10 minutes while playing games too. Are you facing a similar issue here?

I've tried everything, from fresh installs to DDU, nothing seems to work

[weak oxide]

@bright juniper @pure elm

[bright juniper]

Good morning. In my case crash never happens under heavy load - gaming, stress tests, benchmarks,...so I'm not sure conditions are similar to yours. Mine always happens randomly in some office work...i use my laptop as job laptop also. Currently i'm stable for over five days, i'm still monitoring everything...

[pure elm]

I haven't gotten time to play any game to check that. I'll let you know.

[pure elm]

ran quite a few Cinebench Throttle tests (10 mins each) and didn't notice any crashes so far. I think the issue should be fixed.

[weak oxide]

same here, but games continue to crash

[weak oxide]

Someone help me fix this error please

I'm at my wits end

[dusty crag]

from my observation, i can say that if your crashes happen only during loads, then might be a thermal limit issue or a frequency limit issue. I would check the temps of your GPU if the error is event ID 153. if it is anything above 85c, then you might need a repaste. Or if you are using msi afterburner, check if the voltage and clock you set it too low, mine crashed as a result of that. once i increased the voltage a bit, the crash didnt happen.

[weak oxide]

Yeah I'm planning to get a repaste and hopefully it works out

[dusty crag]

i would also suggest to get phase change pad like the ptm, unless you are an expert in pc repaste.

that will eliminate a lot of headaches. saying this from my own experience 😅

[bright juniper]

Yeah I second this, haven't used pads yet, but paste can be messy. 😄

[weak oxide]

What exactly are those? Also, im gonna rely on a service professional for the repaste and not do it myself lol

[dusty crag]

https://www.amazon.de/dp/B0DB84CQW6?ref=ppx_yo2ov_dt_b_fed_asin_title

this one

its a pad instead of a paste, so you just need to place it on the chip and close the cooler. it will auto change to liquid phase when the chip heats up

unlike normal paste that generally pumps out when external heat is applied

but its not like pastes are bad. For desktop, paste might be better, for laptop due to die size, a pad makes more sense

of course, only if you can afford it

[weak oxide]

my issue is that my games sometimes take like 15-20 mins to crash and then other times its just within a minute. would a paste/pad really help me here or is my gpu just gone

[dusty crag]

check the temps of the gpu first. Like use hwinfo and maybe send a screenshot, then i can easily say

[weak oxide]

like how exactly? should I send a screenshot of the temps shown in hwinfo (never used it before) as my games crash? cuz my entire system freezes when that happens

[dusty crag]

like this

[weak oxide]

which of these am i supposed to be looking at 😭

[dusty crag]

close it completely and checkmark only on sensors

thats the general info

[weak oxide]

gotcha, so should I take a screenshot as soon as a crash happens, correct?

[dusty crag]

correct

i suspect a spike during crash, cos ampere acts that way

[weak oxide]

just after a crash

[dusty crag]

open your gpu performance limiters

also shows some pcie express error counters

107 is insane

[weak oxide]

what does that indicate

[weak oxide]

@dusty crag took way longer to crash this time (not complaining, around 30 ish minutes) and all these screenshots are from 2-3 seconds apart after the crash

and this is after the game completely crashed

game tested was expedition 33

[bright juniper]

Another update from my side, was stable now for almost a week, now another crash...however no restart or BSOD, just screen turned off and on again. Brightness on internal screen was on full, no load, regular office work.

I'm getting tired and fed up :), I think I'll just RMA it at this point. Let them check and repair whatever is causing this...

[bright juniper]

I just checked mine 786 pci express error counters...so...

No loads, just light usage, uhm, LOL?

I have already sent a RMA request to see what they will answer...to local service centre.

Ah OK, that's not so problematic, quick google research, oh well.

[dusty crag]

yeah looks honestly like a hardware issue

because my hwinfo does not show this option, just shows the option called hardware issues

so might be the same

[dusty crag]

everything looks normal here temps wise

i guess you got same issue like bungo

[weak oxide]

Im just gonna put faith in the repaste and hope it works for my sanity

[bright juniper]

Update from my side - I RMA'd, tomorrow I'm taking laptop to service. I will see and let you know what they will do...it was briefly mentioned in email (after I gave them short description) the motherboard will be replaced...

[dusty crag]

i also posted my image here how stock paste looks like. Reapplied with ptm. So you can see how your temps are after service and also maybe check if they reapply ptm. Asus loves to not reapply any paste and ships back with old paste lol. idk about lenovo
#legion-7 message

[bright juniper]

Update from my side, received the laptop from service - replaced motherboard. It's now maybe a week or two - it's stable. Regular office work, regular day use, I did some benchmarks, no gaming for now... From everything I can conclude in my case it was hardware related. If any of you still have warranty I suggest you look into that. Before RMA I tried almost everything, nothing worked.