*** Official Clevo P75xDM and P77xDM/Sager NP9758-G and NP9778-G "Batman 2.0" Owner's Lounge ***

Are you running the GPU stock? If you are overclocking it, stop and see if the problem goes away.

The problem you are describing sounds like a 980M that is dying. @ssj92 and I both had 980M GPUs where the screen would go black and it would start beeping. If that is what you are experiencing, contact the place you bought it about getting the GPU replaced.

 

Thanks no over or under clock here. Interesting development though :

I'm currently on a business trip at hotel and after multiple successful runs of firestrike decided to try OverWatch again.

No issues after about 45 minutes, which is unheard of.

Is it possible that the power outlet I was using at home wasn't supplying enough juice?

 

Maybe unstable power delivery from the wall home. 300-500w is nothing from a wall outlet today. Depends of course of the Amp. But so weak Amp is not normal to use nowadays.

 

I'll keep that in mind.


Also, does the 4K screen used in the P750DM or P770DM offer 100% adobe / srgb?

 

Panels can change over time so best to ask when ordering

 

I've updated my graphics drivers this weekend and noticed that the screen flickers about every ten seconds when playing fallout 4.. Only when unplugged. I've disabled the battery saver in gforce experience hoping that would fix it bit it didn't. Any ideas what could cause this? Thanks

Sent from my HTC 10 using Tapatalk

 

Have you looked to see what your GPU and CPU clocks are doing? Graphing them over time is the best way to see how they are getting on.

 

Thanks..I enabled the battery saver in gforce experience and set it to 60 fps and the flicker is gone. I guess disabling it didn't actually stop it somehow.

Sent from my HTC 10 using Tapatalk

 

Have anyone tried to use a razer core with P7xxDM?

 

Iirc, @XMG said a while back that for now an egpu setup isn't possible.

Edit: you'll most probably have to wait for some Prema magic.

 

internal display is not usable yes, but what I'm interested is if the GPU will output to an external monitor or not

 

At the risk of being narcissistic I have quoted myself where we were discussing it previously!

 

It's a valid point, and worth quoting again.

 

Just thermal pads. No other heatsink?

Just a thermal pad? Do you have a suggested part that works well?

 

Yup nothing else... Max temps today in 2.5hr gaming session so far is 57C.

 

Clevo P750DM Thunderbolt Firmware Update v16

 

I have also encountered the beep-shutdown issue while playing WoW. Is there an exact gpu temp for shutdown? 80c?

 

90C+ usually.. I would stop playing and immediately clean your fans and raise the bottom of your notebook (use some bottlecaps).. Might also want to order some thermal paste and repaste your CPU+GPU..

 

I'm sure my gpu temp never reached 85c. That's why I'm comnfused.

 

Are you by any chance running the Prema BIOS mod? I would randomly get a flashing power button, high fans and shutdown after flashing that. I haven't 100 percent pinned it to the vBios yet but I haven't noticed it since going back to stock. Also my gpu runs like 3-6 degrees cooler without it (maxes at about 68C).

 

Will this work on my machine https://www.ebay.com/itm/252422807658 ? I asked lpcdigital for the model they used on my system and they sent this pic the same models on ebay run quite expensive as compared to the link i posted thats why i was thinking of grabbing the rams i posted at 30usd for a 8gb module.

 

So long as it matches the speed and cas timing of your current ram it should be fine.

 

How can i make sure it matches the speed and cas timing?

 

Download something like CPU-z and look at the current memory speed.

 

I reverted to normal vbios, and maximum temp for 980m while running fire strike stress test is 75c instead of 85c.

A wild guess, maybe its because I am using 230w power adapter, and prema vbios unlocked more power than it could handle?

Edit: It seems WoW is more stressful than fire strike loop..

 

@hyno111 in the Prema VBIOS the power limit is disabled and the temperature target is set to 92C. I would guess any lower temperatures seen would be the card throttling.

 

the theory about the power tho seems somewhat plausible imho, i had the issue too and asked @jaybee83 in the ZM forum how he oced his 980m, cause when i tried DOOM with only +10mhz core clock i got the blinking and fan spinning issue, was the only game tho i noticed that issue, so far Overwatch e.g. wich seems extremly picky with OC (at least my 980TI on my desktop seems to crash even after 30 sec ingame, Witcher 3 e.g. took 10-15mins to crash, tested with knowingly unstable OC tho) runs nice at +150/+300.

 

Yes when removing power limits you are allowing the card to run faster and hotter than intended. It is a form of overclocking.

 

That power limit is there for a reason, which is why I wouldn't recommend running the Prema vBIOS, at least if you don't have good cooling and a 330W adapter. Running Furmark with the stock bios I get a max temp of 68C after about 20 mins. You can see the card throttling in GPU-Z with the perfcap reason 'PWR', which keeps the chip cool and under it's TDP of 125W. With the modified bios, after about 2 mins my card is at 74C and rising, and black screens a few seconds later and then the laptop shuts itself down (the shutdown seems more related to power draw then temps). It sucks because I would really like to use the modified vbios but the high temps and instability scares me. But it's not too bad as you can still overclock with the stock bios, just not as much.

 

Nothing wrong using Prema´s vBios, but using Furmark is a big no-no...It places an unrealistic stress on gpu´s and is known to have killed some. That´s why nvidia has implemented some protections against it.

Clevo has more than adequate cooling on it´s mid to high-end machines and the 230w adapter is enough for oc´ing the gpu (provided you don´t oc the cpu at the same time)

http://nvidia.custhelp.com/app/answ...ther-stress-tests-with-geforce-graphics-cards

There´s plenty more if you google "furmark" and "nvidia" .

 

Yea I don't make a habit of running Furmark. I basically use it for troubleshooting purposes and to help make a point regarding the power limit being removed and how temps can get much higher than they normally would. Maybe it's just my gpu but I've experienced enough issues that I don't experience on the stock vbios to stay away from the mods. 70 - 80C isn't bad temperature wise by any means but if I can keep my gpu under 70C then seeing it go over with no real benefit based on what I normally use the laptop for annoys me. I may revisit it one day when the 980m just can't cut it for newer games and I need to overclock more than what the stock bios allows for. I've never had any issues with the power throttling in real world usage anyways.

 

@TheGreatAnonymous that's great but what was the GPU core clock when running Furmark?

 

With the stock vbios when running Furmark after awhile the throttle kicks in and the knocks the core clocks down to the ~800mhz range, can't recall the exact number, so there is no doubt it's being throttled, which is why the temps remain so good. RAM clocks continue to operate at full speed. This may sound bad as the 980m is supposed to max out at 1126mhz on the core, however Furmark is an outlier and I've never seen this kind of throttle happen during normal gaming, or anything else. However, with the modded vbios and Furmark, there is no throttle whatsoever, however either the gpu or the power supply doesn't like this, because everytime after about 2-3 mins, usually at exactly in the 70-74C range, the screen will go black, and a few seconds later I get the spinning fans, blinking power LED, and automatic shutdown after about 30 seconds. This doesn't just happen with Furmark when running the Prema bios, but randomly when playing games as well (even with stock clocks), just without the black screen. It just causes the fans to rev up and the forced shutdown counter to trigger. And it's not overheating because my laptop has never seen anything over 78C on the gpu. I've also noticed that the gpu power reading in HWinfo will exceed 125W when running Prema, which is supposed to be the 980m's max TDP. Not sure how accurate this reading is but with the stock vbios it never exceeds 115W.

 

@TheGreatAnonymous that was what I was expecting. The non-recoverable black screen is due to hitting a hard power limit of the VRMs and is a known issue with the Maxwell mobile cards. I expect this with the VBIOS mod since the power limit is disabled anyways, and you should not use Furmark to test your GPU stability.

 

@TheGreatAnonymous: Did you check your Thermal Pads on your VRM's? Maybe they don't make good contact, maybe stability improves by swapping the Pads with high Quality ones (e.g. FujiPoly 14-17W/mk. Your rig shouldn't shut down by reaching 70°-80° on your GPU. VRM's tend to become hotter as the die, so placing some good Thermal Pads is crucial as well.

For instance, I can oc my graphic Cards with +350/+480/+112,5mv out of the box for benching purposes or for playing games with +150/+220/0mv and temps will never reach the middle 70-ies. I never had a black Screen, only freezes when pushing the Cards too much.

Just my 2cents.

 

just curious are you using aftermarket Thermal pads? Maybe i will get the 330W brick from CEG and some thermal pads or just wait for the 1080m, cause like i wrote some Posts before i had that shutdown issue too with Doom and only +10mhz on core, thermal pads are sitting nice on the card and using Gelid Extreme, card wasn't roasting either.

 

The maximum power reading I got from hwinfo is about 147w with prema vbios and 117w with stock vbios.

 

Jupp, I am using FujiPoly 14W/mk and have CLU on the dies.

Did you by chance used a Watt-Meter during playing Doom? Not that this tiny little oc should strain your brick too much.

Didn't try playing Doom with oc on my cardz, but can give it a shot this evening.

 

@CaerCadarn

Thermal pads are just the stock pads however they are positioned well and make good contact. OC'ing is kind of a dice roll depending on your chip so perhaps you got lucky with yours. Thing with Nvidia cards is they only have a temp reading for the core, I had a 7970m that would show ram and temps for other parts of the gpu. There is definitely something about the Prema bios that my system doesn't like, and I can't really say it's a problem since everything operates beautifully without it. Stable as hell and hardly ever see temps over 65C under normal circumstances, and that's living in Houston with 100 degree weather.

 

@TheGreatAnonymous

That's not normal. You may have to reset/update the systems EC. What version is it on?

EC shutdown value, as per NVIDIA ODM guidelines, is a GPU core temp of 94c sustained for more than 30s.

Furmark: Even though it will show full clocks and 100% GPU load, it is not using 100% at all.
If we would disable the actual Furmark lock bit then the system would simply die.
Short version: Stock or Mod, Never use it!
Long version:

http://forum.notebookreview.com/thr...0-owners-lounge.776183/page-114#post-10102336

 

The system has already been in to Sager and had the motherboard replaced, so EC and Bios is up to date. Funny thing is the motherboard replacement was probably unnecessary as it continued to shutdown after getting it back. The only thing that seems to stop the shutdowns is using the stock vbios. Obviosly Sager isn't going to want to hear anything regarding issues with a modded vbios, and may even void my warranty if they find out. Maybe there is something up with my gpu that the unlocked bios exposes, but so as long as the sysem runs ok with the standard vbios (including with the allowable overclock) I'm going to have to chalk it up to the vbios mod as causing the issues.

 

hmm my power supply died on my Desktop rig any chance to get my Asus VG236H to work with 120hz on my Notebook? I thought about Duallink DVI (monitor) to Displayport (notebook) adapter but i'm not sure about active or passive, problem is that active converters are a bit expensive, somebody got a bit experience about that?

 

The adapters should list maximum resolution but likely active is required.

 

Yep i guessed so, i looked and looked but everything indicates that i need an active adapter, just some Apple mini-dp adapters seemed to work. I even thought about dvi-d to Hdmi 2.0 cause the 980m and the NB should have them, but i not researched yet if the cable has to be 2.0 compliant too.

 

No I believe the HDMI is 1.4b, the mxm cards were not 2.0.

 

The P771DM has HDMI 2.0 compatibility due to 980M and 6700K.. So yes, it is HDMI 2.0.. I run HDMI 2.0 4k@60Hz on my P771ZM due to a similar set up..

 

Just for more data: both without and with a modded vbios, I'm hitting what appears to be the VRM power limit failsafe (black screen, fans max, beeping until shutdown) at a gpu core temp of around 84.

I'm currently dealing with the issue by capping my CPU at 80% and setting a thermal target on my gpu of 77*. Any higher on that target and the issue occurs.

Clearly gimping my system isn't an ideal solution. Will look into EC flash, repast, and custom pads.. Thanks for the discussion!

 

Can anyone think of a possible reason why a samsung 950 pro wouldn't work in my P750ZM? Everywhere I've read about people putting them in a P750ZM has said to simply run prema's mod, load defaults and do a uefi boot with CSM disabled.

Despite this, I've had zero luck getting it to be recognized in any flavor of Windows 7 with slipstreamed nvme hotfix and loaded samsung nvme driver, Windows 8.1 and Windows 10, and I've tried every combination of uefi/csm and changed bios options just for giggles even though prema himself has said it would be plug and play with defaults, but it still refuses to show up.

I know the m.2 port isn't shot because I have a 256gb SM951 pcie AHCI m.2 card that works fine in the port, and I know the 950 pro works because it shows up and I can copy files to and from it on my desktop with a pcie x4 adapter.

I'm at my wits end here with it, is there something, ANYTHING anyone can recommend me to try to get this card to work before I return it and give up? Can someone who actually got it to show up show me how it looks in windows, what they have set in the bios, how they installed windows to it etc.?

Should I try reflashing the bios even though I'm 99% sure that won't do anything?

 

I dont think it will show up in the BIOS, since there is no dedicated hardware NVME controller? (someone correct me if im wrong) But there is a guide @Phoenix posted somewhere in the forums on how to install Windows 7 with the 950 Pro. Might want to search for that.

 

I read about that, but the problem is I can't see it during windows 10 or windows 7 setup, even when I slipstream the hotfix and load the drivers.

I currently have it set to uefi/no CSM with all other settings default, and windows 10 installed and updated to a 2.5" sata SSD which is the only other drive in the system, and the samsung driver setup fails to find the drive so it won't even complete the install.

I get zero indication anywhere along the way that the drive is even attached to the port.