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

It will be on the first page of Chrome bookmarks . Are normally very very careful with Windows update patches. None of these patches you said is installed. Is selective in terms of what I install of this junk from Micro$oft .

 

Windows 10 is not the best thing since sliced bread, and it's clearly horrendous that a combination of NVidia Drivers & Windows 10 has killed the LCD screens of certain Alienware models, but it is not the hell-on-earth OS that you describe. I'm just putting this post here for some balance against your negativity towards it. I think the privacy issues are a little overblown too, and for me the OS works well, I'm staying with it - there's things that could be better, like a slight increase in CPU performance required to meet the standards of Windows 7, but otherwise I like it. Unforgivable that it's killing some Alienware laptop screens, they need to sort that out urgently big time!

 

Well...
1. After first Activation of Win10 camera turns on and Win10 sends 35MB of data.
2. Microphone in Win10 can get enabled when it wants to.
3. All you text in Win10 is monitored and sent to MS.
4. Geolocation data to find where you live.
5. Telemetry.
6. You can't disable it through hosts file because all addresses is written in binaries.
All these info according to http://liberatum.ru/e/windows-10-spy-analysis which is according to http://aeronet.cz/news/analyza-wind...rmaci-o-uzivateli-jeho-prstech-ocich-a-hlasu/

 

hi, i didnt read all in the thread coz its too long.. LOL.. just wanted to know which nvdia driver is causing the LCD problems. thanks.

 

Right now it seems all are, nothing is safe right now.

 

I am currently using 353.62 and so far no issues. i saw an update today which is version 355.60 and most people have problems with it. good thing i checked forums 1st before updating. can you specify a version causing LCD problems? is it the previous of 353.62 or it is 353.62? thanks

 

It is all versions of Windows 10 specific drivers for Nvidia cards in Alienware M18X R1, R2, and 18's. Some Alienware 17" models are affected, but it appears that only those that are running 60 Hz LCD panels. A single Winodws 7 user and a single Windows 8.1 user have had LCD failure using 353.62 drivers as well.

 

Thank you very much.

 

Also had this happen to me two weeks ago on an M18x R2 with dual 675ms. The GPUs had to be replaced. The LCD appeared to be burned out, but I was able to flash the A03 BIOS (Take out battery and AC adapter, hold end while plugging adapter back in with a USB loaded with extracted A03 BIOS file in the eSATA/USB port) and restore booting for a short enough period of time to switch to integrated graphics, which ran unaffected.

 

Welcome to our community. Thank you for posting.

You are very fortunate. I tried that process on the M18xR1 and R2 and it did not work. I'm glad it worked for you and others should at least attempt it to get their machines running again.

 

Hi @Mr Fox... Take a look at the post I wrote.
Can I which has Aw17R1 and 120hz 3D display also get the same problems with Windows 10 (problems with Nvidia drivers or other drivers), like many others with 60hz screen? Has already downloaded Windows 10 iso file. I'm sitting on the fence . 4 weeks computer free vacation, so I'm not as up to date.
http://forum.notebookreview.com/thr...alienware-owners.779449/page-60#post-10066352

Edit: As most people know, I'm already on Win 8 . Have no Win 7 license so what is the alternative for me?
See what he's writing about the temperature change in Win 10... CPU and GPU temps drop 10% with Win 10 http://forum.notebookreview.com/thr...s-lounge-thread.721578/page-539#post-10053902

 

I'm run unaffected Win10 x64 for more than a 2 week 24/7 at my AW18, LCD refresh rate set to 96Hz, GTX 980M SLI, inf for this video card modded by me. Then reinstalling Win8.1 and wait safe solutions.

 

This forums: http://forum.overclockers.ua/viewtopic.php?p=1677630#p1677630 have some info about this problem, nVidia drivers wipes EDID in display, then people successfully reflashed and fully restored LCD.

 

Welcome to our community. Did you have to re-flash your EDID information?

So far, no reports of issues with 120Hz display on Alienware that I know of. Windows 10 performance in some areas is worse than Windows 8. See the link in my signature for my review and benchmark frenzy.

Until DX12 has been around long enough to matter, there is honestly no reason whatsoever for anyone to feel any level of compulsion to install Windows 10 if they are looking for improved performance... it's just not there at this time. The UI is definitely better than Windows 8/8.1, but the OS itself is unremarkable in every other aspect... a big to do about nothing.

The part that sucks most is how we lose control of too many things and Micro$lop assumes control of things they have no business being in control of. With Windoze OS X they have new abilities for harvesting data in areas that they need to keep their filthy noses out of. The security risks might be overstated, but how much is too much? The OS is a designed ground-up to be a data collection bot, plain and simple. Why, for the love of God, would the world be OK with this? Has stupidity become so common that we don't care? Is it really free? I don't think so. It only appears to be. They are doing what is best for themselves... and for those don't have a clue yet, we are nothing but a data mining source and revenue-generation stream in their conquest.

 

No, I dont flash my LCD. He run correctly all this time.

 

Dunno if already posted, but seems Aw 18 R2 is not supported: http://www.dell.com/support/article/us/en/19/SLN297954/EN#Alienware-NB

 

I am assuming you mean the M18xR1 and M18xR2, as there is no such thing as an 18 R2.

Unfortunately, the Alienware 18 is on the list of "supported" (meaning they tested it, nothing more than that,) and there are more reports of the Alienware 18 with Windows 10 failures than any other model. You can take lists such as this with a grain of salt because they are worthless. They are only an indication of what they tested, and in this case the testing was not sufficient to identify the issue many are having.

Their support is not required in terms of functionality. All of the upgraded GPUs running in thousands of Alienware laptops are also "not supported" even though they work as good or better than the original equipment GPUs. All it really means in practical application is "don't call us for help" if you have problems with X because we don't support X.

 

Sorry Mr.Fox, you right. The lists is only for factory configurations.

 

But given that the AW18 is a supported device and has these problems, the company is willing to help out. It's not so bad, and it's something. Consequently, Alienware indirectly aids those with un-supported products. Outside of these problems, your statements stand.
Given that Dell is still clearing their AW18 stock for the whatever-time, this is something they do have to pay attention, or it'll be more trouble. Again.

 

Mr. Fox
Thanks for the link. Maybe there may come a performance gain after a while? http://www.notebookcheck.net/Windows-10-gets-its-first-major-update.147506.0.html I still hate Win 8 but want cpu power. I'm interested in the cpu score i your reviews of Win 10. Seeing that Cinebench 11.5 and R15 score with Win 10 is the same or slightly better than Win 8. Wprime 32 have 0.81% higher scores in Win 10. Wprime 1024 has 0.37% lower scores in Win 10. Fire Strike physics has 0.91% lower scores in Win 10. Sky Diver physics has 1.76% lower scores in Win 10. 3DMark 11 physics has 9.76% lower scores in Win 10. 3 Dmark vantage cpu has 0.05% lower scores in Win 10. Total Slightly lower CPU score in nearly all cpu tests. 3DMark 11 Physics score in Win 10 have the greatest difference in CPU score.
Edit: If this is correct as one has tested this; The CPU and GPU temps drop 10% with Win 10 is something I like .

 

Can someone enlighten me please how a driver or Windows itself can physically break a hardware like a LCD monitor? How would that be possible. I dont believe this is the case, it is just a driver or uefi problem. Why? Because I had similar problem yesterday on my desktop... Nearly excatly like you had it with your laptop:

1. installed new gpu I got GTX970, everything worked fine, for abour 24h, then I did a final last reboot, and PC went into black screen.
2. wasnt able to get black screen away anymore, reset, power off, nothing worked, thought my PSU broke but no, it didnt
3. then I did a bios reset and got picture again... and I was able to debug what was causing the ptoblem, because I just went into the bios, and did all my optimized settings, saved, aaand black screen again
4. after some toggle each setting I did, I was able to file the problem to fast boot/secure boot
5. bios reset, let CSM on (no fast boot), and pc working since then

I have absolutly no idea why or how this happens, because like I said... the PC totally worked fine, rebooted, powered off, everything, for abour 24h, then bam, broken, and fast boot doesnt work anymore. Also PC would beep and do on off loop after ding resets.

Maybe this could be related to the problem here too, the sad thing is, you cannot turn off fast boot and turn on CSM on most laptop bios. And you are stuck.

Also there is no one "black screen issue", I know of at least four totally different bugs which just manifest as a black screen. Most important: There can be black screen bug right after powering on the PC (no bios logo, no oem logo, no windows logo), or, like in the video, after windows loading screen.

 

Its writes something into the firmware of different devices, ( I am guessing here) , but if your careful and use the right tested drivers should be okay.
Its not intuitive though, theres like no way to just GUESS that doing things properly would cause an issue. But this is most certainly because nVidia or MS have been neglecting something somewhere.
I shipped out thousands of graphics cards, and I have had only a single user that bought from me, that had an issue. So, it doesn't affect many people.
I never recommend the latest drivers. I am in fact still telling people to use 344.75 drivers and 352.84 or 353.17 , the last two for different setups, if they have games like DragonAge or The Wicther3 etc..(newest games)

 

Sadly, there are no answers yet, only lots of questions. If we knew exactly what was causing all of these performance and functionality issues for some laptops and even desktops there would be less of a need for threads like this one. All we would need to do is simply avoid or fix the problem. Unfortunately, none of us have a crystal ball and it is not clear why or what needs to be done to avoid or remedy the situation. Other than one Windows 7 and one Windows 8.1 report that we know of, (which might be hardware failures of a different origin,) all the other Alienware owners having the 8-beep/no boot problem are having it after installing Windows 10. We have only speculation now until the engineers at NVIDIA, Micro$haft and Dell/Alienware determine what is causing these things to happen.

Circumstantially, it would appear that (a) Windows 10 or Windows Updates to facilitate Windows 10 are allowing something to happen that should not be allowed; or, (b) the more recent NVIDIA drivers include some changes that are simply not kosher; or, (c) a combination of both is having deadly consequences for some models because they are affected differently than the rest. We already know (b) to be fact because of the 780M SLI throttling problems that exist under Windows 7 and 8 using drivers released after 345.20, and anecdotal observation suggests that perhaps changes in GeForce drivers that are geared toward Windows 10 support are doing multiple naughty things. Based on timeline, around February and March to be more specific, it certainly appears that a lot of the current drama involving desktop NVIDIA GPU malfunction (like the black screen issue mentioned) started getting the most out of control as Windows 10 support became a high priority task for NVIDIA.

 

Maybe you also believe that there is no virus which can destroy hardware?

 

You must be referring to Windows 10.

But, that's OK because it's free, like all other viruses. No extra charge for the damage... on the house.

 

@maffle - All of the Alienware SLI machines have manual graphics switching, and the bulk of the machines affected are those models.

Those with SLI do not have any access to the Intel iGPU while using discrete graphics. You can only use one or the other, and the graphics subsystem which not in use is totally disabled, not even visible in the BIOS. On a single GPU laptop (excluding those with 120Hz eDP 3D capable displays) that use Optimus or Enduro the Intel and discrete GPU are both visible and accessible at the same time. The pre-Haswell Alienware machines with an unlocked BIOS can have the Intel GPU manually disabled, (an awesome feature to have,) but most laptops cannot make this claim to fame, including the Haswell-powered Alienware laptops with crippled BIOS and flash protection. Those folks are screwed as far as having proper control over their systems is concerned.

 

Thanks, very helpful .

Yes, you have the same screen as Daniel's (LTM184HL01). Yet your registry export was fine, so it will help nail down the difference in the installed drivers. Better include all files (.ini, .pnf and .log are valuable too). Found a G-Sync panel validator in one of the installers, btw .

Unlikely, far too coincidental:

It could easily. But it's the combination with Secure Boot that's in error here, it might even simply be the AW bios handling things erroneously, like an earlier Samsung UEFI brick debacle:

However, the more important thing is this:

Spoiler: Secure Boots

Now, the situation here is a bit different; if only the nvram was affected then a bios flash (or reset to legacy) ought to have worked. Yet even swapping the display to a non-affected system still resulted in ' defective display'-beep. What might have happened is that SB somehow merrily wrote its 'security' bits to the edid's eeprom. Since the MSI GT80 isn't having the same problems (so far), it would point the blame towards Dell. Also, because an EDID is optional, not mandatory, and I've checked an LTN154X1-L02 (Samsung) pulled from an old Dell; it had no edid at all, not even an eeprom; the two data-lines for edid communication were non-connected. Hence, on this system; if there was an edid, then it was embedded in the (v)bios (or EC?) ... need motherboard schematics to find out.

The best way to verify a corrupt edid would be to use programmer with a dead display -> offer still stands .

With any luck, there may be another method, one that doesn't require a programmer. Sourced some DOS-based tools and made an ' auto edid-reader'-thingy with these:

http://www.filedropper.com/autoreadedid

Instructions:
Export the .7z and write the .img to a usb stick (min. 1GB) using USB Image Tool. You'll probably need to clear cmos first (to disable Secure Boot) and find some method to force booting from the usb drive. If it has successfully booted (and run both readers) then there'll be a 3-second beep and you can shutdown the system.

Tested it on both an LVDS and eDP system and both showed the same ' 0D' byte, which is out-of-place; length should be 128 bytes. So pretty sure this can be dismissed as a bug with the second reader. Also; if the second-to-last value in the .bin is non-' 00' then you have an edid with optional extension blocks (beyond 128 bytes), these tools will only pull the first section. This doesn't apply to the currently affected displays, though. Upload the .bin somewhere and we'll see what's what.

The other way around; it almost always has an eeprom, since otherwise it'd be too system-specific.

Actually, the DDCW tool on that auro-reader image can read and write via DVI (and VGA). Can even make it auto-kill monitors, if you'd want ... properly parsed you'd make an out-of-spec edid and watch the display go bonkers.

Spoiler: ps.

It's nice the issue is being looked at by the authorities responsible for this utter mess, but one thing is for sure; their f*up will get buried 300 ft. deep, if they have any say in it. Better have evidence stored somewhere, especially with auto-updating drivers ...

 

@t456 - I think we might be on the wrong path with the Secure Boot and NVRAM thoughts, and here is why. I actually LOATHE UEFI and Secure Boot and use Legacy BIOS. I also successfully crisis recovery flashed the BIOS twice with success and after the flash completed the machine failed to POST with 8 beeps when the hand-off of the display task went to the BIOS. The GPU HWDIAG screen and Intel Option ROM appeared, then as the BIOS splash logo would have first appeared, the 8 beeps and not POST picked up where it left off before flashing, so this was after literally everything had been flushed out of the BIOS and NVRAM and all factory defaults were implemented.

Changing the display panel instantly fixes the problem, so it is definitely something happening on the display panel itself. Also important to note, the problem follows the affected display panel to an otherwise functional machine. If you install the affect panel on working machine... 8 beeps/no boot... install the unaffected panel back to the unaffected machine, everything works fine again. Install the unaffected panel on the affected machine and it also works fine Just don't run the good display on the machine with Windows 10 and NVIDIA drivers or you will end up with two non-functional display as I did and @Arestavo did.

As both of my machines were and still are running in Legacy BIOS mode. I suspect what @Zero989 experienced may have been coincidence in terms of timing. Or, maybe the simple act of turning on Secure Boot triggered it sooner rather than later on. It would have happened at some point either way because those affected had it happen at some point without changing settings. Both times it took at least 24 hours and multiple cold boots for my M18xR2 to manifest an issue. With the Alienware 18, it took less than 2 hours. I was not even finished installing software on a clean Windows 10 installation when it hit that machine, and it was running in Legacy BIOS mode (not even UEFI with CSM, but pure Legacy BIOS).

The only thing that we saw different about BIOS settings was @Daniel1983 was able to boot with the external display in UEFI+Secure Boot mode, but could not boot with the external display disconnected. What I am curious about is if Daniel1983 had never connected the external display so that something was saved in a configuration file or NVRAM whether or not he still would have been able to boot from the external display. Or, if a different external display, or the same one connected to another port (say DP or DVI versus HDMI) that had never been connected before, would it still boot or fail to POST with 8 beeps the same as the internal LVDS? Also curious if he were to pull all of the memory modules and let the machine boot and beep until it flushed NVRAM (the only way to clear the BIOS on an Alienware 17 or 18 BTW) and restarts, would Daniel1983 still be able to boot from an external port or not? We don't know this unless someone has tested the theory.

Edit: @t456 - sent you a PM about the possibility of sending you an affected display to try to flash it and make it usable again.

 

i wonder if EDP does anything different? is there something inherit to EDP that might work to one's advantage?

 

Good question. The only Alienware systems with an eDP port are the 17" single GPU models. The bulk of the affected machines are the 18" SLI models and they have no eDP port on the motherboard, only LVDS. It appears that the 120Hz eDP machines are not affected. I am not aware of any reporting this issue, but I don't know if we can say with 100% certainty that they are exempt from the issue, but until we see an example the assumption is they are. As far as I know, the only affected 17 was one with a 60Hz LVDS display, so there is that piece of information as well.

 

http://www.vesa.org/wp-content/uploads/2010/12/DisplayPort-DevCon-Presentation-eDP-Dec-2010-v3.pdf

confused much?

 

Looks like there are many advantages to eDP versus LVDS.

 

less wires needed, uses less power looks interesting.

 

Looks like Aorus users have bricked machines after win 10 upgrade also. So it's not just us AW guys only.

 

got links?

 

Yes, links please. Sony has asked people not to upgrade their notebooks until further notice as well.

 

http://forum.aorus.com/forum/gaming-laptops/support
Seems like it might be multiple things, but the forum looks new compared to other branded forums, so I didn't link the specific threads.

 

Looks like some people in the Aorus forum are using the work "brick" differently that I would. It appears that some are having complications with the Windows 10 upgrade process, but nothing is getting buggered up with hardware.

Edit: Here is one more example of Windows 10 CPU performance inferiority.

​

 

Should be fun to see how the troll reacts to my post in this thread, LOL. Feel free to respond if he crawls out from under the rock.

 

I know for a fact that Windows 10 Nvidia drivers 352.84 WILL cause this problem. DON'T USE THEM. At this time DON'T USE ANY WINDOWS 10 SPECIFIC NVIDIA DRIVERS. You will more than likely brick your display.

 

what happens if you use optimus or is that broken too?

 

Nope. I won't use any new nvidia driver unless they state is as FIXED

I won't risk my superb Samsung 184HT display. Im very satisfied with it.

 

Hey guys, I was using windows 10 for quite a while now(was part of the technical preview) and used all the NVIDIA specific win10 drivers. i just recently rolled back to win8.1 as i don't want the same thing to happen to me but the thing is I didn't experience any issues while still using win10. @Mr. Fox any news so far?

 

@PC GAMER - no... nothing so far. I think you were smart to do that. I ran the Technical Preview for a while on both machines and had no issues. Something has changed and it's not a safe product any more. Wish we knew for sure exactly what the problem is.

Nobody can find out once their display is buggered up. That would be something you'd need to try before it happens. I'm sure most folks would not be willing to keep on breaking displays with experiments... just too expensive, even though it would be nice to know. If I had the knowledge and tools to fix them over and over again for free, then trying different configurations to see what screws things up would not be a big deal... might even be fun in a sick sort of way.

 

That is the best fun, when it is done in a sick sort of way.

 

@Mr. Fox I am sorry to hear about your displays being burnt up those aint cheap at all! FYI I upgraded to Windows 10 on my AW18R1 with Intel 4800MQ and SLI GTX 770M always set in to Nvidia GPU never Intel graphics. I have been using it a little over 2 weeks now on a daily basis taxing the GPUs with Witcher 3 on ultra settings with no problems. I wonder if this just pertains to 780M, 880M and 980M ? Or have users with 770M also reported having the problem? I installed my Windows 10 several days after it was officially released after the nvidia automatic graphics driver install bug was fixed. Im worried I should revert back to Windows 7 , I really do love the new layout of 10 and the upcoming xbox one streaming.

 

Hi @Tony V - Happy Saturday night, bud. I appreciate the condolences. Thanks to the extreme kindness of two outstanding individuals in our community, two of the machines are fixed now with donated LCD panels. I am humbled and grateful to both of them.

I'm not sure what to tell you, buddy. If you're under warranty just ride it out and see what happens. If you have no warranty and money is tight, going back to Windows 7 is probably the smartest thing to do. I'm definitely not risking it on my Alienware systems any more. Unless and until we know with certainty what is causing it and there is a proven fix to prevent it from happening again, Windows 10 is not welcomed on my Alienware notebooks. Obviously, this doesn't kill every machine every time, but only 1 player in a game of Russian Roulette dies and most people don't want to be the lucky winner. I know I sure don't.

 

Happy Saturday night. I am glad you where able to get new displays very thoughtful of those members in the community. At this point looks like im already in the game of Russian Roulette so may as well ride it out at this point. Winning!

 

Mr. Fox is definitely the worst at Russian Roullette. Not many people loose 3 times in a row lol.

 

Are there still no reported problems with Alienware 17 single GPU laptops with the 120hz 3d display?
I upgraded to Windows 10 from 7 two and a half weeks ago and have had no problems (with my screen) so far.

Now, I wanted to wipe my system and do a clean install of Windows 10 in order to change from legacy BIOS to UEFI (so I can later install a 980m). Does anyone think I'd have any trouble doing this? Do I risk frying my display if I haven't had any trouble so far?