MSI Laptops Drivers

Sorry, I'll try this later.
I've been up all night messing with this stuff.
Going back to firmware .313, trying to use FPT in DOS, listening to the great Prema in PM, everything

If I may ask, why do you thikn that disabling overclocking, enabling fast boot and re-enabling the ME will somehow fix this? sorry but i have a huge headache from sleep deprivation and I was scared of bricking my laptop and making @Papusan even more angry

 

Mind you, My son bought MSI. And he was happy to have a Papa (→Papusan)

Not angry, just sad.

 

Because when I asked Bug Bunny from MSI why is the Fast Boot feature not sticking as enabled when I set it to enabled, he said it's because I enabled overclocking and thus the overclocking parameters have to be re-initialized every time before the taptop starts to boot. I am not saying that this will fix it just try, nothing to lose, maybe it's not saving the setting to enable it due to some reason so try with Overclocking feature completely disabled, then enable it and see if it sticks.

 

I did what you said and fast boot works, but its as slow as with fast boot disabled
ME is still "temporarily disabled".

I'm telling you, it's a bug. It is some bug with spread spectrum at 0% causing a warm reset, because I get this exact same 'fast' rest (like 1 second power off and on) when I disable the ME or "Attempt" to re-enable it. I'd bet $100 it is something at LEAST Loosely related to that. Or maybe its a bug and the failed reenable forces spread spectrum back to 0% ? i dont know. my head hurts.

But I no longer have access to the spread spectrum in Intel ICC anymore. But I sure do NOT remember this happening if that spread spectrum was at default way back. But at default, the BCLK was 99.750, which everyone hates...So...
Whether this is the 11.8.50.3425 ME's fault, MSI's fault or both, doesn't matter. Something's broken.

Only way out of this mess is to somehow reflash the ME. Which the MSI UEFI updater won't do. It won't downgrade it. FPT in dos gives CPU permission descriptor error, same as FPTw64 in windows, etc.

I'm convinced a forced reflash of an older ME firmware or even the same one over itself would 'fix this' but I dont know how to do that without trying that creepy "Setup_Var 0xE9A1 0x01 (forgot the exact values) command at an efi boot prompt to try to 'gain' CPU write permission.

 

Yeah... as I said before, things like this mess with my sense of wellbeing

 

yeah me too, this is pretty serious as it's a freakin' BIOS update so all I think is hmmm, what was so screwed up in the previous BIOS then? and what about all those people that did update and don't know that they pulled the update, replaced it with an older version, but different version than what they had before....

 

Phoenix, this is MEMANUF -verbose.
As you can see, the ME status isnt corrupted (apparently) but is "temporarily" disabled and won't re-enable.

memanuf -verbose:

----------
CurrentState: Disabled
ManufacturingMode: Disabled
FlashPartition: Valid
OperationalState: Transitioning
InitComplete: Initializing
BUPLoadState: Success
ErrorCode: Disabled
ModeOfOperation: Temporary Disable mode
SPI Flash Log: Not Present
Phase: BringUp
ICC: Valid OEM data, ICC programmed
ME File System Corrupted: No
PhaseStatus: UNKNOWN
FPF and ME Config Status: Match


Error 86: Communication error between application and Intel(R) ME module (FWU client)

Error 125: Internal error (Could not determine FW features information)

Error 117: MEManuf Operation Failed

 

The bios update is fine.
It even allows you to change the DDR RAM multiplier without using the "MSI Hidden Fun" unlock combo.

HOW DO I KNOW IT'S CALLED MSIHIDDENFUN?
@Papusan
By looking in UEFITOOL above the Setup region...It may be a 4 finger salute but that's what msi "calls" it.

They pulled it because Intel asked everyone to remove that buggy ucode CPU microdcode update that is crashing people's systems. That's why. the 31E version didnt have that ucode yet.
BTW they never even had the .31E up for download even when it was ready. They like skipped it for .320.

 

very nice analysis Mr.

Yes they never had 31E before I was wondering where the heck that came from

 

Of course, old news. You have probably seen awful performing AW systems after Dell's engineers play the firmware game and put in more than needed.+ bricks due re-flash http://forum.notebookreview.com/threads/new-bios-for-15r2-17r3-v1-4-4.812964/page-13#post-10669826

 

Good job educating the uninformed. It's a tough job, but you're great at it.

I liked the look of the Steelseries keyboard on the 16L13, but favored typing on the Clevo. Chicklet keyboards are not my favorite. They just seem too flat and weird to me. I also disliked how it is made into the palmrest with holes for the keys to poke through and not easily removable. That was not a particularly intelligent engineering move from my perspective. I did not find typing to be any more or less pleasant, so from a functional view it was neither a plus nor a minus. All things considered, I viewed that as one of the weaknesses of the 16L13 based on undesirable design only. But, it was really pretty to look at compared to my Clevo keyboards. The LED lighting was better (more vibrant). My Steelseries Sensei wireless mouse is definitely awesome. Best mouse I have ever owned.

 

I prefer typing on my MSI versus the Clevos I've had previously. Yet I completely agree that it sucks how you pretty much have to completely disassemble the laptop in order to get to the top side of the motherboard when on a Clevo you can easily remove the keyboard assembly by itself. That's why I ordered 4x8 GB of RAM from the outset so I wouldn't have to worry about it down the road. The new Clevo P751TM1 is an intriguing proposition - Coffee Lake, an available GTX 1080 (when ordered from Eurocom right now), and two 2.5-inch SATA hard drive bays on top of two M.2 PCI-E ones. It's ever-so-slightly smaller all around than the 16L13, though it weighs about a pound more.

 

Intel Rapid Storage Technology and RAID driver downgraded to v15.2.0.1020 due to stability issues with every IRST Driver after that like random system freezes

 

New ain't always better Why always go for new if you can't see any benefits? This on a fully working computer I have always wondered why people run into screwed up systems

 

Because the new one says it has native NVMe PCIe support so I expected it to be better

 

Yeah, exactly All that glitters is not gold - Wikipedia

All that glisters is not gold—
Often have you heard that told.
Many a man his life hath sold
But my outside to behold.
Gilded tombs do worms enfold.
Had you been as wise as bold,
Young in limbs, in judgment old,
Your answer had not been inscrolled
Fare you well. Your suit is cold—
Cold, indeed, and labor lost.

 

Attempting a Windows 7 install now.... for the 3rd time...let's see if this works....

 

Exactly. Never buy from the Phoenix

 

Sometimes you won't know until you try it. What if it works 200% better?

 

BIOS Version E17A1IMS.320 for GT73VR 7RF was deleted from the website... WHY?

 

Because MSI

 

The cat ate it.

 

Are you have a problem with this bios version? Because I installed the bios

 

Please, read!! forum.notebookreview.com/threads/cpu-vulnerabilities-meltdown-and-spectre-kernel-page-table-isolation-patches
Long thread but I'm sure you finally can see the pict The whole web is full of same topic.

Edit. Or I can make it easy for j00 two
Stop installing our buggy Spectre CPU firmware fixes, Intel says

 

@Vasudev

To enable Write Back Cache, you need to:

1) Open IRST > Performance tab
2) Set Link Power Management to Disabled
3) Set Write-Cache Buffer Flushing to Disabled
4) Now the Write Back Cache will be ungreyed and you can enable it, after you do that, reboot your taptop

 

I have only Link power mgmt option in IRST.

 

is your RAID Array a RAID 0 Array?

What SSDs are you RAIDing and what is your IRST version and favorite food?

 

Was it for RAID arrays only?
I'm using AHCI.

 

Yes of course, that's a RAID only feature

For AHCI, just use PrimoCache (you have to pay to play though)

 

I've read it but I can not really understand it. so it's better if I re-install the bios of previous versions?
Thanks

 

Yes.

 

I got those settings in Device Manager under Disk section.
If I enable them, I will have worse battery life. An hour of reduced battery runtime. I will try testing battery runtime on Write Through Cache feature set.
My PC eats 230V, 50/60Hz as its food. Peak wattage is at 200-210W without GPU OC. With OC, it can hit 220W-225W thanks to undervolting.

 

nVIDIA GeForce Driver updated to v390.77

 

The release notes explains why I had ****ty scores on 3dmark benchmarks.

 

Why do you need to read the last Release Notes to get an explanation that something is wrong?

 

I was on driver 390.65 and had lower 3dmark scores. Reverting back to old driver fixed it. As always, newer isn't a whole lot better these days!

 

@Mr. Fox @Papusan @Falkentyne

As you guys may have known, any IRST Driver after 15.2.0.1020 causes my system to randomly freeze so I had to stick with the old driver.

Today Samsung released a new firmware (4B6QCXP7) but only available through Samsung Magician so I am installing Windows on an external USB Flash Drive now using WinToUSB and will try to flash the firmware since Magician can't recognize my SSDs in RAID mode.
Stay tuned....

Phoenix >>>

 

And they never heard from him again.....

 

FTFY

(Tales of Innocence).

 

I get 6878 TimeSpy 16025 FireStrike. So I don't see a problem. I need to check old values.

 

Ok so here's the story:

1) First attempt was to install Windows onto my Lexar 128GB USB Flash disk which is like one of the fastest Flash Disks you can get, even though its fast, Windows was very slow and laggy to the point where I couldn't even use it so 2 hours wasted there

2) Next, I deleted my RAID Array, installed Windows in AHCI mode, installed Samsung Magician and what do you know! It said I was already on the latest Firmware I was so frustrated after having had deleted my RAID Array and wasting all this time! I felt I was stuck.

3) Then an idea popped into my head I installed hide.me (my VPN) and logged into to the USA East Coast server then relaunched Samsung Magician and what do you know! I was immediately offered the firmware update!

3) After updating both my SSDs, I re-created my RAID Array and reinstalled Windows.

4) Unfortunately, the latest IRST Freezing bug when any excessive writes are happening was still there so this turned out to be an IRST problem not the firmware problem but at least we ironed out the issue now and we know it's the latest IRST which support Optane, all of the recent one have this problem. The problem doesn't happen if you don't have Write Cache Back enabled, but then if you can't enable it properly then what's the point so I reinstalled the old RAID Driver 15.2.0.1020 and all is good now.

Benchmarks were roughly the same, higher in some areas and lower in some areas but that's margin of error but this is a safe update.





oh and @Papusan I was finally able to reveal the secret , since I installed Magician in AHCI mode, I saw my SSD usage....

Morale of the story: Be brave...be like Phoenix...

 

Amazing performance!!

 

Your only gain.... j00 saw your SSD usage And what areas got decreased aka lower performance? 4K ?

after increased performance in the dark seems being wasted time

 

IRST freezing bug on RAID 0 with write cache back enable is on 15.5, 15.7. 15.0, 16.0 beta... From almost a year right now.... when will Intel fix this recurring and critical error? I have the same problem...

 

Thanks for confirming it's not just me. any version above 15.2.0.1020 has that bug. What's weird is the newer IRST Driver are actually made with NVMe PCIe SSDs in mind. Classical Intel Swiss Cheese

 

THIS is what INTEL use it's precious/valuable time on nowadays. As the new Win 10 Services

 

Sorry for thread jacking but since i got a similiar freeze on AHCI mode on my HDD, what kind of freeze is it? Temporary or it hangs up the machine completely?

 

in my PC, where I have one nvme Samsung 960 Pro and 2 SSD in RAID 0, even with IRST 15.2 driver, i have another kind of bug, To completly shutdown windows it takes around 30-40 seconds every time. If I downgrande to 14.x, for example IRST 14.8, I dont have this problem. I'm on X299 platform. Strangely the 14.8 IRST driver is from 4/2017 and IRST 15.2 from 9/2016. 14.8 is newer then 15.2 and works OK in RAID 0 even when I do the shutdown on Windows, it does instantaneously. Soo many critical bugs on IRST drivers ((((

 

It hangs the machine

 

Any reason you have the 960 Pro in RAID, other than doubling the storage space on one drive letter? The performance gain doesn't seem much higher than the stock speeds due to bandwith limitation of the DMI.