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

I actually sent @Prema a before (corrupted) and after (good) EDID of my display with some suprising or not so suprising results. It's actually the nV driver pulling this sh*t off, as far as i unterstand it, right now i'm using the 355.84 dev driver from guru3d and like on my desktop pc they work pretty well on my ZM too

 

Sorry, but that is incorrect. You've both been led on a goose-chase:

The reason for the v1.3 and v1.4 difference has nothing to do with nvidia, but with the source you got the good edid from:

It's quite normal for the same panel to have different edids. The archive we've gathered together so far has several examples of that:

Spoiler: edid revisions

What happened is that your panel had a more recent edid. This got corrupted (with random changes) and you fixed that with an older revision edid. Nothing more; you've always had a v1.4, until you flashed it to a v1.3.

Furthermore, all posted corrupt edids had random changes, nothing that would indicate an intent on nvidia's part. The 5555-pattern changes the manufacturing date and a serial number. Plus, more serious, it converts it to a CRT-type. This is not something that'd indicate purpose and so far there's only been the three 5555 changes confirmed along with one different, more random corruption.

 

Well, that turned out to be more of a personal sacrifice that I had counted on. I was about ready to do a Windows reinstall on my Clevo because the new GeForce cancer drivers are so unstable. I could not get back into Windows again because of all of the STOP errors, LOL. Finally did, got that filthy 355.82 driver uninstalled with DDU in Safe Mode, rebooted and waited while Windows Disk Check ran on all three of my 1TB SSDs and my 2TB HDD. That's one nasty driver... bad, bad stuff. Same experience I had last time I tried using it, too. The Jolly Green Giant isn't very good at his job any more. Too sad AMD sucks so bad, 'cause if there was a better place to go I would already be there.

 

The amount of redditors I've been seeing who have had 355.82 near break their GPUs... this one guy tried to OC using ASUS' GPU tweak and his card crashed once then refused to clock up past 405MHz on core... across reboots. He had to use DDU to get above 8fps in Bioshock Infinite.

Other people put their PCs to sleep and wake them to constant white artifacts all over their desktops. It's absolutely insane.

 

is this still in win 10?

 

I want to know why we aren't talking about a class action lawsuit yet.

They either fix my laptop, or provide a comparable one if they can't.

 

And a big AMEN to that - From your mouth to Gods ears - Lol,haha.

@Mr Fox - Off topic I know, but I sold my "ALIENWARE TOASTER ADD"

Yeah, I get the feeling this is going to be swept under the carpet.
Did anyone from Micro$haft forum reply to the posts made by our friends here about this issue?

 

I just parsed my edid and got this result. Valid checksum true and edid header ok. But I will not reinstall Win 10 . LOL

 

Andrew, g'day mate. Our Microflaccid forum questions got no official reply, just a generic one from a fanboy/flunky regarding Activations where 10 sniffs around for the hardware id but doesn't write a new id on your bios chip; that answer will ward off future remarks from their side so 10's antics regarding activation's a dead-end & no answer worth quoting was given to dead lcd-screens, just knee-jerk blame-shifting over to NVidia. My thoughts are now on whether 10 does some voodoo / Santería during the 'upgrade process', not that they'd answer. I'll be the 1st to clap stomp & cheer if it turns out 10 does nothing it shouldn't do to your computer parts. If it does something, it's there waiting to be discovered so long as someone knows to look for it. @James D noted a material change in his Bios settings right b4 his lappy crapped out, one by one little clues & anecdotes pile up & 10 starts to look guilty of something, huh?

In today's news we learned about the $900 million fine for General Motors over the bad ignition switches & 124 dead people, how they admittedly lied & covered it all up. Deny everything, explain nothing til the very end. I think @t456 aka Mr.T pity's the fool that can't get to the bottom of this in 2days time. NVidia's 'inability to reproduce the problem' seems laughable after 5 weeks, but understandable if they maintain that position as we move forward. Big b'ness & the Bill Cosby's of the world admit nothing til the men in the black robes force 'em to.

How inspired have u felt this week as the momentum builds up in the forum's favor & you know it in your heart Fox Mr.T & Prema are hard at work in their man caves fit to crack this wide open.

 

Ah ok then, but still besides the fact my Panel working again and still is and i'm happy about that, it looks like that nothing of course is coming from M$ and ngreedia's part, we the users have to fix it again, and without knowledgeable guys like you and Mr. Fox and Prema and all other's we would still be looking at a blank screen.

 

Anyone willing to do repairs yet?

 

I feel sick - Our car industry closes down for good next year (Thanks to the fantastic FTA we signed with the US ) and we'll have this rubbish as a replacement. Australian cars dont catch fire when you turn the key, we've only had one factory recall between 3 manufacturers in the last 15 years - Sorry - I'm on my soapbox, very sore point with me

VERY - GO GUYS, SMACK EM' AROUND A BIT

Ahhh social media and forums, they must be starting to be the bane of big business

 

You have to understand, there are multiple sides to admitting fault, especially a month and two to three weeks after release. 1) cost of liability for destroyed hardware (helping injured parties), 2) opportunity costs (cost of non-adoption of a new OS, potential lost revenue from their "store" revenue, cost of companies not doing whql certification, cost of license sales, cost if companies shy away from directx 12 because of lack of adoption of win 10 etc.), and 3) the shareholder lawsuit for false and misleading information relating to risk, litigation and damages pool, etc. To admit fault means pleading guilty in the first and third points. The second point is a consequence thereof.

Edit: this is a simplistic explanation. But if enough evidence comes forward (and my laptop results suggest bricked, not simple hardware failure (I didn't have another laptop ready to go to test the screen, so just filed on warranty)), then I'd like to explore some options...

 

Got around to steps 3 to 6;

Spoiler: Clevo W230SS - bios eeprom re-programming

Spoiler: Clevo W230SS - bios eeprom re-soldered

Result:

Spoiler: 05 - bios pre-flash vs. full bios write using eeprom programmer

Good, but now the reboot:

Spoiler: 06 - bios pre-flash vs. full bios write using eeprom programmer after reboot

No good then , but at least it's just inside the bios rom itself and not in the flash program (afudos). Needs a bios mod then, expect next week.

 

I received the USB programmer, but it only came with Windows XP 32-bit drivers and I cannot find any 64-bit drivers on the internet for it. There may be something for another programmer that will work, but I am not sure what to look for. So, until I get drivers for it, the device is of no use.

I have Windows XP 32-bit OS, but I'm not going to waste time installing it to use this programmer. I don't know if it will work in a VM or not. I guess I could try that if no drivers are available. I've reached out to the seller to ask that they send me drivers that are needed to make the product usable.

@t456 - any suggestions? Do you think 64-bit drivers for most any USB programmer might work?

 

Its amazing how some of you guys work so hard to get to the bottom of this and have so much knowledge. Some of you should be working for these companies instead of the clowns that write these pos drivers and have been trying for a month to figure out what is going on.

Maybe then we would get some decent, real,actual support with answers to the questions asked and not some generic answers read by a script that are absolutely irrelevant to the nature of the problem.

A big thanks again to all of you

 

Windows 7 allows for a free 32-bit XP vm if you use vm from Windows. Used to use that to play old dos games!

 

Microsoft's virtual machine is unlikely to work. Best bet is VMware Player. It's kind of annoying that they would send a programmer that doesn't have 64-bit drivers but then again this is China.

 

VM should do, just needs to support usb in order for you to give it commands. It's the programmer that does the actual nitty-gritty, not the driver. For an x64 driver, might try a bunch; these things look so much alike, there's bound to be lots of sticker-stamping going on. Use the FlyPro from this list, but there's a few others as well, one of those may be a match.

Funny thing; mine has a new hardware version out (supporting more ICs), but that seems a firmware-upgrade only. So ... flash the programmer's eeprom and ... oh, right ... Also fixed soldering iron a while back ... with another soldering iron.

 

Try these (might be wrong, but worth a shot):
https://www.hightail.com/download/bXBZTkZnUzh3NUxyZHNUQw

 

Hi all
Can any one give me edid of Samsung ips full HD screen 15.6 SDC 4852

Sent from my HTC One M8 using Tapatalk

 

I tried about a dozen 64-bit drivers and software for other programmers and none were compatible with this device.

It did not work with the MSVM, but as @Ethrem suggested VMWare did. I installed Vista x86 and got the driver installed. The database doesn't have my EEPROM, but I think I found a compatible chip with the right specs. I was able to read/write/verify. I haven't checked to see if the panel is bootable. There were actually several that seemed compatible and would read/verify before writing. We are looking for 256B capacity and 8B page size, correct?

@t456 does this look kosher?


Edit: Yay! It boots without beeps.

 

OK, here is a squeaky-clean EDID dump from the brand spanking new Alienware 18 PLS display that has not been touched by any OS or drivers yet. And, a screen shot showing the details.

 

Glad VMware worked and awesome work getting the fix working!

 

Exact same raw as the one on my (working) AW 18 (on which I've never installed Win 10)

 

Now to find a way to write protect I suppose. While you are at it, you guys should just write new open source drivers to replace Nvidia and we we won't need them anymore

 

@t456
Does Windows still store the EDID in the registry? I've been trying to find a way to contribute and I keep reading about how you could find a backup of your EDID in the registry and use that to fix a broken one.

Why would it need to be in the registry if it's on the chip itself? Does it read from it at all? If so, it may be as simple as clicking "read only" on the file inside the registry.

Edit: Complete noob speculation btw.

 

@Sspawn26 - Windows does store EDID in the registry, but that likely won't have any effect on the failure to boot with 8 beeps unless wrong values in the registry are being written to the EEPROM. Making it read-only with correct values may stop the wrong values from being stored in the registry and later written to EEPROM. It's certainly worth trying, just to see what happens.

It might also be worth investigation the registry values immediately before and immediately after installation of the latest GeForce drivers. Saving a backup of the key, one might be able to restore the correct values before rebooting to finalize the driver installation.

Spoiler: Registry Export

Code:

Windows Registry Editor Version 5.00

[HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Enum\DISPLAY\SEC5448]

[HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Enum\DISPLAY\SEC5448\5&e2060f0&2&UID1048848]
"DeviceDesc"="@oem37.inf,%productid%;Samsung SEC5448"
"Capabilities"=dword:000000e0
"ConfigFlags"=dword:00000000
"HardwareID"=hex(7):4d,00,4f,00,4e,00,49,00,54,00,4f,00,52,00,5c,00,53,00,45,\
  00,43,00,35,00,34,00,34,00,38,00,00,00,00,00
"CompatibleIDs"=hex(7):2a,00,50,00,4e,00,50,00,30,00,39,00,46,00,46,00,00,00,\
  00,00
"ContainerID"="{00000000-0000-0000-ffff-ffffffffffff}"
"ClassGUID"="{4d36e96e-e325-11ce-bfc1-08002be10318}"
"Driver"="{4d36e96e-e325-11ce-bfc1-08002be10318}\\0000"
"Security"=hex:01,00,04,90,00,00,00,00,00,00,00,00,00,00,00,00,14,00,00,00,02,\
  00,60,00,04,00,00,00,00,00,14,00,00,00,00,10,01,01,00,00,00,00,00,05,12,00,\
  00,00,00,00,18,00,00,00,00,10,01,02,00,00,00,00,00,05,20,00,00,00,20,02,00,\
  00,01,00,18,00,00,00,00,10,01,02,00,00,00,00,00,05,20,00,00,00,2b,02,00,00,\
  00,00,14,00,00,00,00,10,01,01,00,00,00,00,00,05,04,00,00,00
"Class"="Monitor"
"Mfg"="@oem37.inf,%vendor%;Samsung"
"Service"="monitor"

[HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Enum\DISPLAY\SEC5448\5&e2060f0&2&UID1048848\Device Parameters]
"MaxResolution"="1600,1200"
"DPMS"="1"
"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,00,4c,54,\
  4e,31,38,34,48,54,30,32,53,30,31,00,00,00,fe,00,4c,54,4e,31,38,34,48,54,30,\
  32,53,30,31,00,a0

[HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Enum\DISPLAY\SEC5448\5&e2060f0&2&UID1048848\Device Parameters\EDID_OVERRIDE]
"0"=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,00,4c,54,4e,\
  31,38,34,48,54,30,32,53,30,31,00,00,00,fe,00,4c,54,4e,31,38,34,48,54,30,32,\
  53,30,31,00,a0

[HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Enum\DISPLAY\SEC5448\5&e2060f0&2&UID1048848\Device Parameters\MODES]

[HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Enum\DISPLAY\SEC5448\5&e2060f0&2&UID1048848\Device Parameters\MODES\1600,1200]
@=""

[HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Enum\DISPLAY\SEC5448\5&e2060f0&2&UID1048848\LogConf]

[HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Enum\DISPLAY\SEC5448\5&e2060f0&2&UID1048848\Properties]

[HKEY_LOCAL_MACHINE\SYSTEM\CurrentControlSet\Enum\DISPLAY\SEC5448\5&e2060f0&2&UID1048848\Control]
"ActiveService"="monitor"

 

If you happen to suspect that something bad is being written to a registry key...
You might be able to use Process Monitor, to see if any of the driver installation processes touch that registry key. Or even just keep Process Monitor running with a filter, to see exactly when/what touches that registry key. Just a thought. It's a pretty powerful tool.

 

@Mr. Fox Sorry to go off topic, but what program is the CPU/GPU display on your desktop? I want to say you've told me before, but I just can't seem to remember what it was.

 

AIDA64 Sensor Panel.

Alienware 18 is alive again. I am restoring a Macrium Reflect drive image from about 6 months before Windows 10 cancer inflected it.

Edit: back in the saddle... not bad for the first run at a wimpy 4.3GHz sitting in Papa Fox's lap. Let's see one of them BGA turds do that, LOL.

 

And who looked like the cheshire afterwards.

@Mr Fox and t456 - Great work guys

You've just given hope to at least 50 stressed out laptop owners

 

Great work. Im glad that you fixed this issue! It's idiotic that driver works only in 32 bit (anyone uses 32 bit now lol?) but you did a workaround. I used the old driver for my Samsung Omnia 1 same way with windows XP several years ago and it worked too.

Still waiting for frickin' answer from nvidia. What is taking so long? My win8 is up for about 3 years and it's really "dirty" atm so Im waiting patiently for win10 fix.

 

Perhaps, but which specific panel is that? There's an SDC4852 in the MSI GE62, but that's a 4k display, not fhd. Best method to id would be the stickers on the back of the lcd.

Same as the ' SDC4C48_new' in the archive.

Too bad the eeprom id wasn't successful, it'd be easier to find the chip and figure out its write-protect mechanism. So far we only know how to wp two of the SEC5044s (LTN173HT01-301 and LTN173HT02-D**).

Yes, it won't help after 8-beeps (Daniel1983 tried this already). And it won't help prevent 8-beeps either; the corruption is random and the purpose of the registry is merely for convenience (plus colour translation and compatibility with external monitors). It'd make no sense for nvidia to write to registry and then write those values to the eeprom for good measure. And from the data gathered, there was no targeted attempt to change the edid by nvidia, it was the by-product of something else they've have attempted to do (vcom adjustment or a vbios update?).

Backup would certainly be a good idea, mainly if your panel is not already in the list:

Spoiler: archive correct edids

Code:

PnP id  notes   interf  panel nr.      edid eeprom
------- -----   ------  -------------  -------------------
AUO219D !       LVDS    B173HW02 V1
CMO1720 !       LVDS    N173HGE-L11
LGD01CA         LVDS    LP173WD1-TLA1
LGD0226         LVDS    LP173WD1-TLC2
LGD0285         LVDS    LP173WF1-TLC1
LGD0289         LVDS    LP173WD1-TLA3
LGD02C5         eDP     LP173WF2-TPA1
LGD02DA !       LVDS    LP173WF1-TLB3
LGD02FC C       LVDS    LP173WF3-SLB2
LGD0343         LVDS    LP173WF1-TLB5
LGD0391         LVDS    LP173WD1-TLE1
LGD03FB         LVDS    LP173WF1-TL**
LGD046C E       eDP     LP173WF4-SPD1
SDC3654         LVDS    LTN173KT03-W01
SDC4C48 !       LVDS    LTM184HL01-C01
SEC314A         LVDS    LTN184HT03-001
SEC4A4B         LVDS    LTN184KT01-J01
SEC5044 !?AW    eDP     LTN173HT01-301  Winbond 25X20BLNIG
SEC5044 !?AW    eDP     LTN173HT02-D**  Winbond 25X20BLNIG
SEC5044 A       eDP     LTN173HT02-P01  "" ?
SEC5044 A       eDP     LTN173HT02-T01  "" ?
SEC5443         LVDS    LTN170CT08-D01
SEC5448 !       LVDS    LTN184HT02-S01
SEC544B B       LVDS    LTN173KT01-***
SEC544B BD      LVDS    LTN140KT**-***
-----------------------------------------------------------------------------------------------------
!  = known bricked panels
!? = bricked, but unknown which one
*  = unknown part id
A  = highly suspect: multiple variants exist, perhaps the others are safe ...
B  = multiple variants, flash the correct one!
C  = EliteBook 8**0w DreamColor, 10-bit, for fun ^^
D  = 14.0" version for M14x, just in case
E  = G-Sync approved panel
W  = write-protect possible

If multiple edids exist for one PnP id; flash the most recent edid, unless indicated otherwise.
-----------------------------------------------------------------------------------------------------

 

Ran into an additonal incentive to remove that Secure Boot thing from the bios ...

From MS's ' Secure Boot Overview' page:

Hesitant to use that word, but ... that's a backdoor ... right into the bios !!! Make sure to read that closely; there's three databases inside your bios to which Microsoft has any-day, any-time access to. Not just to read from, but to write to as well.

Noticed those KEK-thingies before, but anyway ... just need to remove that KEK-database and the 'special key' is gone as well; no key, no access, no backdoor. Then short the WP-pin to ground and set the appropriate wp-bits and the entire 4MB eeprom is read-only.

Might be easier to short the Data Input pin to ground as well; write-data can be sent, but it won't reach the eeprom. However, this is unlikely to work with a bios; to increase read/write speed most are implemented as dual or quad spi, meaning the pins do double duty; read +write to every pin. With the edid eeprom this may not be the case; speed is of lesser concern, after all. But there's no way to be certain; the package is the same for either implementation.

 

Can you please give me this one of the 4K SDC 4852 ?

Sent from my HTC One M8 using Tapatalk

 

There is a reason I never spell their name correctly and always use unflattering nicknames. These are not ad hominem assaults. For a while now, I have been warning of this "technology" and it's supporting filth, such as UEFI Class 3 (no CSM) and Secure Flash, all of which present a serious danger. It is so nice to have someone else posting in the Alienware sub-forums to help expose this evil filth for those that will set down their glass of Kool-Aid long enough to get the crap scared out of themselves... and for very good reasons.

These imbeciles are evil and they are absolutely not our friends. They do not have our best interests at heart. Windows 10 and UEFI are part of their filthy plan to own us and solidify their Nazi monopoly. If they control the firmware, they can control the hardware, and all of the settings we can use with the hardware. In turn, they control not only customers but their competition and anyone who would dare to resist them. The Redmond Mafia charges a fee, which is tantamount to extortion, for WHQL Certification and firmware signatures.

But, youz guys don't teenk of dis as extortion. Dis is da fee... for protection... badaboom badabing.

The old man is telling the truth. He's not crazy. The door to the ark is going to close. When the rains begin to fall, the best we can do is pity the fools that refused to hear or believe.

 

Let's not forget our jolly green giant friends at nVidia... those wonderful boys and girls who have designed drivers that are fully capable of IGNORING SYSTEM BIOS OVERCLOCK BLOCK CODE FOR THEIR GPUs.

This is a driver, that is capable of ignoring your system BIOS and your video BIOS settings. Let's think about that for a minute. Your OS can write to the BIOS, and your driver is capable of doing whatever it feels like regardless of what your BIOS and video BIOS say.

Now let's sit back and enjoy our notebooks...

 

Or an underclock, as well. Say ... because its on its last legs and you can't afford a new one; nvidia would deliver that coup de grâce for you.

Sure: SDC4852 - LTN156FL02-L01 edid.

Be sure that that LTN is, indeed, your panel; the pnp id alone is not definitive.

 

Screw this, I think we've waited long enough - Thread started 2nd Aug, that's over 6 weeks.
Anyone in Australia who has had a LCD panel blow because of this cancer please PM me if you are interested in pursuing legal action, time for a visit to the small claims court.
The more AU residents we can get involved, the more chance we have of proving our case and winning compo( upto $10,000AU per individual, so new laptops)
New laptops may be required to prevent future LCD panels blowing, as in Mr Fox's case.
The bigger they are the harder they fall, none of that bulls*@t legal Microsoft agreement will wash here.
Not really worth doing it by myself as I've only had one panel blow ($90 for a new panel or $60 for the small claims court fee, almost not worth it) And the cancer has not "reblown" my new 120Hz panel - Yet...........................

 

maybe just don't use win 10...stick with 8.1

 

Hey @Mr. Fox , sorry for posting off topic stuff again but can you upload the nvidia inspector tweak for Mad Max please? Just installed the MGSV tweak and it works flawlessly albeit some stuttering here and there, nothing major though

 

You can download it over here.

 

pretty amazing u guys come up with way to re-write the broken EDID in display. thumbs up to t456.

 

Hi guys, i have 2 weeks with W10 and 355.85 zero problema. Maybe my LCD Panel is not affected? the only weird thing when upgraded 780m the hdmi out never work the hdmi out. mdp vga and hdmi in working good.

M18X R2 (A12 Bios) with Secure Boot Fast Boot UEFI.
i7 3920XM
32GB RAM
2 Dell GTX 780M (Bios Stock 80.04.AF.00.03)
mSATA EVO 840 500GB
HDD 1TB

 

Samsung LTN184HT02-S01. Known bricked model. I give you week max. Edid may be already corrupted, dump it on the forum (not image) and you can still save this display.

 

my data string looks different.

 

SEC5448 in M18x is 184HTS02-01. I have same display in my Win8.1 M18xR1 that im currently typing on.

Check older posts in this thread. There is 100% pure healthy 184HT edid. Also check it in EDID web reader. If it shows checksum valid : FALSE and yours edid is different then it's is already destroyed, count hours to 8 beeps. You've been warned about Windows 10 for a month.

 

Compare this edid to posted before healthy 184HT