*** MSI 16L13 (Eurocom Tornado F5)/EVOC 16L-G-1080 15.6" Owner's Lounge ***

That is strange that your new memory is failing as well. I let my system run memtest86 for hours a while back and had no errors.. I suppose it could be a CPU issue now? I'll let the others chime in as i'm not sure what it could be but even testing memory the CPU needs to be used and it will be stressed as well in the process.

Edit: Also, try to use the program @Mr. Fox mentioned above and see what happens..

 

We just added new video showing how to install memory under the keyboard: . Enjoy it.

 

If you are running the CPU undervolted too much that can produce errors with memory. Set stock voltage on the CPU and test again. Also, test with TM5 if you have not. As it runs in Windows it will sometimes find errors that Memtest86 misses running in DOS mode.

 

I don't believe he's running any undervolt.. all stock and getting mem errors. Two different sets of memory.

 

I repasted.. yet again.. Like no difference in temperatures.. Make sure I covered all angles and did the criss cross method, etc. etc.... Oh well lol. However, it looks like I forgot to tighten the VRM bolts down last time I repasted so many they were overheating causing the crashing? Time to test!

 

Nope, still locking up at -150 at 4ghz even.. This is ridiculous.

 

Stop talking to yourself!! jk
Well maybe .140v

 

You can shave of a few degrees in idle, but thats it. The gains are small

 

Sorry for the double post http://forum.notebookreview.com/threads/clevo-overclockers-lounge.788975/page-647#post-10389079

 

When I pulled my stuff apart my paste wasn't flat or anything weird.. It was definitely squished out some like it should have been. Shrug, maybe there was a bad batch of the stuff. What I did notice though is that the consistency after a month was kind of wet looking, definitely a different look to it from other pastes. I need to order more now though, completely out!

 

I've seen a few with not so good experience with Kryonaut. I haven't seen the same with Gelid Extreme/Icd. Maybe a bad batch, but we can't know for sure.

 

I just repasted with Kryonaut... fingers crossed it lasts at least a few months.

 

I'm running Kryonaut since 8 months without degradation in temps nor performance!

 

What paste are you using

 

Kryonaut, as discussed.

 

Well I repasted again with full CLU on the die and on the IHS.. I saw where some people where applying a light coat of the liquid to the upper side of the heatsink so that it's liquid-to-liquid and it like it's making a pretty significant difference on my temps now.. I'm at -145/-145 right now and about another 4-5c cooler than before..

I also had a weird heatsoak problem where the CPU sits under the keyboard and it seems to be a bit cooler to the touch now to. Seems like I made a better seal finally.

Edit: Just kidding the heatsoak comes back over time.. i guess theres nothing you can do to fix that, the cpu are just gets very hot and bleeds through the keyboard. The GT62VR doesn't do this which is strange when you realize both CPUs are drawing around the same watts at idle but this one still bleeds heat to the right side of the keyboard.

 

For someone who has the i7-6700 "non-k" cpu, how were temps at stock? I'm seriously considering upgrading to the i7-7700, which is rumored to have the same TDP (65W).

 

I don't know but with liquid and a de-lid on my 6700K I still can easily get to 100c with wPrime, Realbench, prime95... OCCT runs a little cooler. And even with -140 with Realbench I get to 94c at certain points of the test then it backs down to the high 80s.. Pretty crazy really.

Granted, when you're pushing 90+ watts.. this isn't surprising. I feel like the F5 CPU cooler setup is better suited to up to 60 watts before it gets too oversaturated with heat when using auto fans. You can drop an easy 12c or more on max fans and everything is fine.

I used to think you can compare to other laptops but you can't because of fan curves alone.. CFM of fans, how robust the heat pipes/heatsink setups are, etc... I just know that even at almost 100c, the auto fan curves are very quiet still.. and it still takes about 5+ minutes before it finally gets up into the 90s on my system. The upside, i'll hardly ever be using software that will stress my machine like that so my temps roughly stick around in the 70C range in games.

When I do my rendering and encoding, it will sit in the 80s at 4ghz with an undervolt.. hardly scary for a laptop, but my mind is all screwed up and I expect it to run a lot cooler than it does but I have to step back and breath and remember this is a laptop running at 4ghz and doing it well. You can overclock our machines, but you'll have to run max fans to do it.

 

Liquid metal all the way. As always
Improved Wprime 32M score - 4.137 sec. We socket machine enthusiasts have to stick together
http://hwbot.org/submission/3370273_papusan_wprime___32m_core_i7_6700k_4sec_137ms?recalculate=true

 

I get 83C with running BOINC at 100% without undervolt running at 3.7 with auto fans. With full fans its at 76C.

With undervolt of .155v and auto fans its at 67C, with full fans at 63c. I let it sit for 5 minutes between each test.

 

Are you running BOINC CPU and GPU CUDA tasks at the same time?

Which Project(s) are you running tasks from? SETI has CPU and CUDA tasks available.

 

Wow, awesome temps! Looks like I'll definitely be picking up a 7700 if we get a BIOS update.

 

It seems after 20 minutes of full fan at 100% at .155v undervolt it drops down to 59C, I forgot to put it back to auto fan.

 

No just processor, World Community Grid does have any GPU work units right now. Its been over a year since I had one.

 

SETI@Home and SETI@Home Beta have CPU and CUDA tasks available last time I checked.

 

I dont like running work units on the GPU. I once did the the graphics card only lasted 9 months at 100% when wasnt used. I had to send $450 on a new 970m but sold the broken one for $110 on fleabay.

I am running the stock thermal paste from Eurocom.

I like running world community grid since a friend of my died from cancer and I prefer to run the cancer work units.
And over at the techpowerup forums, they have a daily contest on who gets the most points wins a piece of virtual pie.

 

That's too bad, I have run both CPU and GPU for years and not lost either.

But, losing a GPU would tend to keep me to short term load, temperature, and stress testing only

If you do want to run long term, running at less than 100% is a good idea, keeping the GPU running well below it's limits. Same for the CPU too.

Watch those temps

 

I dont think humans are ready to learn about aliens yet, we barely get along with certain aliens in the world.

 

Do you run BOINC when you play game.

I still run my and play a game called war thunder and still get 60fps and without any hiccups. My last gaming laptop it would have some hiccups so I had it set to 50%.

 

It's a numerical task to run on CPU and GPU for stress testing, no special "belief's" are required, the code runs as is either way

 

My GT80 SLI-263 handled 2 x CUDA + 1 x CPU tasks while streaming videos, browsing, running VM's doing the same things for days at a time.

I wouldn't game and run 3x 100% heavy loads on a laptop, it's unnecessary. I prefer to run games without anything else running.

I did it a few times for fun, and it ran great, but the loud fan noise - not normally needed for gaming only wasn't fun.

I don't think your F5 will have any trouble doing 1 x CUDA + 1 x CPU tasks.

I don't run BOINC 24x7. I don't like living with constant loud fan noise. Brief short term testing is ok, it's part of initial testing for acceptance of purchase.

When I run distributed compute loads, I prefer to do so on servers with excellent cooling in distant environments

 

There are other BIONC Projects with CUDA tasks:

BOINC CUDA - GPU computing
https://boinc.berkeley.edu/wiki/GPU_computing

Attach to projects with GPU applications
Projects with NVIDIA applications:

• Collatz Conjecture (Windows and Mac OS X on Intel)
• DistrRTgen (Windows, Linux 64bit)
• Einstein@home (Linux, Windows and Mac OS X on Intel)
• GPUgrid.net (Linux 64bit and Windows 64bit or 32bit, requires a video card with Compute Capability 1.3 (CC1.3) or higher)
• Milkyway@home (Double precision GPU required, so compute capability 1.3 or higher, meaning a Geforce GTX 260 or better) (Linux 64bit and Windows)
• Moo! (Driver 256.00 or better, Compute Capability 1.0 or higher, Minimum device memory 384 MB - http://moowrap.net/forum_thread.php?id=16)
• PrimeGrid (Proth Prime Search (Sieve), Linux 32bit, Linux 64bit Windows and Mac OS X on Intel; Cullen/Woodall Prime Search (Sieve), Linux 32bit, Linux 64bit, Windows and Mac OS X on Intel)
• SETI@home (Windows and Linux only)

Projects with ATI/AMD applications:

• Collatz Conjecture (Windows, Windows 64bit, Linux 64bit)
• DNETC@Home (Windows and Linux 32bit) Project retired
• Einstein (Linux, Windows on Intel; Mac OS X under active development)
• Milkyway@home (OpenCL support and Double precision GPU required, so a Radeon 48xx, 47xx, 58xx, 69xx, FirePro V87xx, FireStream 92xx) (Windows only)
• Moo! (Driver v10.4 or later, ATI Runtime (not older AMD), Minimum device memory 250 MB - http://moowrap.net/forum_thread.php?id=16)
• PrimeGrid (Proth Prime Search (Sieve), Linux 32bit, Linux 64bit and Windows)

 

Another good distributed computing task for running CPU+GPU project is Stanford Folding@Home :

https://folding.stanford.edu/

I had runs on Folding@Home as part of my GT80 SLI-263 testing, with 2x GPU + 1 x CPU tasks.

 

I wont be doing any gpu work units anything soon, esp. with this F5 laptop that if the gpu breaks its over $1000 to get a new one.

I thinking of building a dual socket folding farm for community grid but I'll wait until the amd Zen for servers comes out.

 

That laptop runs kinda hot.

 

Nope, it runs just fine. That's with 2 x 100% GPU and 1 x 100% CPU running for 4 hours with CPU temps in the mid-80's C and GPU temps in the mid-70's, on stock MSI paste.

You don't just look at peak temp's, you look at the average temps over the 4 hour run. And, current temps, as that test was still running when I took the screen shot.

That's why you run a log file when tracking temperature to watch the real-time temps across an entire run, the peaks temps are rare.

If you judge a CPU run only on the peak / Max reading instead of viewing the reading every second or two you are getting a skewed view of the run.

Let's see how hot that F5 runs with 1 x 100% CPU + 1x 100% GPU for 4 hours of FAH

 

The GPU is only $500, not $1000, and it's not gonna break by just using it

Besides it's going to be under warranty for quite a while, you should exercise it early to find any weak components so you can return it, or get it fixed.

Why do you buy high power hardware and then not use it?

 

The warranty is only a year. For me it seem everything breaks down just after the warranty ends.

I have a small business laptop, w230ss, that the graphics card fail twice, the first time under warranty and the 2nd time just a few weeks after warranty ended. I had to buy a new motherboard to replace the gpu. I will be selling it soon since this F5 is going to replace two laptops.

And everywhere I have search those 1070 are easy over $1000.

 

Yup, one's personal experience does strongly influence opinion and future actions.

I guess having worked with many computers over the years, my opinions are formed over a much larger pool of experience.

I don't have a problem with running hardware at it's design limit's - knowing their limits lets me be safe but still optimize the hardware to the maximum of it will do.

And, I know small clusters of failure have their reasons and aren't applicable over all pools or samples of all laptop brands and models.

You had an unfortunate experience, but that's not how GPU's usually perform, just because it happened to you in that way it doesn't extend out as such to your new laptop.

Use that warranty to prove that to yourself, enjoy tuning your hardware to it's fullest - get back on the horse and ride!!

If you plan on keeping that F5 longer than 12 months, you should strongly consider getting an extended warranty plan from that vendor.

 

Those are inflated prices due to the newness of the Pascal GPU's.

The 980m went for a range of $500-750 and that's about what the 1070 should go for. I do barely recall the 980m also being higher in price when it was new and scarce.

 

Run a wPrime 1.55 1024 (make sure to run as admin and click advanced and set threads to 8).. And upload a screenshot of your temps + the wprime score please?

 

So would you say that with correct application (CLU on the die as well as IHS), you end up winning around 10 degrees celcius with the delid compared to simply undervolting?

I'm trying to build enough courage to delid mine, since I already have everything needed (tool + clu + kryonaut), but I'm not sure I want to do it for only 5c.

Also, what is the easiest way to protect the VRM during the operation? A remember a video of a guy who used standard "Scotch like" tape to build a grid around the cpu die then removing it after application of the TIM. I'm just scared to short things when putting the IHS back on :s

 

I guess i'm going to reinstall windows for a third time. I'm still getting black screens. I played all yesterday eveningnd after 2-3 hours of benchmarking this morning had 2 black screens back to back with the same generic:

"This indicates that an exception happened while executing a routine that transitions from non-privileged code to privileged code. This appears to be a typical software driver bug and is not likely to be caused by a hardware problem. The crash took place in the windows kernel. Possibly this problem is caused by another driver that cannot be identified at this time."

I don't know what the problem is with my unit. I install windows 10 from a flash drive I created on the windows website. I install the drivers on the eurocom disc that came with my unit and I have nothing but alleged software driver bugs that cause me black screens?

 

Did you get replacement sticks? Is anything overclocked? What does it say in Event Viewer?

Get the latest LAN driver from the Killer website. The one on-disc is an older one that can cause problems. Have you used Driver Verifier before? You might be able to identify which drivers are causing the crash using that tool (but make sure you have a boot disk on-hand).

 

That was my replacement Ram and I couldn't duplicate the single fail in 10 more runs. What are the odds I got 2 bad sets of Ram in a row? What are the chances I could have a ram slot or CPU issue that caused a memtest86 fail? I'd really just like a replacement laptop. I'm tired of troubleshooting. I'm not exactly trained for it. I'm really starting to regret ever making this purchase.


nothing is overclocked
I got the LAN driver straight from killer
I tried using driver verifier and it crashed me to the point where I couldn't boot. Had to recover using a windows recovery file.

 

Don't get too down on yourself bro. Being a computer enthusiast isn't easy, especially on these high end laptops/DTR's. In a way, it is a part of the experience of owning one of these machines. I love the challenge of taking something that has a problem and finding a way to fix it.

Obviously, something is amiss with your machine. You are within your warranty period and I suggest you RMA and see what Eurocom can do. They will have more ram sticks of various companies lying around and if they can reproduce the issue, they will likely replace your machine.

 

It's unlikely, but not impossible that you have two sets of bad RAM. A fault DIMM slot is also rare, but not impossible. I agree... troubleshooting sucks. If you want to continue troubleshooting, read below. Otherwise, I would contact your vendor before too late and inquire about your options. Some times, you can get around fees if you ask to exchange for a different model, rather than straight-up cash return.

What settings did you use with Verifier? Assuming that Verifier was set up correctly, the fact that it locks your OS on boot suggests that software is the culprit behind your issues, not hardware. If this is the case, getting a replacement laptop of the same model won't help if you end up installing everything exactly the same way. Verifier is designed to cause BSODs if there is a driver conflict. Using a program such as BlueScreenViewer or WhoCrashed, which driver does the minidump say is causing the crashes? By the way, if Verifier locks up your system, you don't need to do a system restore. Just use a boot disk to enter safe mode, then disable Verifier using "verifier /reset" in an elevated command prompt.

 

I don't understand how a driver could/should be the problem If I'm only installing the OS and then the drivers included on the Eurocom CD. I would think these drivers are already tested in-house before they ship them out. After a clean install I shouldn't have any drivers that differ from what they have @ Eurocom right? And also My computer black screens so I have to "hard reset it" so when I check WhoCrashed all I see if the power loss failure which I suspect is just from me turning it off and back on and I feel like a piece of hardware is causing the lock-up (gpu).

Does that sound like a reasonable doubt? I just want a working unit. I don't want a refund. If I had a tornado F5 that would operate normally like @Diversion and everyone else's, I would be happy

 

You would think that is the case... I'm not targeting Eurocom specifically, but I've had manufacturers include bad drivers on the included disk before. PCI-E wifi adapters are notorious for this. The software on-disk is usually buggy, and you have to get the updated version from their website to get it working.

Yes, the "power loss" is most likely due to your reset. A freeze with no blue screen can possibly mean a memory issue. Just making sure, double check the voltage is correct for your RAM. If your RAM voltage is fine... this is going beyond what I can advise you. I was hoping Verifier would be able to cause Blue Screens, which would then reveal the culprit. At this point I would post to a dedicated tech-support forum (such as BleepingComputer.com), or contact Eurocom.

 

So a black screen (not blue) with the laptop still running fans and all is indicative of a memory issue? I was thinking it might be a GPU issue. I'll double check my RAM tonight before continuing with the RMA.