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

This is with CB R15: 1197; CPU and GPU Fans at 150%, I left my background tasks running (OneDrive, MSI, SCM, TrueColor, Dashlane on). Ambient is 27C.

CM MasterGel Maker Nano, and Artic Thermal Pads.

I did try one 'Fix' which was to bend one of the tri posts of the HS screws but didn't get any better (I believe it was one of the fixes that were recommended for one of the Alienware laptops - iunlock)

My next step to improve differentials would be to lap the HS? What do you guys suggest?

 

ugh I was afraid of that. MSI is a nightmare to RMA for isnt it?

 

yes they are. That's why we recommend buying from a trusted reseller like HIDevolution or GenTech PC who do extra quality control and testing to ensure everything is running smoothly before they ship out the laptop to you and also in such cases, they handle the repair so you won't have to go through MSI but anyway. Let's stay positive. All the best and keep us posted.

 

Resurrecting this.
Anyone have issues with their keys not changing colors or being stuck on one color even when changing settings?


Sent from my iPhone using Tapatalk

 

I'd like to second this recommendation. Recently I had a screen issue with my GS65. I reached out to Ken from GenTechPC. Within a couple of hours I had a RMA number from MSI. Shipped them the laptop. They repaired it in 2 days and shipped it back via express shipping. Ken stayed on top of the situation the whole time. Simply couldn't ask for better customer service!

 

What was the difference in Temps before you bent it? I haven't done anything yet but undervolting, and I get 12c difference... Really bad for a laptop in this price range

 

I had some instances of 18. But I always used the max temp differentials, not sure if this is the best way to measure.

Or is current or average temp Delta a better representation?

Sent from my BLA-L09 using Tapatalk

 

Hi all --

First, thank you for this amazing thread. I got my gs65 with 1070 on black friday at best buy -- 1799 with another 90 bucks cash back in best buy credit.

I ordered another half tb ssd from newegg along with 32 gig of ram (arriving today.) My main use is machine learning and deep learning -- Ubuntu is going on to the new nvme drive where I need the increased read/write. I've done a fresh windows install, updated everything other than EC (going to do today) including BIOS. Using silent option instead of dragon center, but I did have dragon center installed.

Bunch of questions as follow:

So far I've done a clean windows install and followed all the UTX undervolting info in this amazing thread. Couple jumps from -100 to -195 and I got a BSOD in firestrike, so I dialed back to 185:

1) -185mv undervolt in UTX. Runs indefinetly at 3.88 ghz without any throttling in the UTX stress test.

• Range is about 8 degrees for my cores, which seems about normal? As usual, 0 and 2 are the culprits.
  
• ran UTX stress test over night, no reboot or bsod.
• ran each of the prime95 stress tests for 15-20 min, no errors reported (left AVX on because it is needed for tensorflow.) Lots of throttling here (power and thermal, depending on the stress test.)
• Did idle test over night, no reboot or blue screen.
• Been doing normal work day, no issues.

Am I good on the Windows undervolt? Should I switch to TS and decrease multipliers? CPU performance matters to me for non-DL machine learning applications...​

2) Can/should I move the undervolt to the bios since Ubuntu is going to be where I do my most intensive computing? All the undervolt options I've found for linux are rando folks' github repos... If yes, does anyone have any links to a solid example/guide for doing this?

3) Did the brief install of dragon center mess up hidden BIOS settings that I need to deal with?

4) What's a well padded case I can get for this in the USA? The best example in the thread is UK Amazon.

Thank you!

 

Do you guys have any solutions to the fingerprint stains? I really like the laptop but I hate seeing a smudge immediately after spending 10 minutes to clean it. Would skins be a good option? Maybe apply an oleophobic coating similar to what phones have?

 

Also a quick note -- My best buy version came with dual channel 2x8gb ram... But apparently it is either 2666mhz, or it is stock over clocked?

I need 32 gb or I would keep it....

 

Dragon Center modifies TDC and ICCMax in BIOS, you would need to unlock BIOS to modify that.

As far as a case, TomToc 15.4 inch from Amazon worked for me (SKU: A13-E01G)

https://www.amazon.com/dp/B01HCF15HC/ref=cm_sw_r_cp_awdb_t1_SHhaCbQ03YXWJ

Careful as others users have reported that it didn't fit theirs, however when we compared SKU we noticed they were different. I bought the gray version.



Sent from my BLA-L09 using Tapatalk

 

Dragon Center modifies TDC and ICCMax in BIOS, you would need to unlock BIOS to modify that.

As far as a case, TomToc 15.4 inch from Amazon worked for me (SKU: A13-E01G)

https://www.amazon.com/dp/B01HCF15HC/ref=cm_sw_r_cp_awdb_t1_SHhaCbQ03YXWJ

Careful as others users have reported that it didn't fit theirs, however when we compared SKU we noticed they were different. I bought the gray version.



Sent from my BLA-L09 using Tapatalk

 

Warning:

Don't do what I did and leave GPU fan on overnight.

I measured the dB at each fan%, found that 30% for me on GPU was almost imperceptible so I left it there default (1st slider for GPU on 30%) thinking I would improve cooling and overall longevity.

I ran PCmark 8 battery benchmark over night (approx 7hour test) and now occasionally my GPU fan will rattle (sounds like something is caught in the blade, obviously it isn't I checked). It does it for a while, then it doesnt for a week or 2. Then comes back for like an hour. One time it lasted 3 days.

Sent from my BLA-L09 using Tapatalk

 

Warning:

Don't do what I did and leave GPU fan on overnight.

I measured the dB at each fan%, found that 30% for me on GPU was almost imperceptible so I left it there default (1st slider for GPU on 30%) thinking I would improve cooling and overall longevity.

I ran PCmark 8 battery benchmark over night and now occasionally my GPU fan will rattle (sound like something I'd caught in the blade, obviously it isn't I checked). It does it for a while, then it doesnt for a week or 2. Then comes back for like an hour. One time it lasted 3 days.

Sent from my BLA-L09 using Tapatalk

 

Sorry to hear that

This is why I don't recommend overnight stress tests - it doesn't help anyway - with about 5 minutes to 15 minutes maximum as enough heating to find issues.

Long stress testing is especially wearing on laptops - fans in particular.

That's also why I don't recommend mining on laptops, even with reduced tuning it's still spinning up the fans and cooking the components.

You could report the grinding fan to your local MSI support and see if they will send you a replacement fan.

 

@hmscott


RMA this? 3 months into my purchase, almost 4. This is with AIDA64. About a 2 minute run. Fans @ 150%.
Repasted with CM Mastergel Maker Nano.

PRIME95 - Small FFTs with AVX off

 

Welp, all they did was tell me to update my bios and software.............. or do software recovery. Standard MSI I guess....

 

Hello all,

New member here and a new owner of this beauty.
I got the best buy version I am currently setting it up and being paranoid about temperatures, first thing I installed was Intel XTU.

in XTU I notice that my package temperature is constantly jumping between 50-70 degrees celcius. Is this normal?

It often bumps above 80-85 for short burst and then XTU tells me that it is thermal throttling.
I am literally doing nothing on my laptop. haven't even set it up completely.
not updated the graphics driver yet.

Is this normal?? or just XTU acting weird?

any suggestions about installing/updating/uninstalling stuff before I move to undervolting?

I also notices that the led beside the power button is usually amber(indicating 1070 is being used) it does turn white for sometime before turning amber again.

any help is appreciated,

Thanks,
Vikas Singh

 

Please post the hwinfo from before the re-pasting...

Once you re-paste you are kinda on the hook to fix it yourself... how many times have you re-pasted trying to get the temps down?

If you have done it a bunch of times already, then yeah, open a case with MSI - post the pre-repaste images that show it thermal throttling, say you tried to re-paste on your own, but it didn't help, and ask for an RMA.

BTW, I didn't notice you mention it, you did already undervolt the CPU, right?

 

[double post]

 

Hello all,

New member here and a new owner of this beauty.
I got the best buy version I am currently setting it up and being paranoid about temperatures, first thing I installed was Intel XTU.

in XTU I notice that my package temperature is constantly jumping between 50-70 degrees celcius. Is this normal?

It often bumps above 80-85 for short burst and then XTU tells me that it is thermal throttling.
I am literally doing nothing on my laptop. haven't even set it up completely.
not updated the graphics driver yet.

Is this normal?? or just XTU acting weird?

As simple as triggering a driver download in GeForce experience causes the package temperature to bump to above 80 and start thermal throttling. Right now, with just a few tabs open in the edge browser and GeForce experience in background, my package temperature is 75+.

Package temperature usually idles around 50-55

should I just wait for the laptop to settle in or is this behaviour concerning?

any suggestions about installing/updating/uninstalling stuff before I move to undervolting?

I also notices that the led beside the power button is usually amber(indicating 1070 is being used) it does turn white for sometime before turning amber again.

any help is appreciated,

EDIT:
As simple as triggering a driver download in GeForce experience causes the package temperature to bump to above 80 and start thermal throttling. Right now, with just a few tabs open in the edge browser and GeForce experience in background, my package temperature is 75+.

EDIT2: During and righ tafter startup, dragon center show fan1 @ 4000+ rpm
fan2 @ 2000+

EDIT3: running on -0.120v , takes under a minute to reach 90 degrees celcius, but stays there for only a second or so, usually settles around 85 but still thermal throttles.. how is that possible?

EDIT4: running stable at core offset voltage of -0.16.. cpu stress test idles at around 85, ocassionally hittin 90, cpu throttles but max frequency is around 3.77-3.8.

Am I experiencing normal behavior here?

I am thinking of RMAing it.
I know it has just been hours since I have this laptops, but I feel mine is surely running hotter.
I have seen people's post in here with much better temps after undervolting to -0.16


should I just wait for the laptop to settle in or is this behaviour concerning?


Apologies for the double post

Thanks,
Vikas Singh
​

 

It fluctuates as soon as you open or do any activities, it's boosting to higher frequencies and getting hotter because of it, it's normal.

Thermal throttle starts 90°C. It may alert you before but frequencies shouldn't drop until the reaches 90+

Dont worry about it. Seems normal .

I believe GPU fan 1st slider activates at 60°C +

Sent from my BLA-L09 using Tapatalk

 

My fan 1 is always on at around 1900-2000rpm at all temps even below 50(cpu).

What I don;t understand is while the cpu frequency bumps are okay, how is it so instantaneously hitting 90+ that it has to throttle?

Is there anything that I can export from XTU that might be useful? like some graph, logs etc?

on default volts, even browsing on chrome makes the laptop go 90+??
Just after startup, the fans run crazy high, 4000rpm+ temps near 90s and thermal throttling.

All of this in a 20C room temperature.

is everyone facing the same behavior?



Thanks,
Vikas Singh

 

You can set a custom fan curve so that fans speed up earlier and thus you avoid the throttle.

Sent from my BLA-L09 using Tapatalk

 

could you share your temps pre and post undervolt?

I also cannot seem to disable turbo boost in XTu. its greyed out.

 

asking because my idle temps at stock volts are no where near 45 degrees that people have here, mostly above 50

 

Hi,
I got an MSI GS65 since april. It was running smoothly so far, but I had not play any game for quite some time now.
Yesterday as I played I had a very unpleasant surprise: after some time playing, the computer makes some crackling/stuttering sound. It is the sound you obtain when you brutally mute/demute your computer, you can find an example of it at 1mn25 in this video: (sorry I could not find anything better).
When I have some audio playing, the sound is quite irregular, it happens only at some specific moment (for example I was listening to a podcast and the crack always happened at several specific point... + some irregular points. And it is not a problem of the audio file apparently: when I'm not on "best performance/plugged in battery" mode, there is no problem with the audio file.
When I cut the audio file and/or put my computer on silent mode, the sound becomes regular: one crackling every 2-3 seconds.

The problem only seems to occur when I have the computer plugged in the mains/not on battery or possibly when I'm on "Best Performance" mode on battery (but I'm not sure about that one). I have to play games for it to occur as far as I can tell. So I supposed it is related to either the graphic card or the heat of the computer, but I can't tell... It happens only after I play for some time (e.g. 30 minutes approximately), that's why I was mentionning potential heat issues.
I am French and bought the computer in France but right now I'm in the US with the battery plugged into US plug, so I don't know if that might be a problem of difference between the regions but I really doubt it (and I've been in the US for a month without any problem so far...).

By the way, apart from the "crackling noise", everything is running smoothly when it occurs no freeze whatsoever for example.

I don't know if any of you faced this issue before... I don't know what to do, I'm worried the computer might explode or something (my previous MSI gs60 laptop had an exploded battery after 3years, that's why I changed to this one).

Thanks in advance,
Sorry for my poor English.

 

I know this is an old post but just wanted to confirm that my GS65 does this as well, so I assume it is normal, but I was a bit worried as well, but now I'm not, - thanks!

@ vdroid1211 I think it is complely normal . It is thin laptop, and it gets really hot. However, the throttle should not be severe - you shouldn't notice any hiccups in games. And a lot of people in here post their temps/benchmarks with cooler bost mode activated, and these temperatures are much different (and the laptop much louder )!

@Nzil: It could be driver problem! Have you tried reinstalling/updating audio drivers? How about the bios and firmware?

 

I tried to uninstall/reinstall Realtek Audio sound drivers, it did not work. I also updated the GC drivers with no results.
I don't know it there's something else I should try to update. (via the device manager everything seems up to date, but i did not check in details every drivers)

EDIT: i also installed a new update for the firmware msi without any result. (not sure I found the most upto date version though)

 

@Nzil
You mention GC drivers, - did you update both the Nvidia and Intel drivers from their websites? I can imagine that these drivers can cause the trouble as well because their clock speed change when you plug it in. Finally, I would update the bios. If none of this works, i would try to reinstall the computer from scratch. Try to play games with cooler boost activated - does it happen in these scenarios as well? If not, your theory about the heat could indeed be correct, and maybe a quick fix would be to clean your computer from dust?

If nothing works, you could consider to send it in.

PS
You cannot trust device manager - these drivers are often outdated. You should go the manufacturer's websites to be sure that you got the latest updates.

 

Where is cpu core 6 placed physically? Can someone show me in a picture? I want to try to tighten the screw nearest core 6 to see if the difference in Temps between core 1 and 6 decreases. 12c is way too much..

 

If you find it let me know!

Sent from my BLA-L09 using Tapatalk

 

I'm thinking about updating the ram and adding a hard disk more

please need recomendations

 

They are usually labeled 0 to 5, rather than 1 to 6, in most HW monitoring programs.

But with the laptop opened and ports facing up, core 6 should be close to the bottom right. Core 1 top left, core 2 bottom left, core 3 top middle, core 4 bottom middle etc.
Note that core 6 is not at the very edge of the corner of the die (in fact none of the cores are at the far edges). Around the cores are iGPU stuff and caches, but you get the point.

 

Yeah, I know that they are labeled 0 to 5, but I was talking about physical placement. Thank you for letting me know how they generally are placed, but my knowledge of what things look like on the laptop motherboard is limited. Can you point out where in this picture core 6 is placed?

 

According to Intel, under 100C is within spec.
​

You are misreading something. If the package temperature never reaches 100C then there is no thermal throttle. Most likely you are tripping the power limit throttle. Which is normal. There is plenty of information about power limit throttle and what you can do about it in this thread, on reddit and elsewhere so I won't repeat it.
Ignore that. Stupid BDPROCHOT.

 

And you think all the OEM's follow specs? ​

http://forum.notebookreview.com/thr...scussions-lounge.815216/page-97#post-10742687

http://forum.notebookreview.com/thr...scussions-lounge.815216/page-54#post-10729965

 

Effing BDPROCHOT.

 

I can't. I don't have any pictures. I can only give you this. Some old picture I had of a 4 core and I don't have one without the black circle either.
Let that be your guide.

 

The activation of Cooler Boost seems to work. If I play with Cooler Boost from the start, I never get the problem (at least after 2 hours I got nothing). If I play without it I get it after 20-30 minutes. Then if I activate Cooler Boost, the problem stops in about 30 seconds.
You think there might be some dust stuck somewhere inside the computer? (I keep it very clean but you never know something could go in I suppose) But then why does the problem only occur when the heat is high?
Btw what is the best way to clean the computer (I only clean the keyboard/screen regularly, but never opened it to see the hardware).

EDIT: the problem seems to occur when GPU temperature reaches around 80°C (which is easily cooled down by Cooler boost). I don't know if that's normal or not though, neither why it happens.

Thanks for your reply

 

Anyone?

Also,
on stock voltages and fan set to auto in dragon center, I get a firestrike score of around 11900
with -0.150mv it bumps to around 12300 with gpu graphics score from 13000-13900
temperatures easily above 88, often hitting 90-92

Are these fine?

Thanks,
Vikas Singh

 

OK

Looks like I sent them a sternly worded "That isn't the issue" and I got an RMA. The RMA location is close so I should be able to get it out there pretty quick so hoping for a quick turnaround.

I wanted to confirm that this is abnormal before shipping it out.

At -.150 undervolt I am experiencing around 17-20% thermal Throttling at 150% fans on the CPU while playing COD black ops
At -.150 undervolt and changing the core multiplier to 39-39-38-38-37-37 I'm seeing about 10% thermal throttling

This is pretty identical to doing a stability test since the game seems CPU intensive. I never noticed it before I think because I mainly play overwatch which was a much lower CPU utilization and it never throttled.

 

Sorry I use Tapatalk on my phone and lost your reply amongst the new posts. I did undervolt to -164 mV. I went ahead and repaste, and bent the tri post fastener above the CPU. Temps are a lot better now the hottest core is core 2, but differentials are 6 or 8 at most.

Now that I solved the thermal limitations I was able to do the IMON SLOPE and OFFSETS since I don't instantly shoot up to 88 or 91 now.

I think the problem may have been when I tried to clean the fans without unmounting the heatsink I may have inadvertently raised it off the die and allowed air to get in between the die and HS. Or maybe the repaste was done to thin.

What I did this time is spread it out with the spatula and after placing an even coating put a drop of TIM in the middle and mounted the HS back on.

 

It looks like you have a handle on the tune and solution, that's great.

 

Anyone with gs65 Ubuntu undervolting experience?

Waaay less straight forward than Windows

 

Alright -- found the voltage offset in the advanced BIOS menu.

Anything I need to do other than set the offset to my XTU - tuned value (I've uninstalled XTU right now) and set the sign to '-'?
Do I need to set the cache offset somewhere in there? Couldn't find an option/setting....
Thanks!

 

Cache and core voltages are linked and should not be separated nor is it necessary to separate them.
If attempted to be set separately in throttlestop, the 'highest' voltage of the two will be used.

The only time the LOWEST of the two voltages will be used is when an AVX load is detected, and only for the AVX load. This is an unsupported configuration and no one can help you with this if you try something this bizarre.

It's not worth it to deal with this type of thing to save a few watts.

 

Yeah -- I know they get linked in xtu/throttlestop (thanks to you + others posts!)

So -- reading the second part of your post -- setting the core voltage offset in the bios *won't* be equivalent to the same setting in TS or XTU?

To clarify -- all I'm trying to do is implement the undervolt I was running via XTU in the bios instead of via software.

Thanks.

 

To further clarify -- if I set the top of these to to 185 and the bottom to '-' will that be equivalent to a -.185 undervolt in XTU/TS, or do I need to do something else?

Thanks, sorry I'm a bit thick.

 

It's the same. Set prefix to negative and value to -185 and it's a -185 undervolt.
Both core and cache are linked to the same value.
Don't try to unlink them unless you want....ahem..interesting things to happen, and I'm not touching that.