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

I didn't notice anything after I installed the latest firmware.

 

I have the newest BIOS and EC installed but am not noticing any differences.

 

Monster Hunter World pushes this laptop's CPU a lot. Touching 90C now and then.

 

The TS interface allows separate numbers for core / cache undervolt, and maintain different numbers in the registers for the CPU to read, but the CPU only uses 1 setting for both Core and Cache undervolt, and it picks the lowest setting - in your case -100mV.

XTU maintains the CPU split core / cache up through Gen 5 CPU's, and starting with Gen 6 CPU's it complains if you try to set different Core / Cache settings - or won't let you.

I'd recommend keeping both core and cache the same value initially when you are testing for 100% load / Idle stable undervolt value - then when you have this locked in, vary the Core undervolt higher in TS to see if that gives you any benefit.

AFAIK the higher Core undervolt can give a higher benchmark by a few points, but isn't noticeable in real use, the difference is too small. Others have noticed less performance in some tests as well when varying the Core undervolt higher than the stable value - while keeping Cache at stable value.

You might get a degree or two more decrease in temperature by using a higher stable undervolt than you are effectively getting as set as above, with Core -150mV and Cache -100mV. Try starting both at -150mV, and increase / decrease as needed by +5mV or +10mV - increasing / decreasing the undervolt until stable.

See above response to an older post of your's, it might be helpful to increase real undervolt to more than -100mV, with both Core and Cache set to the same value.

Please let us know what you fine.

 

Got it, thanks man

 

Pleease, it will only take a couple of minutes

 

Here you have a screenshot from the review in Nootebook check:


This latency is probably due to the Killer Wifi suite, so without installing it should not be any problem. Also I must say the XPS 9570 has no problem to work with audio, same thing, just need a fresh Windows installation. I have colleages with 9550/9560/9570 and they have no problems with them. A fresh install of windows usually solves all of those problems and sometimes using generic windows drivers instead of using the specific "brand software package" is better. Whatever, I would recommend the GS65 over the XPS for audio production due to extra abilities of being able to control fans, has 2 ssd slots and usually MSI laptops performs better than other brands. Just my 2 cents as a mixing engineer.

 

With CPU adjusted just right, is it safe to play with GPU performance at all? It doesn't hit high temperatures with stock settings so I'm wondering if overclocking it with MSI Afterburner is a good idea. Can it do any harm to the graphics or shorten it's lifespan?

 

Thanks for the responses JRey and przybytek

 

Usually the higher the temps = shorten life of electronics, but if you retain them in a safe zone we can't really now the life span.

 

Maybe someone here that has done tuning for both CPU and GPU - including undervolting the GPU (if possible on this model), can help?

Well, what I do is OC both independently. OC for best CPU tune, then set CPU back to defaults, and then OC for best GPU tune, and now I have best case settings for both without consideration to either added power and cooling requirements when doing both.

Then I try the best CPU tuning + best GPU tuning, and see if they are compatible together, run the same CPU only and GPU only benchmarks, they should both be better or the same.

Then try benchmarks that use both, like games. Compare both OC'd against your best stock game benchmark scores.

You may find that your laptop can do both for most applications, but some benchmarks may suffer with both fully OC'd, or it might become unstable - not enough power for both at full performance, or it might become hotter - not enough cooling for both at full performance.

Usually it's the cooling that can't keep up in thin laptops, but it can be power starved too.

So you have to decide which you want full power, and which is going to run a bit off full performance.

If the GS65 pascal GPU can be undervolted and that will actually not only improve performance, it will also reduce power and cooling, something to try first before using MSI AB to increase clocks.

https://www.youtube.com/results?search_query=undervolt+msi+gpu

 

Could be. I have the BestBuy edition without Killer WiFi (mine has the Intel Wireless-AC 9 chipset). My statistics:

Spoiler

_________________________________________________________________________________________________________
CONCLUSION
_________________________________________________________________________________________________________
Your system appears to be suitable for handling real-time audio and other tasks without dropouts.
LatencyMon has been analyzing your system for 0:02:18 (h:mm:ss) on all processors.


_________________________________________________________________________________________________________
SYSTEM INFORMATION
_________________________________________________________________________________________________________
Computer name: BELLDANDY
OS version: Windows 10 , 10.0, build: 17134 (x64)
Hardware: GS65 Stealth Thin 8RF, Micro-Star International Co., Ltd., MS-16Q2
CPU: GenuineIntel Intel(R) Core(TM) i7-8750H CPU @ 2.20GHz
Logical processors: 12
Processor groups: 1
RAM: 16228 MB total


_________________________________________________________________________________________________________
CPU SPEED
_________________________________________________________________________________________________________
Reported CPU speed: 2208 MHz

Note: reported execution times may be calculated based on a fixed reported CPU speed. Disable variable speed settings like Intel Speed Step and AMD Cool N Quiet in the BIOS setup for more accurate results.


_________________________________________________________________________________________________________
MEASURED INTERRUPT TO USER PROCESS LATENCIES
_________________________________________________________________________________________________________
The interrupt to process latency reflects the measured interval that a usermode process needed to respond to a hardware request from the moment the interrupt service routine started execution. This includes the scheduling and execution of a DPC routine, the signaling of an event and the waking up of a usermode thread from an idle wait state in response to that event.

Highest measured interrupt to process latency (µs): 333.449430
Average measured interrupt to process latency (µs): 4.614389

Highest measured interrupt to DPC latency (µs): 301.913183
Average measured interrupt to DPC latency (µs): 1.349727


_________________________________________________________________________________________________________
REPORTED ISRs
_________________________________________________________________________________________________________
Interrupt service routines are routines installed by the OS and device drivers that execute in response to a hardware interrupt signal.

Highest ISR routine execution time (µs): 116.18750
Driver with highest ISR routine execution time: ndis.sys - Network Driver Interface Specification (NDIS), Microsoft Corporation

Highest reported total ISR routine time (%): 0.002992
Driver with highest ISR total time: Wdf01000.sys - Kernel Mode Driver Framework Runtime, Microsoft Corporation

Total time spent in ISRs (%) 0.004724

ISR count (execution time <250 µs): 32404
ISR count (execution time 250-500 µs): 0
ISR count (execution time 500-999 µs): 0
ISR count (execution time 1000-1999 µs): 0
ISR count (execution time 2000-3999 µs): 0
ISR count (execution time >=4000 µs): 0


_________________________________________________________________________________________________________
REPORTED DPCs
_________________________________________________________________________________________________________
DPC routines are part of the interrupt servicing dispatch mechanism and disable the possibility for a process to utilize the CPU while it is interrupted until the DPC has finished execution.

Highest DPC routine execution time (µs): 432.247283
Driver with highest DPC routine execution time: storport.sys - Microsoft Storage Port Driver, Microsoft Corporation

Highest reported total DPC routine time (%): 0.035682
Driver with highest DPC total execution time: Wdf01000.sys - Kernel Mode Driver Framework Runtime, Microsoft Corporation

Total time spent in DPCs (%) 0.084966

DPC count (execution time <250 µs): 287442
DPC count (execution time 250-500 µs): 0
DPC count (execution time 500-999 µs): 5
DPC count (execution time 1000-1999 µs): 0
DPC count (execution time 2000-3999 µs): 0
DPC count (execution time >=4000 µs): 0


_________________________________________________________________________________________________________
REPORTED HARD PAGEFAULTS
_________________________________________________________________________________________________________
Hard pagefaults are events that get triggered by making use of virtual memory that is not resident in RAM but backed by a memory mapped file on disk. The process of resolving the hard pagefault requires reading in the memory from disk while the process is interrupted and blocked from execution.

NOTE: some processes were hit by hard pagefaults. If these were programs producing audio, they are likely to interrupt the audio stream resulting in dropouts, clicks and pops. Check the Processes tab to see which programs were hit.

Process with highest pagefault count: explorer.exe

Total number of hard pagefaults 1427
Hard pagefault count of hardest hit process: 229
Number of processes hit: 55


_________________________________________________________________________________________________________
PER CPU DATA
_________________________________________________________________________________________________________
CPU 0 Interrupt cycle time (s): 9.721088
CPU 0 ISR highest execution time (µs): 116.18750
CPU 0 ISR total execution time (s): 0.071924
CPU 0 ISR count: 27188
CPU 0 DPC highest execution time (µs): 432.247283
CPU 0 DPC total execution time (s): 1.368943
CPU 0 DPC count: 282240
_________________________________________________________________________________________________________
CPU 1 Interrupt cycle time (s): 6.566135
CPU 1 ISR highest execution time (µs): 6.796196
CPU 1 ISR total execution time (s): 0.000085
CPU 1 ISR count: 40
CPU 1 DPC highest execution time (µs): 118.894022
CPU 1 DPC total execution time (s): 0.008885
CPU 1 DPC count: 955
_________________________________________________________________________________________________________
CPU 2 Interrupt cycle time (s): 4.830835
CPU 2 ISR highest execution time (µs): 0.0
CPU 2 ISR total execution time (s): 0.0
CPU 2 ISR count: 0
CPU 2 DPC highest execution time (µs): 184.949275
CPU 2 DPC total execution time (s): 0.006668
CPU 2 DPC count: 938
_________________________________________________________________________________________________________
CPU 3 Interrupt cycle time (s): 5.606699
CPU 3 ISR highest execution time (µs): 0.0
CPU 3 ISR total execution time (s): 0.0
CPU 3 ISR count: 0
CPU 3 DPC highest execution time (µs): 60.465580
CPU 3 DPC total execution time (s): 0.000473
CPU 3 DPC count: 93
_________________________________________________________________________________________________________
CPU 4 Interrupt cycle time (s): 5.690693
CPU 4 ISR highest execution time (µs): 0.0
CPU 4 ISR total execution time (s): 0.0
CPU 4 ISR count: 0
CPU 4 DPC highest execution time (µs): 62.336957
CPU 4 DPC total execution time (s): 0.008026
CPU 4 DPC count: 1232
_________________________________________________________________________________________________________
CPU 5 Interrupt cycle time (s): 5.994090
CPU 5 ISR highest execution time (µs): 0.0
CPU 5 ISR total execution time (s): 0.0
CPU 5 ISR count: 0
CPU 5 DPC highest execution time (µs): 119.986413
CPU 5 DPC total execution time (s): 0.002067
CPU 5 DPC count: 262
_________________________________________________________________________________________________________
CPU 6 Interrupt cycle time (s): 4.661256
CPU 6 ISR highest execution time (µs): 0.0
CPU 6 ISR total execution time (s): 0.0
CPU 6 ISR count: 0
CPU 6 DPC highest execution time (µs): 95.928442
CPU 6 DPC total execution time (s): 0.001136
CPU 6 DPC count: 218
_________________________________________________________________________________________________________
CPU 7 Interrupt cycle time (s): 5.351660
CPU 7 ISR highest execution time (µs): 0.0
CPU 7 ISR total execution time (s): 0.0
CPU 7 ISR count: 0
CPU 7 DPC highest execution time (µs): 254.670290
CPU 7 DPC total execution time (s): 0.002741
CPU 7 DPC count: 345
_________________________________________________________________________________________________________
CPU 8 Interrupt cycle time (s): 4.377252
CPU 8 ISR highest execution time (µs): 12.798913
CPU 8 ISR total execution time (s): 0.00070
CPU 8 ISR count: 595
CPU 8 DPC highest execution time (µs): 111.528986
CPU 8 DPC total execution time (s): 0.002738
CPU 8 DPC count: 469
_________________________________________________________________________________________________________
CPU 9 Interrupt cycle time (s): 4.877268
CPU 9 ISR highest execution time (µs): 19.423007
CPU 9 ISR total execution time (s): 0.005544
CPU 9 ISR count: 4581
CPU 9 DPC highest execution time (µs): 126.430254
CPU 9 DPC total execution time (s): 0.001694
CPU 9 DPC count: 252
_________________________________________________________________________________________________________
CPU 10 Interrupt cycle time (s): 4.269626
CPU 10 ISR highest execution time (µs): 0.0
CPU 10 ISR total execution time (s): 0.0
CPU 10 ISR count: 0
CPU 10 DPC highest execution time (µs): 146.622283
CPU 10 DPC total execution time (s): 0.002388
CPU 10 DPC count: 285
_________________________________________________________________________________________________________
CPU 11 Interrupt cycle time (s): 4.601062
CPU 11 ISR highest execution time (µs): 0.0
CPU 11 ISR total execution time (s): 0.0
CPU 11 ISR count: 0
CPU 11 DPC highest execution time (µs): 158.822464
CPU 11 DPC total execution time (s): 0.00160
CPU 11 DPC count: 158
_________________________________________________________________________________________________________

 

A little update. So I re-rolled the drivers back to factory drivers (391.48) from MSI's website and I stopped having problems. For now.

I reported the issue to NVIDIA detailing what happens and how it happens so hopefully they can release a fix in the upcoming driver updates if they cared to. For now, I will be unable to update my NVIDIA drivers if I want Overwatch to work properly, but I'm glad I won't have to go through the hassle of RMA'ing or buying an entirely new device. Shame that something as expensive as this doesn't work well with newer drivers but nothing ever really works as intended on Windows machines.

 

What core, memory clock and undervolt offset values would you recommend as a starting point then?

 

I don't have a GS65, so I can't help with these specifics for tuning, maybe someone else here can help you start out.

Or watch the video's and read articles / posts on how to undervolt the GS65 and / or 1060 and see if you can putter along on your own - shouldn't be too tough to start gradually to undervolt, then when you've found a good setting to adjust the clocks if need be - but it might be good enough to undervolt. It also might not be possible on the GS65...



Nobody here has tried it yet? Or, does it not work?

 

Most people stop trying after knowing that GTX1070 Max-Q uses the voltage between 0.625V~0.800V.

Why do they give up? Because in MSI Afterburner, the lowest voltage allowed to play around in the Frequency / voltage table is 0.800V. So currently through the software tweak, you can only overclock but not really doing the undervolting that everyone initially had in mind coming from the way the CPU is using. But think about it another way, overclocking is actually undervolting on the Pascal, because you have a TDP limit of 80W, you were already hitting that before overclocking, so if you add more frequency, you'll find that it automatically lower the voltage to match with the given TDP limit. And you'll also find that the temperature doesn't change much no matter what frequency you use (unless it is like -400 or something), because it'll race for the targeted 80W TDP when more processing power is required by the Game. if you are coming from a mind set of lowering the temperature, eventually you'll find out messing with the frequency doesn't change anything, limiting the FPS is the King in this situation.

Pascal Max-Q is all about the FPS and the TDP, other stuff won't do much, these 2 are the only directions if you are trying to make some changes (Hint: Chip programmer)

 

This guy got some mileage out of voltage curve setting on his GS65 1070 Max-Q:

MSI GS65 Stealth Thin -- (My Personal Unit's) Optimal Settings for Thermals
https://www.reddit.com/r/MSILaptops/comments/8igeb8/msi_gs65_stealth_thin_my_personal_units_optimal/

"...
Graphics Card (GTX 1070 Max-Q):

I was able to sustain a stable undervolt down to 912mV (from just over 1V) which allowed me to run Fortnite at 120FPS (I capped it here because I personally can't see the difference between 120FPS and above) with max AA and textures (no shadows) at 65-68C with a moderate fan curve (65%ish at medium to medium-high temperatures). This was done using the voltage control on MSI Afterburner."

@przybytek

Here's another, more questions than answers, but there's some info:

Help with undervolting GS65 Stealth Thin
https://www.reddit.com/r/MSILaptops/comments/8zq1mp/help_with_undervolting_gs65_stealth_thin/

There's a bunch more reddit and other site posts - more questions, but no more info than already linked in my posts to you, you're gonna have to watch the tutorials, google for others answers as time goes by if you wanna wait, or jump in and start implementing the tutorial info.

Please let us know how it works out.

 

It depends on the TDP, Fortnite may not be requesting that much juice from the GTX1070 Max-Q, if the processing power required by the game isn't able to make the GTX1070 Max-Q pull 80W, then yes it would run at the highest voltage with the highest frequency available, if he is pulling 80W on the dGPU then with 65% fan it will definitely exceed 70C, it is unfortunate that the Author never posted his curve. I think if he runs heavy games things will be different.

 

Hey guys, what would you recommend based on your experience with the gs65?

After decent discounts I'm getting either the aero 15(7700hQ, 1060 60hz FHD) or the new gs65(1060, 144hz, 8750hq, 70$ more.)

I'm not a huge gamer, like I'll occasionally play Cod or fifa etc and also don't care much about high/ultra settings. Priority would be battery life and a good screen! Portability basically. Thanks

 

Thanks for the answer, but the 9570 has a known problem with acpi.sys so clean install w10 doesn't help.
https://www.reddit.com/r/Dell/comments/8rleyj/dell_xps_15_9570_i78750h_awful_dpc_latency_very/
Your colleagues should check their latency because this is a problem with the whole xps line.

So can someone please test Latencymon on MSI GS65?

 

Thank you so much!!! Going for gs65 then.

 

Do you guys think that there will be replacement batteries for this laptop in the future? I'm trying to think long term and I don't know if there will be parts for this thing in the future.

 

MSI is a very large global company that has been around for decades. Batteries, and replacement parts, for their previous models, going back years, are still available from multiple sources online, as well as directly from your vendor, or MSI themselves.

Rest easy

 

Just over a month into using my GS65, and for people like me who travel a lot, work on flights, (basically your laptop spends more time in a bag than at a desk), I'm not sure I'd recommend this machine, because the build quality does not inspire a lot of confidence. I've used my Alienware 13 R2 (admittedly, a tank of a machine) everywhere, and squeezed it into a zillion airplane pouches, but never any hardware/build related issues.
After a month with the GS65, there's creaking all over the palmrest. The front part of my right side screen hinge needs to be re-popped in every other day. After a couple of hours in a (tight but not too tight) airplane seat pocket, I took it out and found that the colours on certain keys have a permanent red hue to them. Apparently the keyboard backlight ribbon cable shifted slightly due to pressure and support says I need to take it in to a service centre to fix it. I've also picked up a nasty chip on the "premium" gold bezel from innocuous contact with my watch. Again, this is just one month. I'm now just constantly paranoid and scared that there's going to be a new creak or warp or scratch on my $2000 laptop every time I take it out after a flight.

If you just want a nice pretty laptop that sits on your desk all day, the GS65 is AMAZING. It's got the performance, the display, the keyboard, the battery life, everything I love about a good gaming laptop. But this build quality is in direct conflict with the main USP of the thin and light gaming machine - PORTABILITY.

If you're the "out and about" variety, I suggest dropping some extra bucks on the new Razer Blade for the build quality, or even the Aero 15X (if you can live with the crap keyboard). I'd have preferred to get either of those, but none have warranty/support in my region, and I didn't want to live in paranoia. Bit ironic, as it turns out. Guess it's two years of kid gloves now.

 

You just confirmed my fears. My first thoughts after unboxing the notebook were how good, thin and - unfortunately - delicate it looks.

 

I travel 3-5x/month (GS65 owner since April) and it has been a great, durable laptop. No better or worse than my previous Razer Blade 14. Of course, I care for it properly - store in a sleeve, and transport in a TUMI case (sandwiched with my MacBook Pro 13 which is also in a sleeve).

You can't do thin and light without having some kinds of compromises. If you want a tank - buy one.

 

I agree there is compromise, but if all its competitors can manage a solid build in the same form factor at a similar price, performance and thermals, makes me wonder why MSI couldn't do the same. I'm not exactly throwing the laptop around!

I also carry my laptop in a sleeve and a nicely padded Thule Backpack, and am very careful with it, but hinges coming apart and ribbons getting displaced from an airplane pouch is really disappointing.

Maybe I've just been spoilt by my previous laptops. Alienware 13 R2 and a Thinkpad T430. The build quality on both was par excellence.

 

I carry mine in a neoprene sleeve tucked into the water bladder pocket in my pack (REI Traverse 35). Unpacked and opened/closed repacked several times daily. No problems with hinges or keys or anything.

If you pack a notebook under pressure you will damage it so I'd call this poor treatment on your part regardless of build quality. Any tight is too tight.

 

Yeah I mean - I never put anything in airplane pouches for 3 reasons:

1) they're filthy and never cleaned
2) they're filthy and never cleaned
3) they're filthy and never cleaned



I hear what you're saying. Personally, if the Razer Blade 15 was available when I made my choice in April I may have gone a different route - so I think that laptop is worth looking at. However, with Razer's QC problems I would make sure to buy at a place with an extended warranty available. MSI has been a great company for me through GPUs, motherboards, and laptops.

 

You have to be brave enough to put a 15.6" laptop in a airplane pouches.

To be honest that's just too much for me.

 

Hello friends, I did some tests to figure out the levels of activation and fan speed RPM in Silent option. Fan activation is based on temp. For example, getting to 56c in the CPU will activate level 2 fan speed.

Here you go.

 

Is it safe(ish) for me to apply Coollaboratory Liquid Ultra to the CPU and GPU? I jumped the gun and already bought it and just wanted to double check since it boldly says I cannot apply it to aluminum.

 

It is if you do the full proper prep work first. Super 33+ tape completely around the GPU core and fully flattened--a LOT of SMD resistors around the GPU that need protection. CPU also needs Kapton or Super 33+ tape as well. You can get away with using 3 coats of transparent nail polish painted over the SMD resistor parts but make sure the coat doesn't crumple up and cause heatsink pressure resistance. (don't use nail polish *AND* tape, use one or the other; tape is easiest).

You -may- also need to sand the CPU heatsink slightly, with the kit I mentioned and linked in the GT75 thread, to help remove the convex shape from the CPU copper plate, then replace and apply 0.5mm Arctic thermal pads to the CPU chokes and CPU VRM and remove the 1mm ones. This works well on the GT75 but i don't know the measurements of the GS65; I think some people in this thread mentioned something about the pads.

That being said, it's a LOT harder and more annoying to repaste the GS65, because it requires full motherboard removal, and if you do something wrong or the paste isn't fully applied correctly or you have very high core temp differences because of imbalanced pressure when tightening, you have to do the entire removal --again-- This is enough to make anyone rage quit. On the GT75VR and GT73VR, all you have to do is unscrew the fans and then the chip heatsink screws and pop, off the heatsink goes, literally a <2 minute job after opening the case and (ALWAYS) unplugging the battery connector. The GT75 Titan is a 5-10 minute job because repasting the CPU requires -full- removal of both the GPU plate heatsink and GPU VRM heatsink before you can even remove the CPU heatsink, all because of that new stupid half radiator for the CPU that is installed *below* the GPU half radiator instead of on TOP of it (absurd), since repasting the CPU requires far more work getting it perfect (including sanding or redoing paste a few times) due to the convex heatsink, while the GPU heatsink is flat and easy to install and remove.

The GS65 requires full removal so this is not something you want to have to repeat if you find your application is not good for some reason. Your maximum full load temps should NOT exceed 2C delta with LM (if not using AVX or FMA3 torture). The GS65 was not a laptop designed to be serviced easily, unlike the Titans and Clevos (which are easy to service).

 

Thank you for the thorough write-up. Just to be clear I just need to do a careful teardown and access the GPU and CPU. apply protective tape to both, sand down the CPU copper plate, and replace the thermal pads correct? And of course apply the CLU.

 

Hi there’s an option at gs65 bios under graphics configurations to change the primary display from SG to peg. Would this disable Optimus and let me use Nvidia always or the laptop won’t work?

 

Yes, but no one has experimented with sanding on the GS65. Only on the GT73VR and GT75VR (the GT75 titan uses the same stock heatplate, just with an extended heatpipe to a new half radiator). So that's purely and fully at your own risk here, and no one has tried sanding a unified heatsink like that yet. Plus spare GT73 and GT75 heatsinks are not hard to obtain ($70 from MSI's parts department from Tom Ho, although they come from China on order).

You could see if MSI would somehow sell you a spare heatsink (contact Tom Ho, do a search for his email) but good luck with that. At least do the tape and 0.5mm thermal pad replacement first (do not replace the thermal pads on the GPU, although I don't know what size pads are used; on the full sized Titans, they are 1mm pads on the GTX 1070 and 1080).

 

No one knows, as no one has tried it. Be prepared to do the 45 second power button press to clear the CMOS if you get a black screen. No one has tried enabling the iGPU on the GT75 Titan, by changing these same options, or by changing the bit in EC RAM (via RW Everything) register "F1" from 00 to 02 either. So you're on your own. I don't have access to any of these laptops.

Try it and report back?

Remember to also consider using an EXTERNAL DP or HDMI monitor before you give up.

 

Ok so I disable igpu and switched primary display to peg and got a black screen. I already cleared cmos 3 time and I’m still getting black screen. External display over hdmi says no cable connected. Can anybody help?

 

As I said that was at your own risk because no one else has tried it.

Press the power button to power the laptop on then hold it down and do not release it.
the laptop should power off.
Keep the power button held down for 60 seconds.

Then you need to wait *TWO* minutes because the entire CMOS will be cleared and the RAM timings will be reset to default, forcing the RAM to retrain.

If *THAT* fails, you need to pop the CMOS battery out and wait 5 minutes.

 

Should I have the laptop charger plugged or unplugged to clear cmos?

 

Best to have the AC plugged in when doing the power button press.

I don't remember if anyone with a GS65 did the clear cmos trick.
That was tested on the Titan Chassis (GT75, GT75VR etc), but there's no reason why it shouldn't work.

There were two ways to clear the CMOS on a GT75VR, a fast clear and a slow clear.
the fast clear only resets basic RAM settings and some other settings.
The slow clear resets the time/date to 1/1/2010.

I know the fast clear is simply pressing the power button to power on and holding it down until it powers off by itself (about 5 seconds) then releasing it, then powering on and waiting 3 minutes.

The slow clear was holding down the power button for 60 seconds, then waiting 3 minutes.

At worst you pull the CMOS battery (please disconnect the battery plug *AND* AC adapter before doing that) and wait.

tl;dr: don't mess with those graphic settings on a GS65. especially if clearing CMOS isn't as easy as it is on the GT75.

(I had to do my own clear cmos yesterday when I tried running CAS 17 RAM at CAS 13...now that was a riot....)

 

Can anyone here send me photos of the bios advanced tab with settings not hidden. Then a picture of SA settings then graphic configurations and if possible the same picture but with primary display set to peg and integrated graphics disabled?

 

The 60 second button press didn't work for you? Did you wait 3 minutes?
Where is your CMOS battery ? It should be on the flip side of the motherboard (unless it's in the front).
If you can't clear cmos easily, how did people tweak their RAM timings so far without risking a permanent black screen?

@hackness should have those pictures. PM him.

 

I did test the cmos reset before trying this and it did reset the igpu pre-aloc memory so I thought it would work to reenable igpu but it didn’t. But I found gs63 bios pictures and they looked the same as I remembered from my gs65 so I followed it to load optimized default and it worked. Thanks anyways.

 

Interesting.
So your CMOS reset only does a "Soft reset"?
E.g. you would be able to reset RAM back to default after a RAM Timing fail?

If your time/date didn't change to January 1, then it's just a soft reset and none of your other bios settings would be changed except basic overclocking settings.

A hard/full reset is required to reset System Agent, and all of the other settings back to stock. I was not aware that you could not do this on a GS65 without pulling the CMOS battery! (I apologize).

It's already been confirmed the "Full" CMOS reset works on the "Titan" series.
Plus I have a SPI hardware programmer if things really go bad

 

On the Titan series there are three resets:

1) RAM reset after a RAM timing fail (60 second black screen+power light turns off then on, another black screen then board boots)---RAM is reset back to stock timings HOWEVER Command Rate (CR) remains at "1" instead of resetting back to 2. Overclock and voltage settings still remain in effect. This doesn't require you to do anything. This does change the "Spread Spectrum" from 0% (if you want 100.0 BCLK) to 0.50% (99.750 mhz) however.

2) Soft reset (changes basic overclock settings to stock, RAM timings back to default, most settings restored); turning on the power button and pressing it down until it turns off is enough for that. Usually does not reset time/date, so a lot of other settings may not be restored. Overclock saved profiles are still there.

3) full complete CMOS reset (acts like you just did a BIOS flash, or wipes everything back to default settings, time/date set back to January 1, 00:00:00, CPU multipliers fully reset, in some cases even the overclock advanced menus get set to "Disabled" and "CPU VR Config" gets set to Disabled too. What's funny is it doesn't always get set to disabled. I've never been able to figure out why sometimes it remains enabled and sometimes it gets set to disabled. Of course I have no access to the CMOS battery since it's behind the motherboard, but that would definitely do that too if it was pulled out.

I do NOT know if this erases saved overclock user profiles or not.

 

DVMT is in SA setting and it did reset with cmos reset. But integrated graphics and primary display settings did not( I held the button for more than a minute and waited about 5 minutes).

 

Did your system time/date reset to 01/01/2016? or 2017.

 

No