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

@Scanner - what I was suggesting was disconnecting the external display and running only on your internal display to see if it gets blown out (if you are still under warranty). If you shut off your system you may not be able to power it up again with the LVDS cable disconnected. That causes 8 beeps on all of my Alienware systems because at POST it checks for the internal display and if it cannot find the display, you get 8 beeps.

@thegh0sts - was that with 353.62 drivers? That is what I was running that killed all three displays. The slider was not grayed out on my M18xR2, but it was on the 18. I wonder if that is why it took longer (both times) to kill the M18xR2 display? I still cannot believe the Alienware 18 display was killed in about an hour.

 

60 hz panel, don't know what vendor but can check if you need, running in discrete


Sent from my iPhone using Tapatalk

 

So let me get this straight because I skipped like 6 pages...

Windows 10 + 353.62 is KILLING AW17 R1 and SLI models of old Alienwares?

WHQL drivers?

Cuz you know, WHQL means that microsoft has tested and verified no damage by using those drivers. Even if nVidia doesn't pay you for the screens, LEGALLY microsoft must reimburse you for it (especially since Win 10 forces those drivers on you)

 

no no, like anything before 353.62 like 353.38 hf, 353.45 hf, 353.49 hf drivers. if we want to eliminate win 10 and narrow it down to a specific nvidia driver version then i would have started there.

if you can like the actual model number. i wonder if the inverter board is busted. there's no sign of models of 120hz panels being affected....yet.

pretty much so far it's been:

- M18X R1
- M18X R2 x 2
- AW18 R1
- AW17 R1
- Lenovo Y50 (as reported by @Mr. Fox)

I don't think GPU is a factor since there's a mix of 780Ms and 980Ms (stock and aftermarket respectively). The common things are:

- 60hz display
- using nvidia 353.62 drivers
- windows 10 x64

no reports of 120hz panels succumbing to the same fate....yet!

 

Hi all! I didn't know about that issue with windows 10 and alienware machines. If it will be useful for help:

I installed windows 10 Pro on my ALW17 with GTX 980m and windows 10 Home in my M15x GTX 780m.
In the first i've upgraded from windows 8 Pro and in the second from windows 7 Home Premium. Both using the App togheter with media creation tool to
get out of messages of incompatible display device.

All are running great. I used NVIDIA driver 353.30 to windows 10 modded by me for use both in 980m and 780m.

I would bet that these things happened because of SLI, but look like that other uses with single card had trouble.

My alw17 is pure UEFI in dedicated video card mode, so I don't use with optimus since i have it with my 780m and after with my 980m.
And the M15x is a very strong machine that i think is hard difficult to happen something like that. It work with a 100w tdp cpu and i tested before nvidia 980m high overclock
and the machine was totally ok. The display can go until 120hz without issues in idle only get scratching on game.

So my feedback is more for the Alienware Ranger.

Hope that help!

 

@Mr Fox I disconnected both internal
Monitor cables from MB. the external monitor is connected, no problem turning on PC. On other point, not willing to blow internal monitor. PC is under warranty but not risking sending it to Dell and being sent down rabbit hole of no parts available here is crappy new model replacement.

 

I disassembled my laptop up to removal of the palm rest and I'm not home now but I will post the exact model number here if you want when I get to removal of the screen. I ordered the 120 hz panel and cable because might as well if I already have to rip everything apart.


Sent from my iPhone using Tapatalk

 

See opening post...

I was using this driver: http://www.nvidia.com/download/driverResults.aspx/87791/en-us

Trouble is, I don't know why or who to blame. And, you know good and well getting Micro$haft or NGREEDA to accept responsibility will be like whizzing into a headwind.

Now I have 3 dead 18" Alienware LCDs... and, I am scared to even consider installing Windoze OS X on my Clevo. If it or the drivers kill my Clevo I won't have any computer left to use. SO, I guess I will either part out my Alienware machines on eBay or wait until I win a lottery so I can afford 3 replacement RGB LED 18" panels to fix my broken Alienware machines.

 

i'm still waiting for my 3d cable to arrive from china...damn slow blow mail crappery! the panel's model number should be behind the panel so you'll need to remove the bezel and remove 6 screws to get to it. you'll need something to pry open the bezel.

and how long have you had your install for? more than a few days?

i think this might be the preverbal straw that broke the camel's back. part the laptops out and move on from alienware.

i won't do this cos this is my only working laptop and i would be screwed if i didn't have it working.

 

Yup, that says WHQL.
That means that bottom-line, Microsoft must reimburse you if nVidia does not (as long as your OS is legit of course). That's what the WHQL branding is for. That's why Windows costs much more than OS X, because you're paying for the security of that WHQL license. Microsoft tests them and deems them fine for release and pushes them out. If damage occurs to your system due to it, then Microsoft is bottom-line at fault for not doing enough testing. If it's a beta, non-WHQL driver, then nVidia would be at fault, however likely would not be required to reimburse you, depending on how their Terms of Service is worded. Of course if updating to their driver kills your GPU at stock, they're likely going to fix you up, since that driver in their GTX 200 series killed GPU fans spinning up and everyone's cards burned out.

Anyway, you do have a legal claim to reimbursement once your OS installs were legit and you were using their WHQL drivers.

As for your Clevo, a few users in the P570WM forums seem to run Windows 10 on them without fault. I'm less lucky; my P37xSMx thread is dead XD.

 

I had that issue before, in my case was that the nvidia driver didn't install correctly. This was the first time I tried to go with a newer driver. After noticing this issue I just went back to the desktop mod and I was able to change brightness.

 

I've installed 2 days ago. Maybe i have to wait more to say something. My screen flicker in black one time from i remember but after i rebooted normal. Can be died after some time?

I don't know if I am in risk but when i installed i didn't knew about that.

 

that's what we're figuring out.

 

Not having a brightness slider on a single display is a HUGE indicator of an issue. Now this is from a fresh cold boot with the laptop only monitor and no slider issue. This means the system is giving you, sensing, no control and the gain is most likely set on its highest for output to the monitor. My suggestion here is ANYONE with no slider on a single monitor to stop using the driver immediately! Every monitor will have tolerances, some better than others, but why gamble.

 

I have been using Windows 10 since release day on my M18x R1, however, both my 580m died some weeks before the install so I have been using only the HD 3000. So far I have not experience any problems with the machine or screen... now I am glad I killed the 580m's.

 

Never, ever, ever, thought I would hear those words from a gamer, not ever...........

 

Hehe, I am getting a single 7970m from a friend for $75 which is much cheaper than the screen. One 580m died from a mild OC while the other burned, with smoke and everything, for some reason... what is sadder is that I already got used to this 18 inch laptop with a iGPU.

 

If you W10, or any other OS for that matter, be sure you have a brightness adjustment. Now it may be that the AMD card in any case just can't overdrive the monitor but better safe than sorry.

 

I warned about this. If you see anything acting odd, like your fans not working or sporadic temps, (rememebr when i said my cards were hitting 90C !!! and no fan action..with windows 10) Anyways, there are some windows updates that are causing it. It seems to only happen with the right windows update. After 2-3 reboots and other windows updates, they seem to go away. I have been using external monitors on all test benches so i did not suffer this issue, with the LCD panel, but otherwise I most certainly would have. I have 6 cases out of maybe 20 recent Alienware/DELL/Clevo/MSI/ASUS users who have similar fates recently.

It is a combination of Microsoft with its forced updates and nVidia that cause this. If you run on the default VGA no issues. If you use the nVidia drivers for Linux or MAC no issues. If you use the recent last 3-4 drivers of nVidia like 352+ and Windows, you will go through the same issue. hopefully you will reboot and get past this specific update that is causing it. but I suggest not spending too long idling on your laptop when you have updates left.

 

haven't I been saying do not use the latest, LOL,

I do not link the latest, neither does Eurocom. We're a few drivers behind for stupid fearful reasons like this. This is totally , however, not me or Eurocom's fault, but we will do what we can to help those who are in need, and we are now much more aware of the issue, and will try and help future upgraders, thanks to everyones help and concern of course. I personally haven't figured out 100% what is causing it, but I stopped trying to experiment heavily with windows 10 when my Videocard once overvolted itself to 1.4v and the fans stopped working. I think i told some people or mentioned it in a thread, no one really took notice of it, when i said it. But I stopped trying different drivers, and as such have only two drivers I have made for AW systems in my drop box, even though there are like 4-5 others I could have. No one I know using the drivers I have linked, mind you, have reported it failed as in 8 beeps LCD issue or whatever.

BTW, the drivers I have linked were just from J95, nothing special. I followed his instructions on making the 352.84 and 353.17 drivers work for dedicated/120hz/SG-optimus. kind of like an all in one driver, so i could give it to my buyers who use Alienware.

 

i'm staying away from both win 10 AND nvidia 353.62 drivers. i'm still on the 353.38 hotfix drivers and that's still good.

 

Hmmmm, no problems yet. I shutdown twice since as well.

 

with 353.62? when i did a fresh install of 10 it pretty much installed only the audio driver, accelerometer, and bluetooth (plus the usual windows updates).

i then went an disabled all driver updates in windows and installed the nvidia drivers. when using the latest drivers i did managed to get a brightness slider.

 

Has anyone tried this with AMD cards to see if it's Nvidia's drivers or if it's windows 10?
I have an M290X and have had windows 10 installed for 3 days the only thing I have noticed is when I run windows 10 with both Intel HD and m290x the brightness it locked at max and cannot be changed at all, but as soon as I disable the Intel GPU, the brightness can be changed but the gpu stops responding every 4-6 hours however it was doing that since I got it so I think it'll die on me soon regardless.

 

I've been preaching against Automatic Windows Updates for a couple of years now. They are a a ticking time bomb and absolutely a disaster looking for a place to happen, but in this case I don't think I can blame that for my experience.

No automatic updates are allowed on my systems, including Windows 10. I have them disabled in group policy and "deferred" in Control Panel. Updates can only be installed manually. With the two M18xR2 displays it took about 24 hours for them to burn out. With the Alienware 18 it took about an hour after switching to discrete graphics and installing NVIDIA drivers from the NVIDIA web site. Everything seemed to be working normally. The M18xR2 (both times) failed to POST with 8 beeps and no LCD backlight after being shut off. The Alienware 18 make a click and shut itself off and hasn't been able to POST, throwing 8 beeps and has no LCD backlight since it shut itself off.

The actual display panels are damaged somehow. There is no backlight on any of the 3 machines now. The first one that went bad caused 8 beeps and fails to POST when installed on the M18xR1, which does not even have Windows 10 installed, so it's not going to get better by rebooting. Either Windows 10 or NVIDIA drivers for Windows 10 is causing the displays or their circuit board to fry. I am guessing the drivers are forcing the display to run so far out of specs that it is ruining them.

 

maybe not the actual LCD/LED bit but maybe the inverter board is busted because it's that bit that controls how the signal is displayed.

 

On the M18xR1/R2 the LCD and inverter board is a rather large PCB that is sealed between the LCD panel and the heavy cast aluminum housing the LCD is mounted in. It's not a tiny little strip shaped PCB like you would find on run-of-the-mill consumer laptops. You would literally have to destroy the assembly to remove it. They are not designed to be taken apart. I haven't taken apart the LCD on the 18 to see if the inverter circuit board has its own part number. I am going to take a wild guess that part is not serviced separate from the LCD assembly on either machine. It may not be on the 17 either, but I don't know.

 

probably not. i am still waiting on anyone with 120hz panels to see if any break the same way.

 

These displays very rarely go bad. It's a shame and a crime they are being destroyed by incompetent software engineers.

 

yeah, but which: M$ or Nvidia? that's the question.

 

Yep, 353.62, installed by Windows...

 

Just created this acc to post on your thread, abut your problem i'll just gona paste my nv forum post, maybe can help you

Did you tried the ePSA mode? (hold "D" or "Fn" key while turning on -key combo varies from model-), if this work and you suffer the problem again maybe you should go to bios settings and test the system disabling the "non-turbo flex support" in the processor tab, i've seen a case where this option enabled cause these 8 beeps code, apparently under certain circumstances that cause troubles.

 

i did a manual install. got a screenshot of the update history?

 

Thank you for your suggestions. I appreciate you taking time to make an account and post your advice.

I am the person that discovered having non-turbo flex support enabled can cause very similar symptoms and I definitely have that disabled. I tried blind flashing the M18xR2 twice and it did not help. After trying a number of things, including testing the GPUs, I decided to check the LCD panel. I found that the LCD assembly is actually "blown" because the problem follows the LCD assembly to another machine that is working correctly with a working display attached. The laptops themselves seem to be unaffected, and it appears that only the LCD display panels are getting messed up.

 

it's hard to determine if it's either: so that's what i was thinking is that someone with win 10 should use older drivers to test if it is nvidia or not.

 

I feel sorry for your problem, just one last option, did you tried to clear the CMOS?, just a month ago i've suffered a big problem with a cmos data corruption (curiously testing win10 with bios fast boot option enabled), begin with bios showing 2 ram modules (i have 4), and after a few test system refused to boot, tried clear cmos with clrCMOS button, nothing changed, then, i suspect bios corruption, then i tried bios chip replace (i have one as backup because my board doesn't have bios recovery functions) and problem persisted, then, i've tried to clear CMOS again, this time using the jumper (my MoBo have both options) this time the problem was solved, so i´ve discovered two things, problem was cmos corruption and that my clrCMOS button doesn't work LOL.......

 

Well my GF A18 have been running windows 10 too and everything is fine. She also have M290x CF.

 

Hello and thank you again. Yes, I did try that. Even did crisis recovery flash and that did not help. Here is the sequence of events:

• Installed Windows 10 on M18xR2 with GTX 780M SLI... everything seemed fine for about 24 hours, maybe slightly longer
• M18xR2 suddenly started 8 beeps, no boot after being turned off - 8 beep code indicates GPU, LVDS Cable or LCD issue - tested GPUs in M18xR1 and they where perfectly fine; replaced LVDS cable and no change
• Removed display from M18xR1, intalled on M18xR2 and everything is fine with M18xR2, installed screen from M18xR2 on M18xR1 and it started 8 beeps, no boot, confirming LCD failure
• I assumed it was just an ordinary hardware failure, was glad I could steal the LCD from the M18xR1 until I could purchase a replacement part... life was good (or so I thought)
• About 24 hours later, the second display is doing the same thing with the good display taken from the M18xR1 and the 8 beeps, no boot problem also surfaced with the second display after the system had been shut off
• Alienware 18 has GTX 780M SLI and Windows 7. I was not planning to upgrade the OS but I wanted to rule out any anomaly involving my M18xR2. So, I installed Windows 10 yesterday running on Intel Graphics and all is well... about 16 hours later (today) I switched over to GTX 780M SLI, installed NVIDIA GeForce drivers and about 1 hour later, maybe slightly more, Alienware 18 makes a click sound and turns off while web surfing. Pressed power button and now it has the same problem... 8 beeps, no boot.
• On all three machines there is no LCD backlight, but if I open and close the lid the backlight turns on
• Removing NVIDIA GPUs has no effect... display malfunctions the same on Intel HD Graphics once the LCD panel is messed up and the damage is already done (M18xR1 has only Intel HD Graphics, no GPUs, and has the same behavior with a bad display attached)
• On all three machines, if I boot with mini-DP or HDMI connected to an external monitor I get the vBIOS HWDIAG splash screen and Intel Rapid Storage Technology Option ROM but 8 beeps begins when system BIOS loads and attempts to begin output to LVDS port

I cannot prove it, but all indicators are that something with NVIDIA's Windows 10 driver is what caused the problem. I think it is not very likely that the OS is directly causing the problem.

 

don't know what to say really other than.....

DON'T UPDATE TO WIN 10!!! (or at least the nvidia 353.62 drivers).

 

I have brightness control with the latest NVidia drivers too.

 

so i don't know what that means. it could mean anything.

 

Totally different GPU and different 900p display in the M17xR3 so I don't know if it is relevant or not. For @Robbo99999 sake I certainly hope it is not. I would not wish this kind of fiasco on anyone... I am pretty upset right now. I just don't see how this could be a coincidence though. Should be interesting to see what transpires in the coming week or so. Things could get a little bit crazy.

 

Yep, I would be very upset if my laptop screen died, it's my only laptop and I don't need a new laptop yet, but if I'd had 3 laptops die like you then that would make me even more angry! Thing is, I don't know what you can do about it, maybe it is worth at least presenting the results to Microsoft or NVidia or Alienware and seeing what kind of a response/help you get - probably best not to rant at them if you contact them, but maybe there's an outside chance of some compensation or replacement. I'm thinking of downgrading to 352.84 to side step any potential damage to my laptop, and if I do that I'll also be able to test if I have the same weird Cold Boot Low GPU Performance (performance fine on a warm boot).

EDIT: downgrading to 352.84 didn't solve the Cold Boot Low GPU Performance Problem. Decided to play with fire and go back to 353.62 aka the Alienware LCD Killer!

 

Is the AW18's panel easier to extract? I mean is it encapsulated in magnesium frame like the M18xs, or it can be pulled out like on most laptops? My guess is that display's processor got burned.

 

Yes, it's much easier to remove. It is not encapsulated. The bezel snaps off and there 8 screws. The frame around the LCD has brackets on it and I do not believe the encapsulated LCD for the M18xR1/R2 has those brackets on the frame.

It has to be something like that that caused it and more likely than not, it is the new NVIDIA drivers for Windows 10 that caused it to happen. The 8 beeps and failure to POST occurs when the BIOS attempts to hand off the display to the LVDS channel. If I hook up a mini-DP or HDMI cable to an external display there is output until the system BIOS tries to load. The NVIDIA vBIOS HWDIAG displays and the Intel RST Option ROM shows up briefly, then the beeping begins. If I start the machines with the lid open there is no backlight. If I press the power button and close the lid quickly, then reopen it the backlight comes on. While it is beeping if I close the lid the beeping stops for a moment and the external display backlight comes on, (although nothing is displayed,) then if I open the lid quickly the laptop backlight comes on, but the beeping is only momentarily interrupted when opening and closing the lid. And, now that the damage is done to the panels, the behavior is the same whether using Intel Graphics or discrete GPUs.

 

Yeah, might be different EDID communication, although I find it strange, since it is free upgrade for W7 machines as well and a lot of them, not to say most, lack UEFI support. Something is messed-up with the communication. Can you get the display out and examine its board for any physical burn? It usually is around the connector.

 

Is this issue only affecting all gen 18's with 780ms? Or are other laptops with 900 series being affected too

 

It's weird to say it like this, but I'm really sad you didn't get to try it on Win 7 and see if it was the drivers directly or the combination of drivers + win 10.

 

I've been using Win 10 on a AW17 with the new NVidia drivers and a GTX880M.

It's a 120Hz panel - Win 10 or the new drivers actually lowered the GPU temp by close to 20% and also increased it's score in 3DMark Vantage by around 15%.

Opposite affect in the same spec machine with a 60Hz panel and a AMD (8970) GPU, the new CCC or win 10 increased the temps by 15 degree celcius and lowered the 3DMark Vantage score

 

BTW - The brand name of the AW17 displays I have are AU Optronics, both the 60 and 120Hz models are the same brand