*** Official Clevo X170SM-G/Sager NP9670M Owner's Lounge ***

That looks nice for a bag, how is the quality and is the shoulder strap supportive?

 

Screws for cpu arrived ( finally ) so I could put them on with anti locker ( or however it's called in english ) application and washers, besides looking far cleaner than it did previously, without re-paste or any further change, I've seen a substantial decrease in temperatures with equal or more wattage used ( topping out at 360W more or less ) . In total, highest spike at 90 C on full combined load on cpu and something silly like 68 C on Gpu ( spike again ).
View attachment 190193

The damage you can see on the cpu heatsink is courtesy to the old screws which would not come off the first time, fortunately it's only the coating that has been slightly scraped off and nothing worse, the choice was to either let the cpu get to 100 C or to try and get them off.
Thanks again to Fox and everyone who suggested the various mods to help with the quest.


Just as a funny addition, will post the CPU only stress test result :

CPU at stock frequencies.
EDIT : ( removed the gpu mod bit as too much pressure on the vapor chamber can cause damage )

 

Quality seems good. Not sure what you mean on the shoulder strap being supportive? I just throw it over my shoulder and it's fine. Doesn't feel like it's gonna snap or anything.

 

You might get away with this for a while but in the long run this can potentially damage your motherboard. There should at least be some wiggle room for the heat sink as the metal expands when it gets hot. The springs provide just that. If you completely eliminate the springs and even apply thread locker on the screws then after repeated hot-cold cycles, components might start to wear out. A better solution could be to put washers under the factory springs to increase the mounting pressure without risking damage.

 

A better solution which isn't available given the sorry state of the cpu screws ( 1 specifically ) which aren't replaceable unless you buy an entire new heatsink.
Also, it's not like I tightened the new solution to the extreme, but the old stock solution was not solid, considering how bad I also found the thermal paste afterwards. Also, I tried your solution for weeks before this and didn't change anything over stock.
What I could try is using spring washers, which I do have in fact.

 

The factory screws are sturdy enough that they can be used with washers, as long as you use a quality screw driver and don't end up rounding their heads. Also, if that's the only thing you did over factory then you might want to delid that beast as well. Plus stock thermal paste is garbage. I'd still recommend having at least some margin for the metal to contract and expand without restriction.

 

No, I changed the thermal paste various times for obvious reasons otherwise I wouldn't have been able to find out about the screws, which originally were so tightly put that I had to resort to extreme solutions to remove it.
I do plan to delid in the future yes but first I wanted a secure solution for how the heatsink was held in place.

 

The coating stops the copper tarnishing so I would not worry too much about that.

 

Nothing a black "Sharpie" marker can't fix if the aesthetic is bothersome.

 

Some straps for the bags do not have good padding like the backpacks so it puts stress on the shoulder.

 

Go for red, I makes the system faster

 

Red could indicate overheating

 

Yes, I was thinking that about red, but if you use Liquid Paper or White Out (typing correction fluid) to make racing stripes it would go even faster.

 

Is it possible ceramic paint could have an impact on temps?

 

It certainly may, as it is used as a finish on automotive exhaust headers, but whether or not it is enough to be meaningful is unknown. I wouldn't bet my paycheck on it.

 

That's more for heat resistance and corrosion protection, any additional layer would insulate.

 

The thought was if you can use ceramics to protect the space shuttle from excessive heat... why not a few laptop components.

I don't even own one of these laptops. I'm still using a ten year old ASUS G73JW.

I just find the conversation intriguing.

I am almost certain that more innovation comes from the users than the engineers.

 

Yes, and if you consider that it fits with what Meaker was saying. It's more of a protective insulating layer rather than heat diffuser. But, it is an interesting thought no less. It could be useful to shield components in a laptop, but not likely to enhance thermal management of the parts that are creating the heat.

 

Just cerakote the parts

 

Or just get a @Prema , @Mr. Fox , @jc_denton , and @Papusan X170 from @Donald@zTecpc and avoid all the hard work.

On another subject, found that me running the Chessbase program (from https://en.chessbase.com/pages/download), and having the chess engine FishStock ( https://stockfishchess.org/download/) running as a kibitzer (analyzing the moves), my X170 gets toasty with temperatures up to mid 90 C. My ambient temperature is about 68 Fahrenheit (20 C). Had the fans running at full with Fn +1 for about 10 minutes.

Maybe next time, I'll undervolt at Bios and see it it makes any difference. Apart from that, while running all my regular programs (Photoshop, Acrobat, Word, Chrome, Excel), I don't even remember fans turning on at all. :-0

 

Imagine running Alpha Zero !

 

Last I heard, Alpha Zero is now co-piloting a U2 Spy Plane. Truth!

 

I am new to this XMP thing for the ram. My new ram came and I went with Crucial Ballistix 3200 MHz DDR4 DRAM Laptop Gaming Memory Kit 64GB (32GBx2) CL16 BL2K32G32C16S4B. So the laptop see the ram fine and I am getting 64gb stable. but it is at the lower 2667 speed. How do I get it to run at 3200?

 

Hi guys, I have an issue when connecting external monitors to my X170SM-G. Could you guys help me to pinpoint the issue?

Problem description: I have 2 identical brand-new external Lenovo ThinkVision monitors (S28U-10) connected to my laptop. Both are connected with identical DP to mini-DP cables. One of the Monitors (let's call it Monitor 1) randomly turns black for one or a few seconds. This happens randomly once or sometimes multiple times one after another. It is very random meaning sometimes it does not occur for hours or only very rarely, and sometimes the monitor is unusable as it happens repeatedly relatively frequent.

Testing/Trouble-shooting:

1. I swapped cable 1 from Monitor 1 with cable 2 from Monitor 2 to test whether one of the cables is defective. It turns out that after changing cables, Monitor 1 still has the issue, Monitor 2 is not affected. Therefore, I think the cables are okay.
   
2. To test whether one of the mini-DisplayPorts outputs of the Laptop is defective, I swapped the mini-Displayport outputs to which the Monitors cables are connected. Afterwards, still Monitor 1 is affected. Therefore, I think the mini-DP outputs of the laptop are okay.
   
3. I installed the WHQL drivers from the Lenovo website for the specific monitor model S28U-10 instead of the Generic PnP Monitor drivers. The issue was not fixed.
   
4. Google-solutions recommend to set "link state power management pci express" to "off". I've already done this previously. So I can't do anything more here.

Possible causes:

1. Drivers / Software: If a driver or a program causes the issue, shouldn't it affect both monitors and/or also the build-in display? The latest NVidia drivers are installed.
   
2. Hardware component of the X170SM-G: If anything inside the laptop were faulty and specifically caused an issue for Monitor 1, shouldn't it change and affect Monitor 2 if I swap the output for Monitor 1 and Monitor 2 around?
   
3. Defective Monitor: Both monitors are brandnew, however one of them could be faulty. But before I got the monitors, I had two older monitors connected, one via VGA- mini DP adapter and a second monitor via DVI-HDMI adapter. The latter connected via HDMI showed similar issues (occasional blackout for a second or turning black and requiring the monitor being manually turned off and on before showing an image again).

I would be very grateful for any advice,
Cheers, J

 

I am not 100% sure, but I can tell you that the HP Elitebooks we have deployed at work have issues with the wrong displayport cables for users running 3 monitors off them so it could be something with the cables even though they work on say a different laptop. I noticed that you left out the GPU that you are using? because you are right, it could be a driver issue also. I just ordered 2 mini DP to DP cables for my laptop and I can see if they have issues, but it will be 10 days before they get here because of Christmas.

 

Hi, thanks for your reply. That would be an easy fix if its the cables. Are there any renowned manufacturers from which I could buy some cables and test? Most of the cables from amazon are no-name cables from China as far as I can see. Hopefully there is nothing seriously wrong with my cables, I've read horror stories that people got displayport to mini-displayport cables with wrong pin assignment causing GPU damage. Gonna double check and measure whether PIN20 is locked as supposed. But on the other hand I already had some issues with another monitor that was connected via DVI-HDMI cable.

*Edit*: I looked up some higher quality cables from Club3D. They got VESA certification and good reviews on amazon. Are they any good? They don't have direct display port to mini displayport cables, so I am considering to get a standard displayport cable including a displayport to mini displayport adapter.

If anybody has some more ideas how to fix that issue, please let me know!! I am thankful for any tipps!

I am not an expert, but as far as I read in the forum here, Memory modules with XMP profiles cannot run at 3200MHz with the stock BIOSS. If you run the stock BIOSS you need Memory modules that support 3200MHz via a JEDEC profile. But I think I've read that the PREMA BIOSS allows to run Memory modules according to XMP profiles. But PREMA BIOSS is only available from resellers that are PREMA partners.

 

Right and I haven't heard anything about a Prema bios for this laptop so far. If someone knows how to get it to work, I would like to know.

 

At these heat levels managed airflow does a better job.

 

Get better connecting cables. Crap cheap cables will do exactly that. I have experienced it before with my 144 Hz Prism+ 1440p curved monitor as well as my Asus 240Hz one and the reason was a crappy HDMI cable that didn't support a high enough refresh rate. Choose a cable that is designed to support high refresh rate because cheap cables cannot keep up with the signal data (poor shielding across different cables internally) and end up distorting the data which causes fluctuations, horizontal blue lines and even intermittent blackouts. If you are using some sort of multiplexing device to connect both monitors simultaneously to the same output port, get one that supports high enough refresh rates and has stable power output that is high enough or it won't work. I suggest that you connect one via HDMI and the other via mini display port, using high quality cables. Looks don't matter when it comes to cables btw. I have noticed that a much cheaper looking cable which is quite light weight but rated for higher refresh rates, works flawlessly as compared to a hefty dud that only supports basic refresh rates.

 

Thanks for your advice! I ordered now VESA-certified cables from Club3D. According to reviews they are very good. Hope this solves the problem

 

Keep it to the length you need as well.

 

Has anyone considered any serious heatsink modding?I kinda got curious after watching Mr. Fox's video

 

The thermal performance is already outstanding with the vapor chamber. The only thing that needs tweaking is the mounting pressure which is doable. Why would you want to mod the heat sink? There is always going to be a form factor bound limitation in place no matter how beefed up the heat sink becomes.

 

the vapor chamber is nowhere near the messier part of the whole cooling system though , I can't say cpu cooling is optimal tbf

 

It is VERY optimal. With full fans my CPU runs at 37 degrees without aircon. If I turn on the aircon, it drops even further. I call that optimal cooling. Besides, you cannot expect a 10900K to run with an air cooler like it would run with a water block. Form factor is a real limitation but it gives 0 troubles during practical usage and gaming (even benching is as good as an air-cooled desktop rig). You could weld a vapor chamber on top of the CPU heat sink if you want to but not sure how much it will help considering the form factor.

 

I'm running a 10600k, and I don't consider Idle temp with full fans as that is not a scenario I would find myself in.

 

If you want idle temps as low as possible i'd recommend a custom dock with high pressure fans blowing on it, you'll have some airflow through it but much less noise.

 

Honestly mine sits on a laptop cooling pad with the bottom off it shaves off around 5 to 7c but I still thermal throttle when benchmarking only gaming I stay around 76c to 80c at 5ghz all core. Unless I'm playing cyberpunk or r2d2 then I thermal throttle hard too stays at 3.6ghz entire gaming and never boost past it. This is with all the mods I've done. No xtu installed no cccancer I use quick cpu. Even with undervolting 75mv which is my limit anymore and it became unstable. I used @S.K thermal pad mod and used @Mr. Fox guide to delid and relid with lm and some of his mods for more pressure and even used the rockit Cooper ihs. It's great idle for sure but in some games it sucks. Idle it's around 29 to 32c but anything else runs hot. Gpu stays frosty always 72c no more at 1905 stable. Minor oc 45mhz and 300mhz memory max oc was 110 with 500mhz but I don't need the extra heat

 

On my own machine I tuned it so cyberpunk does not kick in the full fan speed so my wife gets a bit of a break. 4.4Ghz all core on my 8086k and 1900Mhz ish on my 1080.

 

Rockit cool IHS is thicker than stock and will alter thermal pad profile on CPU VRMs. Also the z-height will not allow the heat sink to sit 100% flat on the IHS which will inhibit heat dissipation which is why I strongly recommend changing it back to stock. Also, 72 is too high for the GPU. Mine runs at 2040 MHz with GPU Memory at 7700 MHz stable and the max i get is 64 degrees Celsius consistent with custom fan profile. I suspect your custom IHS is ruining your GPU temps too.

 

The Rockitcool IHS is the same height as the stock IHS, it just has a larger contact surface area.

 

That is not true. I have measured it with a micrometer screw gauge and there is significant difference in height.

 

+/- 0.1mm is not a significant difference.

 

It's a lot more than that. I wonder which IHS are you talking about. 10th gen Rockit Cool IHS is quite thick compared to stock IHS. I have a bunch of them lying around that I have used in various machines at home. The one I tried on this machine had to be lapped for 2 hours before it matched the height profile and it was still no good because the heat sink mating surface is concave and flat IHS does not make good contact from the middle part if lapped. I hope you are measuring the actual thickness of the IHS from where the IHS sits on top of the die to the other side where it interfaces with the cooler and not measuring from lip to the top. And like I said, your temps are not as good as they should have been and something has to be the culprit.

 

I'm using the 10th gen. I put back regular ihs and the heat got worst! Everything is making good contact though but I don't have my machine now I had to send it for repair because my mobo died. Well not exactly but the laptop would not post bios at all when I powered it on. I tried everything to nvram reset did not work. So had to send it in. Btw I did not seal the ihs I left it loose and let clamp clamp down the ihs.

 

You didn't seal it? Well that explains why the mobo died. Reseal it with some rtv silicon before putting it back together as these motherboards can fry the CPU if they burn out. I have a couple of dead 10900K CPUs lying in my drawer, killed by an X170SM-G dead motherboard.
For IHS, again, since I've spent multiple hours fiddling with these 10th gen Rockitcool heat spreaders, unless they have multiple models for 10th gen, I'm warning new buyers to watch out because they are much thicker than stock (at least the ones i bought from Rockitcool). And lapping doesn't work. Hope you get a better replacement.

 

I did try my 10850k in another z490 mobo and it worked though. I swear I'm gonna figure this out I can't with this machine. I thinking when I get it back I'm gonna sell it and see how I can come about getting a ztec.

What's a good price to set on this machine with a 2080s 10850k delidded with rockit cool copper ihs, original included. 16gb 3200mhz ram cl22 300hz screen no g stink as it's not really need do to 300hz modded pads (on heat sensitive surfaces) lol dual 1tb drives raid 0 (nvme 3500 read 3000 write) windows 10 without all the bloatware. 2
Power bricks and holder.. Obsidian tools since its mobo specific included. Guess it's open box since clevo is replacing mobo @Donald@zTecpc @Mr. Fox

 

Did you by any chance change the bios whilst using Obsidian tools? If so, maybe that is why it doesn't boot up

 

I didn't change anything. I had an unlocked bios but my Obsidian tools didn't have bios access as that has been removed. I just had the drivers downloaded thru there. That's it.

 

Exactly. And the reason for the sligthy difference in thickeness is the use with monoblocks. Of course it will be thicker if you re-lid it with a huge load of rtv silicon/glue See... if you put on too much it will affect total thickness but in a negative way. I always let the lid float on top of the pcb.