MSI Laptops Drivers

Updated!


Sent from my iPhone using Tapatalk

 

Same here

 

I didn't find changelog for new bios in msi website
and it makes me curious

Sent from my CPH1723 using Tapatalk

 

Says it has the updated CPU code for CPU vulnerabilities.


Sent from my iPhone using Tapatalk

 

Confirm spectre and meltdown protected now


Sent from my iPhone using Tapatalk

 

When did you find a changelog from MSI? It's very rare. Just cross your fingers and flash away...

 

That is exactly what I DON'T Want! Performance comes at number one for me. I have ESET Internet Security to protect me and that's all I need. I'm willing to risk a little bit to have my 100% performance. So, I've disabled the Spectre + Meltdown Patch via a reg tweak





@Papusan @Mr. Fox @Falkentyne

 

Very impressed with that VID regulation btw, Phoenix.
You still using IA AC DC loadline=10 or did you be a real man and change it to 1 (the lowest value)?

That being said, the change log is in the text file.

Not updating unless there's an actual new Bios setting. Which there isn't
BTW last 2 bioses have something called "Intersil VR Command" added in the CPU VR section. Yet seems totally pointless. Some google search shows something like this added in coreboot a YEAR ago (whatever that is) because some S3 or S4 state wasn't working in Linux or some silly stuff. Absolutely USELESS option

 

Bruh, you have the InSpectre tool. Please use it and check how the patch affects your MSIBOOK. Of course you must disable all protection once you are finished with the testing. The Microcode is for Spectre vulnerabillity flaws. Aka enable Spectre protection and run Bench bruh See if MSI has put in an extra mickey in their firmware. Same as Dellienware normally prefer.

 

I have not the time for this bro

 

I didn't, I don't remember you telling me to change it to 1 and what's the benefit? Also, keep my voltage @ 1300 if I change it to 1?

 

Nothing new in this Bios except more cancer.

Slightly tighter voltage regulation. Brings VID range even closer. You would have to test for stability again though but 0.1 mOhms to 0.01 mOhms is only about 5mv (estimated)

So if you needed 1.30v (1300mv) for stability yo umight need 1305mv (1.305v) for stability after.
Not a big deal.

Skylake boards, according to Baidu, used ver 1.0 mainboard, and Sirgeorge had bad instability with AC DC values of 1, 2 and 3. Seems fine for Kaby lake.

 

Checked

Sent from my CPH1723 using Tapatalk

 

You don't need to flash new Cancer Bios to get the microcode. You find latest so called secure Intel microcodes in Micro$haft Update Catalog (Skylake-Kaby-Coffee). Ps. for Win X Fall Cancer Update.

 

Your processor is safe? You mean your processor just got pwnd!

I don’t know what are you so happy about? The fact that you got this false sense of security when we know there are 1000 ways of getting infected with or without the spectre and meltdown patch or are you happy about the fact that Microsoft has robbed you from your taptop’s performance?

 

I just reflash my bios . and I do not know about this
it's possible that the bios display details look younger
so what else i do disable on spectre and meltdown via reg tweak
I also use eset just like you

Sent from my CPH1723 using Tapatalk

 

The Intel Graphics Driver link has changed:
https://downloadmirror.intel.com/27484/a08/win64_15.65.4.4944.zip

Still listed as v23.20.16.4944, though.

 

nVIDIA GeForce Driver updated to v391.24

 

Intel WLAN & Bluetooth Drivers updated to v20.40.0

 

very similar to my drivers layout except for the Dragon Center, I hate that cancer program that keeps changing my CPU's VR Current Limit to Auto which causes CPU Throttling.

 

What are you getting? A GTX 1080 or 1070 version?
7820HK or 7700HQ version?

I really hope you didn't get the 7700HQ. That CPU is a dog. Locked down BGA chips, 45W TDP and 3.4 ghz on all cores maximum..

 

Awesome man, so glad you joined the MSI Team.

I decreased my temps by 10C by changing from Liquid Ultra to Thermal Grizzly Coconut after @Falkentyne and @Papusan suggested it to me and that allowed me to raise my overclock from 4.2 GHz. up to 4.6 GHz. but I keep it at 4.4 GHz to be on the safe side and to keep the temps even lower. Good enough for me.

4.4 GHz is the magical number, if anyone remembers the Devil's Canyon CPU 2 years ago, that was one awesome CPU that clocked at 4.4 GHz out of the box with no overclocking or tweaks needed.

 

Run a K version of LGA chips with no overclocking is a crime!!

 

noooooo man, I tried Kryonaut it is a piece of garbage. The temps were the same as IC Diamond but after 3 days the temps kept rising and rising until my laptop was really overheating so I removed that garbage. What you want is Thermal Grizzly Conductonaut (AKA Thermal Grizzly Coconut as it's known here on the forums )

 

Finally you pop up! Welcome back Mr. Poopoosan.......

 

Yeah, I had to, beacause some need to correct you, bruh. 4.4 ain't a magic number, and have never been. 5 bruh, 5.0GHz or higher starts smell like magic

 

it's super easy man. I didn't even use nail polish or electronic tape, just carefully put a pea sized dot, then used a small brush to spread it to create a nice thins and even layer and slapped the heatsink back on.

Be brave.... be like Phoenix....

 

Like this, poopoosan?

 

This benchmark aint stressfull. More like easy not stressfull walk in the park.



Run Cinebench R15

 

I can't! The laptop will shut off *instantly!*

 

bruh. But you as I know, bro @Falkentyne get kicks (wet in the pants) off of good bench scores. From Papusan's Camp 49x@1104cb

Spoiler: 49x@1102cb

 

That's such a good score....

 

New Battery Calibration available
1.0.1801.1701
Release Date
2018-03-23
File Size
7.14 MB

 

What do you think sounds better? Game like a God, or Game like a Boss?

 

But God sounds so much better when you use some English on the sounds.....
I give up. You win this round. Going to go back to BGA hell for a bit. Need to lick my wounds.

 

Shh don't say it out loud!! Shh....or the BGA BEAR will have your head! @Papusan

 

@Phoenix Poopoosan asked me to do this. so:

cinebench x42-x46 runs. Cache -3 lower than CPU. voltage 1.275v (because i used throttlestop to change speeds), fan speed 100%.

 

MSI Battery Calibration updated to v1.0.1801.1701

 

YEES, bro Falkentyne. Scores should be in the driver thread. This can show the others that the drivers @Phoenix post in his thread works Since you changed multi only for one profile, what clocks have you put in the 4 different TS profiles?

 

@Papusan

Profile 1: i change to whatever I set in bios because im always changing the clocks every boot (because i'm weird). Just change multi and hit apply. To get cache ratio higher than 36, I have to set max cache ratio manually in Throttlestop. (or I could just use the bios too whatever).

Profile 2: 4200 mhz
Proflie 3: 3500 mhz
Profile 4 (battery power profile), 2900 mhz, Speed Shift=128

 

None is weird, bruh!! Only me
Why not put max safe clocks and voltage from bios for one of the TS profiles? Use then one for testing. And two for fixed speed.

 

I used to do that then I stopped. Too many BSOD.

If you unlock the voltage for one profile (example: 4700 mhz, 1.275v), then have another profile with "3500 mhz, default voltage" (voltage not unlocked),
When you change to 3500 mhz, the voltage will be 1.275v, because the 3500 mhz profile is using the "default" voltage (which in this case is the last active voltage).
So then you have to set 3500 mhz and an exact voltage, too. (like 1.0v).

One time, I had 3500 mhz @ 1.0v and I selected 4500 mhz, but (Default) voltage. So it used 4500 mhz and 1.0v and instant BSOD.

There are ALSO problems when one profile is set to ADAPTIVE voltage + offset, and another set to STATIC VOLTAGE. or switching back, if one "Option" is not unlocked.

I also did this back before I knew about "IA AC DC Loadline" too.....like I tried 4500 mhz, 1.175v and found I was putting 1.3v into the CPU.......

MSI cancer.

tl;dr: I think the MSI firmware is using "hidden" loadline calibration and maybe something gets messed up when you use Throttlestop to change voltages in windows.

You see @Vistar Shook 's eVGA? He is using HIGHER voltage than me at 4700 mhz but LESS POWER DRAW ....and 4800 mhz, much higher voltage than me but less power draw...and he could not use 4900 mhz even at 1.45v...but i can do cinebench at 4.9 ghz and 1.380v, but the laptop shuts off after a few runs because of the 94C "EC Current trip protection" (this proves that @BeastsForever.TheDragon was telling the complete truth).


I think the eVGA has a lot of vdroop (like desktops with loadline calibration disabled (LLC=off), but MSI is probably using some hidden cancer loadline calibration. Do you remember??
If @Vistar Shook sets IA AC DC Loadline to AUTO (0), he has NO VID BOOST with STATIC VOLTAGE !!! ONLY with adaptive!! BUT MSI HAS VID BOOST WITH STATIC VOLTAGES! CANCER CANCER!! why do you want me to keep playing with cancer?

http://forum.notebookreview.com/threads/evga-sc17-1080-finally-available.807895/page-16

READ please!

Another time when I was changing multiple static voltages in Throttlestop (AFTER I knew about IA AC DC loadline), i had a BSOD during stress test, like the voltage was set wrong.
Like when I used XTU to set power limit to 511W (bios limit was 500W) and time "Unlimited" and hit "Apply" and got another BSOD.......

What works for your premabios Clevo LGA powerbook doesn't always work for BGA jokebooks, Master @Papusan

 

If I understand you correct... You change clocks with every boot (testing or as usual to have different clocks?). And you use override Voltage fixed with no offset. If you find a perfect clockspeed you want have in bios and implement it in TS. You can't use or put offset for the other profiles? Cashe don't do very much, maybe it will be easier with fixed Cashe? Or let it be as default.

All too long since I tested stock bios, but I mean Clocks, voltage and power settings worked well with TS and stock bios.

You can't be 100% of the power draw vs voltage and load in notebooks. Not even Intel theyrself have a fully fasit.

 

Yes Master Papusan.
i have 5 bios saved profiles. I just go in Bios, load profile and reboot.
I have 960 pro and 950 evo ssd. Only takes 8 seconds to load windows
Right I use override voltage, no offset. Although I have profile#2 in bios, set for 4200 mhz, ADAPTIVE + offset +105. (no offset is unstable, because i REMOVED MSI VID BOOST with IA AC DC Loadline=1 ).

With IA AC DC Loadline=Auto, 4200 mhz adaptive + 105mv offset is about 1.32v (ESTIMATED BY TEMPS AND POWER DRAW!!!)
With IA AC DC LOADLINE=1, 4200 mhz + adaptive + 105mv offset is 1.18v.

Now you understand this cancer, man?

 

Another problem I had is that when I had multiple Throttlestop profiles to change from adaptive voltage to static voltage (with each profile having an override voltage), if i had CPU Cache slider changed for each profile (to override x36 cache or to override cache -3 bins lower than core), the laptop would take 20 seconds to shut down/reboot. That was annoying...I found it was caused by me overriding the cache slider (e.g. cache and core at exact same speed). like x42 cache x42 core and a different setting on a different profile. (would not happen if this were set in bios this way in "Overclocking performance menu" e.g. min 42, max=42

 

I know the most. But i expected you could change multi in TS and add + or - offset vs what clocks you wanted put without too much struggle. And @Phoenix say this mess works perfect out of the box

Dellienware crippled their Jokes with the new tool who talked with firmware so it couldn't work properly. See their NEW modern useless overclocking controll apps (who don't work properly due crippled power settings). Msi is in same liga if you change a few settings, bruh. When is enough enough?

 

Yes I can change these settings with Throttlestop.
I just set 4.7 ghz with 1.2754v identical to Bios and did a test run. It was fine.
The problems seem to be when I go back and forth with different voltages. Maybe the loadlines are not working properly sometimes? i don't know. Or maybe I simply wasn't stable?
but I do know something else and i have a screenshot:

4.7 ghz, 1.25v, prime 95 small FFT:
GPU idle: CPU=101 watts.
GPU full speed on desktop: CPU 95 watts.

(but temps are the same).
I Think this is more proof that the MXM slot is supplied by the same or similar power delivery system as the CPU.....

 

Last Q before we live the topic. @Phoenix Can you forgive me?

Have you checked if this happen with lower overclock, the way MSI want you to run the notebook (oc'd by MSI and stock).