*** Windows 10 + NVIDIA WHQL Drivers are Killing Alienware and Clevo LCD Panels ***

Sorry guys but I didn't followed this thread for a bit. Do you also need correct EDID values for research?

I can post valid SEC5448 Samsung 184HT values if you want. You mad need this if you got your M18xR1 display destroyed and want to fix bad EDID. Sorry if this is useless

Code:

[HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Enum\DISPLAY\SEC5448\5&18a51dcd&8&UID1048848\Device Parameters]
"EDID"=hex:00,ff,ff,ff,ff,ff,ff,00,4c,a3,48,54,00,00,00,00,00,14,01,04,90,29,\
  17,78,0a,c8,95,9e,57,54,92,26,0f,50,54,00,00,00,01,01,01,01,01,01,01,01,01,\
  01,01,01,01,01,01,01,29,36,80,a0,70,38,1f,40,18,10,25,00,99,e6,10,00,00,1a,\
  1c,24,80,a0,70,38,1f,40,18,10,25,00,99,e6,10,00,00,1a,00,00,00,fe,00,48,47,\
  54,33,4a,80,31,38,34,48,54,0a,20,00,00,00,00,00,00,41,01,9e,00,00,00,00,02,\
  01,0a,20,20,00,a8

 

@Robbo99999 @Matrix Leader
Started up my computer now. Had to turn off my pc 2 times after boot due to black screen after startup. Pretty hopeless. Trying your solution now @Robbo99999 Post a picture of settings. Are these correct?

 

Both are fine regarding the first 128 bytes. However, there's a further 128 byte extension block:

Those might be ok too, only the checksum byte does not apply to those (it validates only the first 128 bytes). To extract those blocks as well you'll have to use ' MonInfo -> File -> Save as', as bvermeul has done. However, this doesn't work on Vista/7; EDID from registry missing 2nd block. Even then; registry is limited to 256 bytes and, hence, so is MonInfo. Have a CRT that has 32 extra blocks, to get these as well you could use ' PowerStrip -> Monitor information -> Options -> Read data directly from monitor'. If it fails ('DDC error'), try softMCCS ( 'File' -> Save EDID as ...').

That would be a good idea . If EMI is correct, then power-on/boot-time is most critical and a more gentle boot process would alleviate that.

Thanks, it's very helpful.

Though it is identical to the SEC5448 in the 7z, there's multiple variants of the same PnP ids in existence. With every edid posted, the data can be more accurately matched to specific panels. There's also the safety of your own system/lcd to consider; it's ok now, but if at one time you extract edid and checksum has failed then it's time to take action.

The more 10+lvds+nvidia edids, the better.

 

ok i have been watching this for a while, gotten scared a few times and almost went back to windows 7 but then saw people talking about it still happening so figured if i might already be screwed then might as well stay where i am. i am running an M17X R4, aftermarket 680m 4gb (my 675 blew a cap a year ago) with i think its bios A10 unlocked. now i have not rebooted in 14 days (windows doesnt seem to want to install updates on its own for some reason but im not going to force it) but i used the web edid parser and with monifo it says my LGD02DA is valid (got invalid once but most likely due to edid not formatted correctly for the webpage). ill include my moninfo txt file for anyone that would like to examine it.

a few other notes that might help with identifying the problem, i am still running optimus mode (intel is still enabled). i always had brightness control from the start (and keep it at 60%, under 7 i ran 80%). had to enable discreet graphics after windows finished installing some drivers (it did install nvidia but then after a reboot they were gone) then use DDU to uninstall what windows put in then reinstall the 355.30 driver (i think, and no modded driver, have updated to 355.60 when it came out) but i deselected 3d vision since i dont have one. i also have a very limited nvidia control panel (has only 3 options). this is most likely more information than is useful but figured i would at least give a few more points that might help.

 

If it helps...here's my EDID from my Alienware 18. 780m SLI running 355.60 drivers and Windows 10 Pro x64

I extracted it using CRU, so don't know if it worked properly.

 

Both are fine and, funny enough, also different from the other LGD02DA and SDC4C48. These two are older revisions, the SDC especially; it looks like a pre-production panel. There's no model/inventory id and, according to its dimension values, it's a 21.7" display (480x270mm):

CRU does not affect the export; it uses edid override values in registry, but does not touch the original values. That's why it doesn't work with Intel; HD hogs the panel and any changes result in failure. Should think flashing OC'ed edid to the eeprom would circumvent that ... might give this a try ... 75Hz for starters. It's not as simple as it sounds; refresh rate is controlled by seven values, not a single **Hz byte. The refresh rate is derived from those seven, hence it's never exactly 60, but merely something close (mine's 59.95). Perhaps that explains why there's 59 and 60 with Intel? One rounded down, the other rounded up.

 

Hey , has your 18 been affected by this EDID "bug" or is windows 10 working fine for you?

 

Yes, all three of the Alienware computers I've upgraded to windows 10 are working fine, but always proceed with caution. I may have just gotten lucky.

Alienware 18 - 4930MX - 780m SLI - Windows 10 Pro x64
Alienware M18xR2 - 3920XM - 680m SLI - Windows 10 Pro x64
Alienware 17 - 4900MQ - 780m - Windows 10 Pro x64

working fine still

 

Here is another one from my working M18xR2 to compare with what @GodlikeRU provided. I used softMCCS for the SEC5448_EDID.bin file (also saved it using the same name with TXT extension).

 

Just make sure you do not switch to Dedicated only..In Optimus you will be fine.. The minute you switch to dedicated everything starts getting screwed up... Use the modded inf J95 provides on his alienware inf mod thread over on techinferno and use 353.00 or 353.06 and don't install 10 yet...

 

few weeks ago I saw on the nvidia forum that people's displays were being broken from a possible nvidia driver/windows 10 combo do you think it is possible this is the reason why yesterday my display in this laptop crapped out? I have had no problems with the laptop and now the display is totally gone/broken its very upsetting. Just thought I would shoot you a message and see if it is happening in other laptops besides alienware, in my np9377sm-a I do have 980m's SLI so not sure if that has anything to do with it or not? So has anyone else reported this problem or is it only a problem with alienware because what happened to me sounds alot like what has happened to people with AW laptops...was getting weird blank screens skipping the boot screen or bios screen then would load windows freeze then restart and work okay and suddenly the screen just stopped working at all....I have filed for an RMA...I might go back to win8.1 for a while after this RMA is complete...

 

is the black screens like as if the screen is off with no backlighting?

 

When you ask now; I don't remember. LOL
But I think off with no backlighting. Going to check if this happens again.

 

Looks like you are the second Clevo owner reporting a dead LCD running Windows 10 and newer NVIDIA drivers. See quote below. We may see more dead laptop LCD panels before this is over, unless the responsible part stops doing whatever they are doing that is causing it. Going back to Windows 8.1 or Windows 7 and using only older NVIDIA drivers is a safer option if you have already erred in judgment (as I did) by installing Windows 10 to begin with. The safest option is to never allow that filthy trojan malware OS to infect your system to begin with. http://forum.notebookreview.com/thr...ware-lcd-panels.779449/page-108#post-10077594

@D2 Ultima

 

Currently using a M17x r5 (?)

I7 4710mq
GTX 880M

This is what I got from Edid:


Vendor/Product Identification:

Monitor Name :
Monitor Serial Number :
Manufacturer Name : LGD
Product Id : 230A
Serial Number : 0
Week Of Manufacture : 0
Year Of Manufacture : 2010
EDIDVersion : V1.4
Number Of Extension Flag : 0

Display parameters:

Video Input Definition : Digital Signal
DFP1X Compatible Interface : False
Max Horizontal Image Size : 380 mm
Max Vertical Image Size : 210 mm
Max Display Size : 17,1 Inches

Power Management and Features:

Standby : Not Supported
Suspend : Not Supported
ActiveOff : Not Supported
Video Input : 1
sRGB Default ColorSpace : False
Default GTF : Not Supported
Prefered Timing Mode : True

Gamma/Color and Etablished Timings:

Display Gamma : 2,2
Red : x = 0,64 - y = 0,335
Green : x = 0,32 - y = 0,63
Blue : x = 0,15 - y = 0,06
White : x = 0,313 - y = 0,329

Etablished Timings :

Display Type : RGB Color Display

Standard Timing:


Preferred Detailed Timing:

Pixel Clock : 140 Mhz

Horizontal Active : 1920 pixels
Horizontal Blanking : 176 pixels
Horizontal Sync Offset : 48 pixels
Horizontal Sync Pulse Width : 32 pixels
Horizontal Border : 0 pixels
Horizontal Size : 382 mm

Vertical Active : 1080 lines
Vertical Blanking : 33 lines
Vertical Sync Offset : 3 lines
Vertical Sync Pulse Width : 5 lines
Vertical Border : 0 lines
Vertical Size : 215 mm

Input Type : Digital Separate
Interlaced : False
VerticalPolarity : False
HorizontalPolarity : False

Detailed Timing #2:

Pixel Clock : 140 Mhz

Horizontal Active : 1920 pixels
Horizontal Blanking : 176 pixels
Horizontal Sync Offset : 48 pixels
Horizontal Sync Pulse Width : 32 pixels
Horizontal Border : 0 pixels
Horizontal Size : 382 mm

Vertical Active : 1080 lines
Vertical Blanking : 33 lines
Vertical Sync Offset : 3 lines
Vertical Sync Pulse Width : 5 lines
Vertical Border : 0 lines
Vertical Size : 215 mm

Input Type : Digital Separate
Interlaced : False
VerticalPolarity : False
HorizontalPolarity : False

Monitor Range Limit:

Maximum Vertical Frequency : 0 Hz
Minimum Vertical Frequency : 0 Hz
Maximum Horizontal Frequency : 0 KHz
Minimum Horizontal Frequency : 0 KHz
Maximum Pixel Clock : 0 MHz

Stereo Display:

Stereo Display : Normal display (no stereo)

RAW Data:

0x00 00 FF FF FF FF FF FF 00 30 E4 DA 02 00 00 00 00
0x10 00 14 01 04 90 26 15 78 0A F1 95 A3 55 52 A1 26
0x20 0F 50 54 00 00 00 01 01 01 01 01 01 01 01 01 01
0x30 01 01 01 01 01 01 B0 36 80 B0 70 38 21 40 30 20
0x40 35 00 7E D7 10 00 00 18 B0 36 80 B0 70 38 21 40
0x50 30 20 35 00 7E D7 10 00 00 18 00 00 00 FE 00 4B
0x60 36 50 4A 31 80 31 37 33 57 46 31 0A 00 00 00 00
0x70 00 00 41 31 9E 00 00 00 00 02 01 0A 20 20 00 7A

 

PS: Got Windows 10 and 355.60 nvidia drivers installed...should I got back to 8 you say? :/

 

@ssj92 , I've been using windows 10 for quite a while now, was part of the technical preview and upgraded to the official version when it launched. I've been using the official version for only a week though before going back to windows 8.1 due to this issue. I haven't experienced any sort of bugs or screen issues while using it. Do you think it's safe for me to upgrade again or should I just wait?
@Chonwey , if you are covered by warranty then it'd be a good idea to stay with windows 10 and report back on any errors found.

 

Have a look at what has happened to those affected and the 1160+ posts discussing what happened, then decide if it's worth rolling the dice on continuing to allow the Windoze OS X digital cancer to reside on your system. It might turn out fine for you, and we all wish you the best. If you ultimately decide to stick with Windows 10 knowing of the potential for a tragic outcome, you will have only yourself to thank if things end badly. We all hope it doesn't. Good luck.

If your hardware survives the next 60 to 90 days, then you will only need to worry about identity theft and how to recover from Micro$oft trojan OS privacy abortion. Ironically, that problem might turn out to be a whole lot more expensive to fix than replacing an LCD panel.

 

Could you point me in the direction of the other people who have had this display problem with a Sager/Clevo laptop? does it seem like this is mostly happening to systems with Nvidia SLI? My started out as I would go away from my laptop then come back and the screen would be off and I couldn't get it back on without restarting, then when I would restart it would completely skip bios/post screen and the display would be completely blacked out until the windows login screen, then I would login to windows and it would instantly freeze then it would, by it self restart and once again have the blacked out screen until the windows login screen, then I would login and it would be fine, then I was in advance start up options and clicked on UEFI restart and from then on the screen would not come on at all. I am so upset about this, I wish I would have known so I could have reverted to windows 8.1 but from this description does this sound similar to the problems everyone is having with AW systems? thanks @Mr. Fox Will you please add my laptop to the affected laptops list here and on the nvidia forum? Thanks

*Edit*

@Mr. Fox Sorry I am editing this post because for some reason it will not let me message your/start a convo with you it says you aren't available, anyways I was wondering if you had a email address for anyone at Nvidia or MS that I should email about this maybe if enough people bug them and email them they will fix this!? thanks

 

@Brent R. you're the second Clevo report I have seen, so other than what I linked for you, no. It's not limited to SLI. Their could be failures we haven't seen posted in our community. It does, in fact, seem limited to systems with NVIDIA GPUs and Windows 10 having been installed seems to be where almost everyone's problems first begin. Yes, yours and @ajc9988 Clevo failures have some very similar attributes... close enough, being they are serious display-related problems, to suggest the "issues" are not limited strictly to Alienware laptops even if the symptoms are not precisely identical.

 

So Win 10 seems to be the common culprit... I have an uptime of 4 days and 17 hours right now and I've gotten no issues in any games or with overclocks or anything with 355.60, but I'm on Win 8.1 and I have none of the Win 10 updates installed. If I see a single crash I'll instantly go back to 353.06 which I know is stable.

@Brent R. have you been using the 120Hz panel or the LVDS panel?

I want to find out some things CLEARLY:

1 - Has there been *ANY* instances of a 120Hz or otherwise-eDP panel being killed? I have only seen LVDS (which is common). Ajc has the P770ZM, so unless he got it from Eurocom with the IPS panel, he's using LVDS.
2 - Has there been ANY instances of any machines who have *NEVER* touched Windows 10 getting their screen EDIDs corrupted?
3 - Has ANYONE managed to get back a screen into an affected laptop and NOT touch Windows 10 at all but simply switch to Win 7 or 8 and STILL have the screen die on these newer drivers? I.E. is the problem BIOS-deep, and Win 10 alters the sBIOS in such a way that the screen EDIDs can be corrupted without a fresh BIOS flash regardless of OS as long as it happened once before?

 

Yes this happened to me with my 60hz LVDS display with Windows 10 and Nvidia drivers 353.63, I am also interested in learning how to revert back to windows 8.1 with out win10 corrupting my display firmware or bios once I get my laptop back from RMA?

 

Have some in this forum asked xoticpc reps about the screen problems? Maybe they know a little bit of this huge problem?

 

Yes, there are two unmistakable common elements in all verified LCD failures with EDID corruption. Those are (1) initially caused by Windows 10; and (2) only on systems running NVIDIA Kepler and Maxwell GPUs.
1 - yes, one Alienware M17xR4 owned by @andrewsi2012
2 - none that we have identified conclusively, but one or two that appear to have been fully updated and Windows 10 ready may have
3 - anyone willing to run newer GeForce drivers after having to replace their LCD is stinking nuts to do so until we have proof of the cause and fix

I am no longer risking damage to my Clevo P570WM (or my identity being stolen) by running Windows 10. As far as how deep Windows 10 goes with its digital cancer, I'd say clear down to the bone. During Windows 10 setup, at every automated reboot my P570WM opened to the BIOS main menu without my intervention. I had to press F4 to save and exit the BIOS for Setup to continue. I installed Windows 10 with my CPU using my customized default settings with the CPU overclocked and overvolted. After Windows Setup completed, it happened a number of times when Windows 10 decided that it did not like my CPU overclocks. It had NEVER done that before Windows 10 and has NEVER done it again since I eliminated Windows 10. So, you decide... I think it is demonic behavior, and as a matter of principle I don't want the Micro$haft Nazis messing around with my life, my computer or anything else. They totally suck as a company and I now hold nothing except contempt for them.

Here's what I did and my M18xR2 has stayed fixed and working as if Windows 10 had never been invented and allowed to ruin a tiny part of my life.

• Flash your system BIOS
• Flash your video BIOS
• Diskpart clean, then secure erase the drive Windows 10 was installed on
• Avoid using any GeForce driver version that was available for Windows 10, but most especially 353.62 or newer

 

Well if it's a M17x R4 using a regular eDP panel? Or a 120Hz panel? There are only four 17" 120Hz panels that I can find on www.panelook.com and all of them are by LG, and all discontinued by late 2011. If it's under 1080p, then it might be LVDS. I'm just being as thorough as I can be.

This is a very critical point. If Windows 10 touching a system is the catalyst, then while nVidia's drivers may be at fault for doing something they shouldn't, Microsoft is more at fault for building an OS capable of modifying a system BIOS on its own. Not every system has had issues, as there's another user just a page or two ago who said all his Alienwares are working fine, so even if I sit forever and never encounter problems until this system gives up the goat in 2018 or something, it might just be luck. Your Clevo hasn't shown a single inkling of dying either, and if I'm not mistaken you're also on a 60Hz LVDS panel. If it's BIOS level catalyzed by Win 10, then your P570WM is in the strike zone.

I wholeheartedly agree. I'm basically asking for an out-of-pocket blood sacrifice here when I couldn't even pay to get a new screen shipped directly to me myself, far less ask someone to sacrifice a new screen. But this is a critical point of testing that we need to pass, and apparently nVidia is more incompetent than we.

Even if all I'm good for is process of elimination and rationalizing data, they're a multiBILLION dollar corporation that can't figure out why users were getting TDRs a month and a half ago, far less determine why laptop screens are being rendered useless by THEIR OWN DRIVERS.

 

All Alienware M17xR3/R4/17 laptops with 120Hz 3D capable display panels are eDP and 1920x1080 resolution. These are the only eDP models ever sold by Alienware to the best of my knowledge. So, yes. One reported failure by @andrewsi2012 as I mentioned. Everything else they have ever sold is LVDS as far as I know. This is at least true for the models that have had their LCD panels killed by EDID corruption.

Until we know more about the exact cause, I think everyone with NVIDIA Kepler or Maxwell GPUs is in the strike zone for something to go wrong or get damaged. It may require a specific set of conditions to be present, but we don't know what they are apart from a direct correlation to Windows 10 and NVIDIA seeming to be a deadly LCD-killing combination. It's hard to avoid a specific set of conditions when you don't know everything that needs to be avoided, so avoiding Windows 10 and GeForce driver version numbers offering support for Windows 10, and all Windows Updates needed to prepare a system to receive a Windows 10 upgrade, seems to be the safest option in my mind.

My Clevo had little signs (as already mentioned) that had me worried it might be at risk of being damaged. The display was brighter than it ever had been with Windows 7 or 8.1, and display scaling frequently changed from 100% to 120% without my intervention; and, both of those were symptoms I noticed with my M18xR2 and 18 before their LCD panels died.

 

Ok. That means eDP isn't safe. We can rule that out of the equation.

@Brent R. and @ajc9988 were either of you using a Prema sBIOS mod for your laptops?

 

Until the media runs a story on this; we are likely to get nowhere and look more & more like dogs chasing our own tails. Without people being informed of this on a much larger scale; we will never know just how many have been affected. Even if we manage to discover a reasonably 'easy going fix' - that would only solve 30% of the problem here. Until we wake up and take legal action, we will always be defeated & allow these big corporations to just take a dump on us. Channeling our energy in the correct direction (outside the box) can prove to be far more affective & useful.


Sent from my iPhone using Tapatalk

 

Updated thread title:

*** Windows 10 + NVIDIA WHQL Drivers are Killing Alienware (and now maybe Clevo) LCD Panels ***​

 

The problem is, other than Intellectual Property, the W10 EULA prevents law suits. You would need a lawyer willing to fight that battle first.

 

Nope I was not using any form of Prema Bios mod or vBios mod, and everything was at stock settings and speeds.

 

If we were to take our cars to gas stations & they sold us gas that seized our engines; we would sue. This is no different. You cannot offer / sell something to someone that has not been fully tested to not cause damage to personal property. Damaging personal property is a crime. This is not difficult for a lawyer to disclose & establish affectively in the court of law.


Sent from my iPhone using Tapatalk

 

Yeah... except that "installing your "guaranteed safe" intellectual property on our machines has opened them up to hardware malfunction through no fault of our own" is a pretty compelling arguement.

Unless somewhere in the Windows 10 ToS it specifies it can have access to and modify your firmware without your consent OR KNOWLEDGE, then MS isn't safe from this.

 

You have to have this hurdle overcome first.
https://weigoba.wordpress.com/2015/08/18/windows-10-eula-cant-sue-except-for-ip/

Edit; They are not saying no to compensation but to make it individually and quietly go away.

They can, in good faith, disable unauthorized hardware. Still they would have an uphill battle here. Just for the bad press of a looming suit I am sure they would settle out individual claims through the arbiter.

Edit 2; I am playing devils advocate here, not that I agree with them at all!

 

It is illegal to damage other people's property. They can write whatever crap they want to, but a good lawyer can get past that junk without breaking a sweat provided that there is sufficient evidence that proves clearly the OS is causing / enabling the damage.

We just need to reach a clear conclusion as to who is responsible for this, Nvidia with their drivers, or Microsoft with their OS. One of them wrote code that destroys personal property. Intentional or not, is irrelevant. Compensation for damage caused is mandatory along with an apology. I could care less about fixing their mess for them, my concern is more so with making sure they don't do it again to anyone and also that anti-virus companies can come up with affective monitoring software that will safe guard our firmware from any future corruption. Media exposure (after narrowing down exactly who is responsible --MS or Nvidia) will get us on the right track.


Sent from my iPhone using Tapatalk

 

What version of the NVIDIA drivers is the best? I'm on 353.62 with no issues, should I update to 355.80?

Who is this most likely to affect, eDP or LVDS displays, 60hz or 120hz displays, SLI or non-SLI, etc?

 

Hahahhahaha oh god, this is great. Microsoft made everybody agree that they could do ANYTHING that doesn't endanger intellectual properties and never be sued.

That kind of EULA should be against the law. In fact, I think it is, in the UK. I wonder if the ToS is different in the UK.

 

You hear all the time of undisclosed settlements and no admission of guilt or wrong doing and even NDA's. This is just a fact of life of the courts. As far as the fact it must be settled by an arbiter, again this hurdle has to be overcome by a court first to even file the suit.

You have entered into a contract to settle all issues by a federal civil arbiter. This would be brought to a judge who would then, per your contract, drop the suit and refer you to your contractual compliance. This then could be appealed but that is the rub. How much time and effort to overcome this along with expense and the over all possibility you may not even get to a suit.

The idea again is to make a public suit tough if not impossible, to just have it go away as quiet as possible. They are not denying compensation just the manner of how it is to be determined. Keeping this in a legal private federal arbitration procedure but do not count on an apology though.

Edit; As far as international, my guess is this would have to change according to the countries applicable civil laws.

 

The Win 10 EULA only mentions "Class Actions" and only in USA.

Micro$haft probably bought a few hundred lobbyist in Washington and got that pushed through.

Absolutely no way that kind of law would pass in AU

 

@GodlikeRU Well win10 and nvidia cause my display to fail in my laptop so I don't believe it is limited to Dell/Alienware anymore so the whole Dell's display firmware iis unrestricted might have to be looked at again unless my laptop NP9377SM-A which has 60hz LVDS display and the brand is some weird name I forget exactly what it is...anyways I wish there was a simple fix for this...

 

You have a Chi Mei panel. The only panels I remember being sold were a Chi Mei 72% NTSC matte, an AUO 90% NTSC glossy, and a LG 120Hz 72% NTSC matte (Eurocom has a glossy version of this too).

 

Yah thats it, I have the Chi Mei, Won't how long this RMA process will take? And I wonder if I provide a display (my 120hz edp panel) and the cable if they will install that and by doing that they could save themselves a little money by not having to get a 60hz panel and install it....I just hope this process goes smoothly and quickly...never expected this to happen on a machine that costs over 3.5 grand!!!

 

If you send them the panel and cable you'd probably still have to pay for installation... and they'd still have to give you back a new Chi Mei. And your 120Hz panel wouldn't be under warranty either.

 

Yes I know it wouldn't be under warranty, I really hope they wouldn't make me pay for the installation, I mean I shouldn't even be in this situation, I know lots of people buy and spend a lot of money with them, but if they aren't willing to work with me on that then I am not sure if I could recommend them to anyone or buy from them again myself, I mean over $3500 and the display goes out then IF they want to charge me installation for something they would have to do anyways if they were to put a new 60hz display in, plus by not having to use any of their hardware I would actually save them a little money.....but w/e this is all speculation right now until I talk to them more about this, and I am sure Xotic will be there to talk to Sager on my behalf....do you know how long RMA's usually take? 2 weeks? 4 weeks?

 

Has anyone successfully claimed under warranty to fix this issue?

I've still to identify what caused my 17 R2 to suddenly stop working but are getting it repaired regardless in home this coming week.

 

Quote from J95 post in:
http://forum.techinferno.com/nvidia...-driver-support-modded-inf-70.html#post143122

NVIDIA's Windows driver version 355.78 (released 8-22-15) provides beta support for VR developers using Oculus 0.7 SDK.
It provides a bug fix for an intermittent issue found in the publicly released 355.60 Game Ready driver as well as solving an issue with incorrect display timings.

I was under impression nvidia fixed whatever was causing EDID corruption from earlier 355.xx drivers with 355.78 version and that all the fixes were also present in 355.83. This is not the case apparently since 355.83 on Windows 8.1 update 1 still killed the screen.

 

Did you flash your system BIOS and erase your install drive before putting on Win 8.1? If you can get your screen working again (since I think you seem to be able to fix them) can you do that prep and test for us if your screen still dies? I know this is a whole lot to ask.

 

I did not flash the BIOS or vBIOS for 980m, but did erase the SSD completely before installing Windows 8.1. Next time, I will be flashing both bios and vbios.

 

Your testing will prove things once and for all. If you flash your BIOS fresh and your vBIOS fresh then install your Windows copy fresh and do not install the win 10 updates, and your screen never dies on the problem drivers, that concretely proves it's Windows 10 facilitating the ordeal. Because almost every single person I've seen here has had windows 10 touch the machine at some point. Mr. Fox is way too safe to let that stuff happen to him again and I can't ask him to sacrifice a screen again since he simply has to buy new ones.

 

Mr Fox, I plan on moving my 970M from the R4 to the P771ZM.. I don't know exactly how to deal with this problem. I will of course Diskpart and secure erase the OS drive, however for the vBIOS flashing, should I do it on my R4 or on the Clevo? Also how does vBIOS flashing work on Clevo? Like Prema has the automated files however do you need to disable the driver before flashing the 970M like I used to do on the AW? How does vBIOS flashing of 980M work on your P570?

I don't think it's a vBIOS/BIOS flash problem... The driver is the issue... His screen was working fine after he fixed it bu the minute he installed these newer drivers, it broke.. There is something fundamentally wrong in the drivers....