Please help me diagnose an issue with my Sager NP8278-S

Mizugori

[H]ard|Gawd
Joined
Mar 25, 2004
Messages
1,240
***I really appreciate any advice/assistance you can offer me***

I have a Sager NP8278-S (Clevo P170SM-A) which I purchased from XoticPC several years ago. I love the laptop and have never had any major issues with it (aside from the battery dying, which is to be expected after 5-6 years!)

Basics:

-Sager NP8278-S
-i7-4810MQ 2.8Ghz - 3.8Ghz 6MB Smart Cache
-GeForce GTX 870M 6.0GB Video Card
16 GB DDR3 1600Mhz (2x8GB Dual Channel)
2 x Samsung Evo 840 250GB (1 is msata 1 is SSD)
Windows 7 Pro 64-bit

TLDR:

-started freezing & bluescreening, generally while gaming
-laptop started defaulting to onboard video; dev mgr shows the gtx 870m = disabled due to error code 43
-cleaned dust out and verified both fans working
-uninstalled & installed gfx drivers many times
-ran tests on SSDs, memory, cpu - all seem to pass with flying colors
-however, CPU seems to run crazy hot - 80-90 deg C under load
-tried to upgrade from win7pro 64bit to win10, it fails every time due to "an unknown issue"
-please help!

Full Story:

Recently, when gaming the laptop started to freeze sporadically (not just the game, the entire laptop would freeze and I would have to hold the power button down to get it to shut off.) Sometimes it would bluescreen. I also started having weird issues, for example in XCOM 2, my entire game seemed to become corrupt. When I would start the game it would prompt over and over and over to agree to the EULA (the game had been installed for years) and any game I loaded would crash within 2-3 turns, and freeze the entire laptop. I reinstalled the game (including ditching all my saves, etc.) and this did not re-occur. However it continued to freeze occasionally while gaming.

I took the bottom panel off and cleaned it out thoroughly, removing all dust and such from areas where it was trapped (most worryingly, there was a fair amount of dust hidden under this weird sticker that covers one of the heatsink fans, blocking the path the fan is trying to pull air through for that hsf.)

However, this does not seem to have changed anything. At some point I went to launch a game and it was lagging a lot and the graphics looked terrible. I figured out that it was running off the onboard video, not the GeForce GTX 870M graphics card. In Device Manager, the graphics card shows as stopped and says error code 43, basically that windows has stopped the device because it is reporting unknown issues.

I tried uninstalling the drivers and installing new drivers, many times over, and I can get Windows to think the card is okay, but as soon as I try to launch a game from the card it just hangs, Windows does not freeze and the fans spin up like the PC is trying to do something... but the game never launches. Then when I check Device Manager, I find that the GTX 870M has again been turned off citing error code 43.

I spoke to a local technician about sourcing a replacement graphics card, and he seemed convinced that the graphics card is not the actual issue - he thinks it is something else preventing the laptop from properly turning on / using the graphics card. The only evidence I have seen which may support this is the CPU seeming to run hot as heck as discussed below.

The technician suggested using Windows Repair Toolkit which I used to run a variety of tests. The memory passed (main computer ram) with flying colors, as did both hard drives (I also ran Samsung Magician on them and both show as 100% healthy in every category.) The CPU passed the stress test and benchmark with flying colors.

However, the Windows Repair Toolkit panel constantly shows you the CPU temperature, and as I completed various tasks I noticed the CPU hits 80 degrees C under ANY kind of load, like unzipping a zip file = boom you are going from 55 to 80 degrees in 2 seconds. When running the benchmarking / stress tests, it hit 90. That seems crazy high for a cpu in a laptop...??? It is oddly inconsistent though, there were times it would go back down to ~65 while under load. But often even when doing minor tasks, it was hitting ~80.

I tried upgrading Windows to Win 10 a couple of times, but it fails every time very close to the end and says it failed due to an unknown issue, then it reboots and is reverted back to Win7Pro 64bit. ***This is the part I truly cannot understand, because this happens even with the graphics card disabled in device manager and the nvidia drivers uninstalled***

Can anyone suggest how I can further troubleshoot this? What do you think the issue is? I am open to replacing part/s depending on costs; buying an entire new laptop is out of the question at this point.
 
.
started freezing & bluescreening, generally while gaming
-laptop started defaulting to onboard video; dev mgr shows the gtx 870m = disabled due to error code 43
-cleaned dust out and verified both fans working
-uninstalled & installed gfx drivers many times
-ran tests on SSDs, memory, cpu - all seem to pass with flying colors
-however, CPU seems to run crazy hot - 80-90 deg C under load
-tried to upgrade from win7pro 64bit to win10, it fails every time due to "an unknown issue"
-please help!

Those cpu temps are not extremely concerning on a laptop. What are the gpu temps?

Run hw monitor to check the max on both. If both fall below 90c, I would cc clean the gfx drivers and install fresh from nvidia. For the 800 series, you may try a few releases back as well just in case. I've had issues with them and older hardware from them before.

I didn't google the error code, but that may have more help.
 
Thank you all! So generally it seems like the video card is the culprit. Could that explain why the win10 upgrade is failing?

That EuroCom link is great but wowwww they're expensive!! $5-600 for a gpu that is ~6 years old now...? And if you want a new hsf to go with it, that's another $100-150... At that point you are half way to buying a replacement laptop.
 
https://www.ebay.com/sch/i.html?_from=R40&_nkw=mxm+3.0b&_sacat=0&_sop=15

Id prolly contact Eurocom to verify BIOS support for cards not listed on their site, but most likely any MXM 3.0b card would work. You may able to source a reasonably priced replacement on ebay.............

Clevo has been using MXM cards a long time and I'm aware of many success stories involving upgrades/downgrades in their units.
 
As an eBay Associate, HardForum may earn from qualifying purchases.
Wasn’t the 870m known for like mass failing?
 
I have experienced failures more times than not when upgrading from Win7 to Win10. Get yourself a USB stick, and burn a clean copy of WIN10. I am still using old Win7 keys and they are all activating fine. THIS is the website where I get current versions to burn to disk.
 
Thank you all! So generally it seems like the video card is the culprit. Could that explain why the win10 upgrade is failing?

That EuroCom link is great but wowwww they're expensive!! $5-600 for a gpu that is ~6 years old now...? And if you want a new hsf to go with it, that's another $100-150... At that point you are half way to buying a replacement laptop.

Definitely sounds like you're 870m is failing.

You're going to pay a massive premium for an MXM video card, unfortunately. If you really want to keep the laptop, I'd be looking used on Ebay. Double check, but I think you may be stuck at Kepler-level cards or below e.g. 880m. As I recall the memory interface changed on nVidia MXM cards between kepler and fermi. If you can go higher, something like this would give you a big improvement over your 870m for not much cost: https://www.ebay.co.uk/itm/CLEVO-NV...723693?hash=item1a9fa7cbad:g:B5cAAOSwyH9fLrQx

After years of buying Clevo/Sagers for their "upgrade-ability", my conclusion is it's rather cost prohibitive to upgrade anything but the RAM and CPU. You're normally better off buying a new laptop.

Edit: Forget my comment on you being limited to Kepler cards. It looks like from the Eurocom list, you could go all the way to a GTX 1060.
 
Last edited:
Back
Top