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

This is ridi

This is ridiculous. I'm waiting to see if the RTX 3080 with 16gb of vram will be available for the x170. Lookin at some of the early benchmark it's not beating a 2080 SUPER by very much. I still would like the RTX 3080 because it would be ideal to run with a 1440P 165hz display. Hopefully Clevo will be able to source these to put in the X170 SMG.

 

Does the Prema bios support XMP for memory? If so, how do you activate it? I couldn't find it in the bios.

 

So I have been seeing rumors on reddit as well as in this thread stating that there is an issue with Kingston HyperX 3200 MHz memory in this machine if all 4 memory slots are filled. This weekend I had the chance to harvest a whole bunch of 3200 MHz memory modules from some of my lab laptop machines and run a little experiment. I plugged a set of 2 x 32 GB 3200 MHz HyperX modules in the bottom memory slots that are beside the NVMe drive bay on the back. I followed it up with a full session of memtest86 and there were no errors (no surprise as it was a known good set of memory already under use in my machine).

Then I stepped up the game by a notch and installed a third module in one of the slots under the keyboard followed by another memtest86 session. This time, I noticed that I received intermittent errors during test 4 and sometimes during test 5. I even noticed that at times the machine halted because the errors came like an avalanche across some consecutive addresses. I then removed all the memory modules and only installed the latest one (the one that gave errors) so that now my machine was running with only 32 GB of memory. I again ran an memtest86 session and surprise surprise... no errors this time! So it was obvious that the particular memory stick was having a hard time running at higher clock speeds but when run alone, it was OK.

I then restored the known good pair of ram back in it's original place and replaced the third module with another one of the same hyperX part followed by memtest86. This time, once again, no errors. I then installed a 4th module and this time memtest86 gave errors again. I then took the third module out (leaving the 4th one in that recently gave errors) and ran memtest86 again and once again no errors! So it was obvious by this point that some memory sticks were unable to run well with the power provided by the machine when all 4 slots were full.

Right now, I am running 4 x 32GB 3200 MHz Kingston HyperX memory sticks on my machine which are a mixed batch of sticks in a way that some of them I bought specifically for this machine but they failed to work when all 4 slots were filled. I swapped them out with the ones that I already had running in other machines that only had single 32GB module in them and kept repeating the process until my machine was stable at 128 GB 3200 MHz with no errors on memtest86.



The take away from this is, unless you are willing to RMA your memory sticks back and forth with Kingston or you are able to buy a large enough batch to play around with, my suggestion is that don't fill all 4 slots with these hyperX 3200 MHz sticks in this machine unless you have prema bios and you can raise the memory voltage until all the sticks are stable (although they work fine when they do). Hope this helps somebody out who is considering buying these memory sticks for this laptop.

 

I was using my machine (from HIDEvolution) for some serious scientific computation and it ran with full fan blast for several hours. Later, I noticed that the 2nd M.2 NVMe drive is not showing up any more. As in, the system simply does not see it even after a restart/refresh hardware. It's as if it isn't there. The drive is only 128 GB and used for a temporary/back up storage, but at only 4 months of intermittent use I can't believe it would be bricked so quickly.

Could the heat from the workout have killed it? I am worried the heat could do the same to my main drive.

 

They don't even try to hide it anymore it's only about being slim and light and deluding customers they get a top drawer card when in fact the performance of what they get is about equal to the two classes lower desktop cards. We used to have 8, 9 or even 10+ lbs machines with up to 18.4" screens, SLI and close to desktop performance and now we have crippled 3070 dies posing as a 3080 in 5 lbs turdbooks. Very depressing to see this.

It is telling that there werent't any big gaming machines announced from MSI, Acer, Asus, Aorus or HP. Too early yet to say that things will stay this way as currently manufacturers will have a hard time getting any high end chips out of Nvidia but I wouldn't be surprised if we do not see much in the way of replacements for a number of manufacturers.

It doesn't help that we are stuck with the current MXM form factor that was not even intended for 200W cards let alone cards that draw 250 or 300W! If the connector doesn't get wider there is only so much that can be done with the ever increasing power draw and bandwidth requirements of the top cards.

 

Always a good idea to monitor your drive temperatures.

That being said I would swap the drive to another slot and vice versa if that is possible.
Could also be the slot itself that has a defect.

 

Guys I’ve had a talk with several of my Chinese suppliers and they indicate a strict ‘May’ release window for the X170KM-G. They failed to comment on specifics for the rtx 30xx mxm release dates but said anywhere from 3 to 6 months from now!!!

I’ve also got news that the vbios on the 3080 is adjustable, you can still flash higher or lower wattage vbioses on it. I’m guessing they have to be original ones with the same hardware ID though.

For that reason I think I’ll buy another x170sm barebones and slap a 2080 inside to be honest...

 

Did you use the heat sink and cover the ssd drive with thermal pads on both top and bottom? If you did, there is no reason for it to burn out as a result of such a brief power session. However, cheap Chinese junk always burns out no matter how good you cool it. Telling you from experience. When it comes to nvme drives, I never go for any other manufacturer but Samsung. I have a big home/work lab that I maintain and over the years I have had nothing but problems with other brands under load. Samsung drives last and so far I haven't had to claim warranty for a single one of them.

 

What tool would I use for the purpose of monitoring the drive temperatures? Thanks!

 

Try SSD-Z.

 

It's always been about being slim and light. However historically you only got the cheaper chips, eg a GT650M in the smaller form factor. What if the 680M chip imposed with the 650m TDP limits performed better. Well then that becomes an option.

 

Desktops don't deliver more than 75W from the slot. MXM is good for 200w and you have the power connector. No need for major change. And you have clock block 3 or maybe 4 from Nvidia. They will restrict boost and max TGP anyway.

Several tools. But try out CRYSTAL DISKINFO

 

Notebook designs are so custom the slot limits are just less important.

 

With the slot + the power connector it wouldn't stop Clevo or Nvidia offering 225-250W. But they don't let you have it. They prefer that you upgrade the whole machine before time instead for prolong the lifespan with maxed out specs. Higher boost clocks let you get more performance and you could potenially have your machine a extra year.

 

Except they’re charging the same as the full fat 680M, despite it not performing close. That’s the crux of the issue.

 

Apart from a few VRMs it's all the same stuff though. Why even both if you are Nvidia at the moment. Just stick 3060 cores in those and sell 3070 desktop cards instead if the profit margins are lower.

 

So by that token, because the hardware is the same, we should pay the same for an 80W 3080M that performs 30%+ worse as we do for a 200W 3080M. I’m sorry, but that is anti-consumer bullsh*t. We are buying the performance, not the name. The companies know full well that a much cheaper 3060M or 3070M unit would perform the same as the Qrippled 3080M.

 

Yeah, the main drive is Samsung but this was just listed by HIDEvolution as: HIDevolution Approved 128 GB M.2 PCIe 3.0 x4 NVMe SS.

Also, no I didn't wrap it with any pads. I am using whatever stock setup they have.

 

I mean they could just not bother, just offer the 3060 in the chassis and not offer the best for the price increase.

They are limited on the cores they produce so they could just sell them in the desktop parts.

You are buying the chip, not the performance typically.

 

Consumers buy the performance, ODMs buy the chip.

 

Nvidia sells the chip, you want ODMs to then sell that chip at a loss? Just get the mobile 3060 variant of the machine instead.

 

That’s exactly what I tell people, stay away from the Max-Q scam.

 

I see it more as going for the top end trim on a car with all the bells and whistles, you do get more for your money but the return on your money is low. In terms of bang for buck usually a middle of the range car does much better.

 

Any chance Cleo will come up with another 15 inch form like the p750? I really love my P870 but carrying it around is a real challenge.

 

It looks like Clevo has abandoned the modular 15” platform, as the NH5xAx is their only current 15” DTR with a desktop Zen2/3 CPU but it has a BGA RTX GPU.

 

A set of very thick properly sized and precut thermal pads come with the machine in the box that can be used for nvme drives. One on the back and one on the top on the square mark that is etched into the heat sink. That allows the heat to be absorbed by the chassis and the heat sink. Without setting that up properly, you don't have a good enough reason to complain if a cheaper drive burned out.

 

How thick are the pads that come with them, they look like 5mm pads? and then a set of 2mm pads?

 

You will have to figure out which one goes where. First drive is a bit lowered into the slot close to the motherboard while other 2 are a bit higher so definitely they will need the thicker pad underneath. Place the pads where you think they seem to fit well and go from there. On the top side however it seems that all the ssd drives have a 2mm square shaped pad.

 

I just looked into the box and I do not see anything that looks like that. (Can you show me a picture?) I see two partially used bottles of thermal paste, probably from the delidding and whatever, and the paperwork and a DVD. But even if there were supposed to be pads in there, why would you say it is not unusual for a drive that is not being used to burn up? The only way it could have been burned up is from heat coming off the CPU during the intense processing, as no files were being written to the back up drive when it died, only the main drive was in use and it is fine.

 

Got my laptop back from repair from rjtech. Seems they changed my mobo and they updated the bios to 1.04 im happy to report that thermal throttle limit has been removed my package hit 94C on 1 core and it did not thermal throttle plus my temps are alot better too while gaming gpu stays around 56-63c and cpu at 4.7 all core (since my bios is no longer unlocked lol and quick cpu doesnt let me change the turbo limits for some reason) stay under 66c

 

new bios version 1.04???????

 

:: I still want to know how many MM they are.....

Typically the thick one faces the motherboard or inside of the laptop and then thinner one is between the nvme and the case. 3 sets came with my laptop for 3 nvme drives.

 

Failure in the silicon of the drive or its own circuitry can burn a drive that's plugging in not doing anything.

 

My Granddaughter Jessica received her X170sm-g on 1/11/21 and so far she LOVES it. Purchased from Donald at ZTECPC after he spent an hour on 2 different days helping us configure it. He was Fantastic ! ZT-X170sm-g with i9-10900k delidded,cherrypicked @5.0 Gz,64gig XMP @cl16, RTX 2080 Super and 3 Samsung SSD’s. Also has the Mr. Fox cooling magic and the Prema Bios. She is a videographer and it has been outstanding using Adobe Premiere. She has used various desktops at her wok place and she says this blows them all away. Fans rarely come on editing/rendering a large raw file(9 days on a mission trip to Ethiopia).YouTube videos and reviews by Mr. FOX , B0B (B.O.A.T.) and advice from Donald@ZTECPC was very helpful and we Thank them all !

 

Strange! I thought my system was ready to go out of the box. I didn't know that I was going to have to install ipad's on the hard drives! I'm going to have to ask HIDEvolutuon about that, if the pads aren't already installed. Does anyone have a picture of where the drives are under the hood so I can find them easily? All I see is heat pipes!

 

I found it. Seems that the backup drive does have the pads on it - its the main drive that doesn't! Is this normal??

 

They gambled on that the thermal stickers for the Samsung drive was enough. And it should be enough. Just add the thermal pad on top of the controller chips as your second ssd. You risk loosing the warranty if you remove the thermal sticker <if you isn't able to reuse it>>>> put it back properly with a possible later warranty claims. Up to you if you want to remove the sticker and use the thermal pad instead. But pads on top of the thermal sticker should still be better than no thermal pads Again... Up to you.

 

Thanks, but I do find it strange that the one which failed is the one with the pads! And after only 4 months of rare use. Maybe the pads are the problem...

 

Nope shouldn't be. New modern ssd's have sensors that throttle the drive to reduce heat. All ssds and HDD drives can fail. Before or later. Often something wrong with the drive if it fail after short time.

If you mean the Toshiba drive...


Many failed in the Alienware's. They cooked them to deaths.... Pads or not didn't matter. But pads should reduce heat and help to keep up the speed (most PCIe ssds start throttle down speed at 70C due heat) http://forum.notebookreview.com/thr...iew-by-ultra-male.801314/page-9#post-10486496

 

Safe operation of a drive cannot be ensured unless it is properly cooled in an environment provided by the machine manufacturer. You never know what happened during your stress test as windows sometimes does squirrely things to the available empty drives in the background at it's free will unless you disable certain background services. If you do not see any thermal pads in the package, you will have to contact the vendor and ask them about it and they should be able to ship them to you separately. I cannot show you the pictures because I have used them all up and they are inside my machine now. They are greyish white coloured square shaped pads and should be in the large zip lock bag of supplies that comes within the box. The instruction manual and driver CD is also in the same zip lock bag.

Or may be the drive itself. I don't know if anyone else out there uses their machines the way I do where a minimum of 8 virtual machines are running in parallel all the time except for the few hours when I am gaming. My samsung drives are running like a clock and I have never even experienced a single hiccup. Just for your info, my machine has not turned off during the last 2 months or so ever since I received it. I use it as a work station.

No it's not normal. Whoever installed that drive forgot to put the pads on it. If something happened to the unpadded drive, you can file a claim and get a new drive. In your case, since one of your drives burned out, you can still claim warranty (I assume it was brand new). Even the cheapest ones offer some sort of warranty period.

Yes, that's correct.

 

It was the drive with pads, the Toshiba, that died. I moved it to another spot and after startup it is still not recognized or seen. i had some data on there but I guess it is just gone now? The drive looks fine to visual inspection - no burn marks or whatever. Is there anything else I can try to get it back?

 

I cannot imagine that it is very good to have a pad on one part of that Toshiba SSD and not on the other, might lead to a temperature differential?

As for reliability of Toshiba NVME SSDs these came standard with a number of MSI laptps and I have been using those for several years now and I never had one of them fail on me. That being said I always had a pad and some kind of heatsink attached to them and a temperature monitoring program as they can run quite hot and exceed 70 degrees in operation.

For temperature control I like to use HD Sentinel by the way. Love the interface and it has the option to show the temperatures of all drives n the tray. I use a family licence that gives me the professional version for 5 computers and allows me to monitor more than 40 drives with a lifetime license. I consider it money well spent as it also is quite good at monitoring drive health and I am obsessive like that.
https://www.hdsentinel.com/
https://www.hdsentinel.com/store.php

In the case of the Toshiba it would probably have meant a red blinking icon in the system tray and a notification the nature of which you can define to a degree.
This is from a small server I just set up, the red 50 is a Toshiba SSD by the way:



Probably overkill for most and in that case there are free options that have already been mentioned in this thread.

 

Mr. Donald is a real unicorn!

 

So my previous laptop I had a 1TB boot drive (MSata) and then 2x 2.5 inch hard drive bays in raid 0 ( I wanted 1 logical drive) for storage. Then I had a DVD drive adapter for a 3rd physical hard drive/ssd. ultimately I had 6TB of storage and 1TB for OS.

This X170 is where I am finding storage confusing because of nvme drives. So I am using an Evo 970 Plus 2tb for my boot drive (TLC)? but now I need some nvme for storage, and I need high capacity 2x 2TB or 2x4tb but from what I am seeing 4TB nvme is only QLC for now, TLC is coming, but I want to avoid QLC for mass storage?

This is a problem.

 

In mine, i'm using a 1TB M.2 SATA for the boot drive (and documents), 2x2TB in RAID0 as a large drive and a single 2TB for other stuff.

Not sure why you don't just do that?

 

Hey there,

If I do not intend to overclock, what sort of thermal performance should I expect as a simple daily driver? As a workstation? For example, this one:

https://www.sagernotebook.com/Notebook-NP9670M-G2.html

Presuming the only upgrades on the base would be
64GB RAM
2x 2TB hard drive
Win Pro
IC Diamond Thermal Compound

Earlier in the year I tried out a similar build with a thinner bodied machine, and literally it couldn't do anything at all, even copy and paste files without kicking up to 80C and hitting thermal throttling, and running max fans to just open a text editor.

Does this thicker bodied machine handle thermals well enough that it doesn't need to run the fans at 100% to open notepad, let alone run photoshop or Unity3D?

 

I just updated to the 461.09 Nvidia drivers (dated 7th of January) and right after it the colors of the monitor feel different. I can't describe how though. Has anybody experienced this? I'm on the 4K panel.

 

Yes sir. New mobo updated bios. 1.04 is not up on clevos site yet. 1.04 is short of the sn and all the number crap they call it.every is great except thunderbolt my thunderbolt is greed out even thought I've tried everything to get it to work. Updated the driver and app it's greyed and cannot be activated.

Update: thunderbolt issue fixed it was disabled in bios has to turn it on lol. No idea why a port would come disabled from start

 

That is not the issue, I don't want to store large files for editing on QLC nvme's if their life span is no so great, a lot of read and writes.

 

Was it from https://ztecpc.com/?

Is this the model?: https://ztecpc.com/product-detail/81032/zTecpc zT-X170SM-G-8 Gaming Laptop