Official Clevo X170KM-G Thread

[MiRaGe]

5 hours ago, KEYD said:

Thanks for your feedback, so you mean that Origin is the same laptop/config that was sold by PCSpecialist?

 

By the way, where do you find the stock Clevo bios? On Clevo website seems to be gone...

Here you are the stock 1.07.08:

 

https://www.dropbox.com/scl/fi/y06jw0qqqzufu53wsil04/B10708.zip?rlkey=4g75ke1tzvbq2gvhqafl0a0wu&st=5hwrox5z&dl=0

 

[MaxxD]

13 hours ago, KEYD said:

Thanks for your feedback, so you mean that Origin is the same laptop/config that was sold by PCSpecialist?

 

By the way, where do you find the stock Clevo bios? On Clevo website seems to be gone...

 

Try this:

X170KM-G BIOS 

Name/Password: repo

[MaxxD]

My KM-G computer broke down. After turning on the computer, there is no keyboard lighting and no image. It has been tested, the CPU and GPU circuits are intact, but the keyboard (KBC) is not okay. The chip needs to be replaced and programmed first. I hope it will be fixed and usable again. This is not a small time... Has anyone experienced this before? 🤔😥

[kela-slk]

4 hours ago, MaxxD said:

My KM-G computer broke down. After turning on the computer, there is no keyboard lighting and no image. It has been tested, the CPU and GPU circuits are intact, but the keyboard (KBC) is not okay. The chip needs to be replaced and programmed first. I hope it will be fixed and usable again. This is not a small time... Has anyone experienced this before? 🤔😥

on my 170km shorted power bricks even plug in - i changed ITE5770 multicontroller - this is weak point. about your problem - i dont know but i think it can be Multicontroller

[MaxxD]

13 minutes ago, kela-slk said:

on my 170km shorted power bricks even plug in - i changed ITE5770 multicontroller - this is weak point. about your problem - i dont know but i think it can be Multicontroller

 

yes, the ITE will be replaced and programmed.✌️

[kela-slk]

On 5/30/2025 at 1:12 PM, MaxxD said:

 

yes, the ITE will be replaced and programmed.✌️

Did you check dump or cutting adress from bios? when i replaced New chip i dont flash anything (because not have programm tool for ite microcontrollers) and it working. 

[MaxxD]

On 6/2/2025 at 9:13 AM, kela-slk said:

Did you check dump or cutting adress from bios? when i replaced New chip i dont flash anything (because not have programm tool for ite microcontrollers) and it working. 

 

Unfortunately, there is another problem. It won't be that simple. It should have EC, RAM voltage. CPU and GPU voltage are fine. They have been measured. I hope the experts can fix it.

[Teo]

X170KM-G  WATERCOOLING MOD 

 

I stumble on the internet with this Forum and found a lot tips and help to build my own Clevo X170KM-G.

 

I just wanted to share my own experiences as well, I will try to make it short. 
 

Bought the X170KM body case from ORIGIN with its motherboard for a very good price. I had to slowly get the rest of the parts. I like these type of laptop where every part can be remove or swap, gives you a lot room for improvement. 

 

I thought a lot on doing overcloak since the CPU was capable of, obviously I also had to upgrade the Cooling system. 
 

I had in mind the one in Aliexpress but after watching some comments I opted to make my own. It was too expensive and not so reliable. 
 

I made various sketches and pipe routes to get the best outcome and thermal contact. For test I had an extra heatsink. Bough normal copper pipes from a hardware store instead of the already self made pipe from Ali, those were a bit smaller for my likes (4mm (O.D.)X 3mm (I.D.)) I got something similar as my own laptop. 5mm by 4mm for More area covering. 

First water cooler was fine I could score 15265 points on Cinebench multicore and had 90c ~95c around, drawing 180w ~ 190w. And 83c on GPU scoring 6976 points on Superposition (stock, couldn’t make any proper overcloak I am still new on that area) 

 

I also delided my CPU (11900k) but didn’t got any better cooling performance, was poor, still I didn’t know what was wrong (on idling was hitting 40c ~50c, it wasn’t like that before on the stock heatsink but the watercooler performed well enough).

 

I wasn’t satisfied about the first cooler and decided to add an extra pipe, thought it throughly, could had 2 intake and 2 exits but it wouldn’t be so practical and esthetic wise, so had to be inside with some kind of joint. 
 

On the making of the 2nd watercooler I had to practice a lot soldering with torch, I couldn’t find any “Y” joint that would fit on the pipes, those were bigger. After couple of days I manage to solder it properly (note: with solder iron is also good, very strong but it could desolder when soldering back to the heatsink, soldering onto the heatsink you really need a lot of heat 400c about and heater plate + a heat gun, the heatsink does suck a lot heat).
 

I made all the necessary measurements and bending and flattened the whole pipe, step by step, everything had to be done in order. I didn’t have good tools but good enough to do the job. So when all was done I test it for water pressure and flow, it check out to be good, zero leakage, so time to solder it onto the heatsink. Every step was complex, I didn’t expect to runout of solder paste (low type 138c) but I manage to cover all the pipe underneath and add extra on the side, still wanted to add more, since I don’t have much time I had to do it that way. Had to expend about one week 4 hours per day, no everyday I could work on it.
 

I also wanted to paint it as I did but not this time, looks like this heat proof paint doesn’t adhere well on copper, so just raw watercooler. 

 

Prior making the 2nd watercooler when I was removing I did realize that the CPU and GPU wasn’t getting proper contact as the stock one, had like 1mm of gap, the thermal paste was not spreading well on the edges. So fixing that gap gave me the normal temps on iddle, about 30c. 
 

After mounting the heatsink on the laptop I was surprise that I got extra -10c to -15c on CPU and -30c on GPU. I believe the CPU could still perform better if I apply more solder, the top of the CPU area which is not flat unlike the GPU, Is all pipes. I just run the normal profile as the 1st watercooler to see the differences: 

- CPU did perform better on temps 80c~85c, on the scores was nearly the same, (as I did mention I didn’t change the overcloaking profile.)

- The GPU performed a bit better and had very low temps on Superposition Benchmark.

 

Score: 7691, temps: about 56c max.

 

Well so for now I am leaving it as it is after I reorder the solder paste, I am also looking to solve the fan issue, it does start slow on high temps, I wish it could be programable to start early and gradually increasing speed. Some 3rd party fan software mentioned in the forum didn’t help me yet. Cannot identify all the fans. Maybe I am not so familiar yet. 
 

Note: I did use thermal Grizzly Liquid Metal for the delided CPU and MX6 thermal Paste on top, AAIRHUT thermal pads on the rest. 
 

I also made a 2nd watercooler because the bottom cover wasn’t closing properly, I didn’t foresee it was too long. XD the rear RBG light couldn’t fit in. 

 

I did upgrade the BIOS AND EC from XMG, was more stable, no crashes. 

 

Thank you for all the precious information. 

(removed single pipe cooler)

(Pressing softly no to bed the copper underneath, preheating the heatsink)

(I always used a thermometer to control the heat, I worried about the pipes bursting, but they are pretty good at cooling, even at 400c the heatsink was about 90c, the solder never melted properly even with heat gun and solder iron. (3 solder device at once).

 

(Water test and pressure)

 

(Delided CPU, first watercooler setup)

(Freezemod pump with speed regulator, ThermalTake T1000 water cooler)

(2nd watercooler - the bottom cover close properly now)

 

(Benchmark and temps: to the left single water pipe cooler, to the right double pipe.)
 

(First watercooler finishing)

[Please know that everything was done in a controlled environment to avoid self harm. ]

[Teo]

CLEVO X-170KM-G WATERCOOLER MOD PART-2

 

(Sorry for the misspelling and not so orderly arranged pictures)
 

(First score on Cinebench with the single watercooler pipe)

(some tweaking still learning how to use this software, Intel XTU had a lot bugs and issues, was not friendly)

 

(On the process, joints were to thick)

(best option, hard way)

(Welding with silver rod and torch, took a while to get it right)

 

(Not painted, assembled ready to go)

(copper shim added 30x30, thickness: 0.5mm on GPU, this solved the gap)

(Superposition Benchmark 1080 Extreme test, good temps)

 

[Please know that everything was done in a controlled environment to avoid self harm. ]

[Teo]

Hello everyone 

 

Just a new update, I free the CPU from its limitations to test how good it was the new modified watercooling, I guess is not throttling at anything but I think it didn't change much. 
 

Feels something is holding it back or is just my imagination. There is not thermal limits. 
 

- Undervolted -90.8 mV

- SPEED Shift EPP= 0

- C states= disable

- Prohot= disable 

- Clock mod= 100%
- Turbo Time = Max

- PL1 = 1029w

- PL2 = 1029 w

- Turbo Ratios Limit = Max out

- Overclock = Check

- Ring Down Bin = check >>>>???

 

Under this settings I got 15456 points on cinebench with temps ranging between 85c~ to 95c Max. Drew 200watts max.

 

 

[giltheone]

@Teo WOW! You are braver than I am. By the way, that looks exactly like the EON17-X that I bought last year from a seller at eBay. Which brings me to:

I haven't posted in a while --- it's been a very busy last few months. So I apologize for the long post, but I have some catching up to do.

 

If anybody recalls, I wanted to flash another BIOS to be able to use the full 128GB of RAM. I'll admit I've been a chicken 🐔 but then again, I also use the notebook for work, and if anything would've gone wrong, that would have been a rather unpleasant inconvenience.

I managed to dump the whole Origin UEFI-BIOS by doing what @win32asmguy recommended (and referring to the link he provided too). By comparing to the Clevo BIOS and reading about what the (binary) contents of what's inside the BIOS, I could assemble into a single binary image the segments I extracted using FPTW64.exe (the one appropriate for the Intel ME version 15, which is the one in the Clevo X170KM-G).

 

(I have the whole thing, in case anyone is interested in the EON17-X BIOS. Mind you, I had unlocked the BIOS previously, so before I extracted everything, I set the BIOS to default values, relocked it, and then I extracted all three segments of the image, so I am not entirely sure if it is the exact same one that the computer had before I originally unlocked it.)

Now, here comes the big question that I have, for which I have not yet found an answer:

All the tools and files for flashing within the EFI shell come with both a binary image (e.g. a "X170KM.08" file in the case of the vanilla Clevo BIOS, "X170KM.09RTR7-G2" for one the Schenker BIOSes), some .nsh files (e.g. a "FLASHME.nsh") with several commands for flashing and running some other EFI executables. I get what's going on in the *.nsh files, but all the BIOSes I have checked (e.g. the Clevo and the XMG ones), contain in the "FLASHME.nsh" at the very end a "bios_image_name.efi" executable (i.e. a "X170KM08.efi", or the X170KM09RTR7-G2_xmg.efi" for the Clevo and one of the Schenker ones respectively.)

I have no idea what these last *.efi executables do. I have compared the binaries for several of them (two different XMG ones for two of their BIOSes, and the one for the Clevo vanilla), and though very similar, they are not identical, they all have slight differences from each other (even the two XMG *.efi files are not the same) so they are doing something different.

If I wanted to reflash my original Origin EON17-X BIOS, I don't know what the corresponding "BIOS_NAME".efi should be. The most similar BIOS to the original one that my computer has is the Clevo "X170KM.08" BIOS, but they do have substantial differences, so I don't know if the "X170KM08.efi" for the ""X170KM.08" image would work for the image I have of my Origin BIOS. And yes, I did check the Origin webpage, but the bios they have for the EON17-X is for a later model, one that has the same name, but uses a completely different motherboard --- even the Intel ME version is different for the later "EON17-X", they use version 16. They kept the name of the model, but it's a completely different computer.

So, can anyone give me any pointers about what these "BIOS_NAME".efi files are? Does it even matter if I use a different one when flashing a UEFI-BIOS image?

Edited yesterday at 02:46 PM by giltheone
Edited for clarity

[Teo]

@giltheone Hello 
 

Thank you for the comment,

 

 

 

I see you have been struggling a little bit, before I do any update on the Bios I always have a plan B, I do a full dump of the BIOS with a EEPROM PROGRAMMER, I only forgot the EC chip, I didn’t check if is a programmable chip or not or if is embedded onto the BIOS itself, I haven’t touched any further on this BIOS, but I did once on a ALIENWARE 17 R1 (Ami BIOS too on asus Laptops) which has the same H02 Bios (Insyde), they were separated the BIOS and EC chip.

 

I always dump my BIOS directly from the chip, I am not good making dumps from FTPW64 or any other 3rd application, sometimes they are not so easy to operate or not available to the public or not even compatible, so for safer and faster backup I dump it directly from the chip, it has always worked for me so far. (I got the clamp and many other  tools to help me connect directly, is very cheap) in this case I had to solder it directly. 
 

By the way I did update the BIOS of my laptop from XMG vendor using their files, it has been a while ago, I think you got to make a bootable drive. All the files are there, I think they haven’t take down yet. 
 

So going back to each BIOS update from different brands, I can’t tell you 100% how each of them does it, but I know that (correct me if I am wrong) EFI files must be a part of the BIOS, is not the whole BIOS, so just the necessary bits are replaced and like in the Alienware website they say you can upgrade or downgrade your BIOS at will (I don’t know how is it for AEON17X Laptops on their website if you can revert any updates). Like one time I wanted to update my ALIENWARE and I couldn't update it through the application because of some WINDOWS or UEFI protection so I took some tools like EFI64 or AMITOOL or one of those and search what files were different in each section of the BIOS and just replaced it directly with a HEXEditor, there is a Insyde tool, I used to check from there the EFI files what content or section were been replaced and recreated my on updated dump and then later reprogrammed the whole BIOS onto the chip if didnt work I could just flash back the original dump. 
 

I think XMG BIOS are actually a full BIOS, is not just an update. I really have to check again in my laptop as I have heard they have changed the BIOS structure. So I think you just can’t use your vendors laptop BIOS update and just revert to the original BIOS you had before. That’s why is important you keep that Bios dump you have managed to back up. 

 

I do have the whole dump of my Laptop from ORIGIN if you want to check it out or keep a copy. 

 

BIOS: 1.07.08.OBTOPC 

KBC/EC: 1.07.04

ME FW: 15.0.23.1706.

 

Now I think the ME is within the BIOS.

KBC/EC is separate from the BIOS, in some certain laptops they are within the BIOS. 

I can tell you that so far XMG bios is pretty stable, when I started doing some overclocking or doing some heavy stuff with the original Bios I had a couple of crashes and shutdowns, I heard from some other comments that is a problem ralated to EC sensors that weren’t well programmed, so I guess XMG team fixed those issues and who knows who others already did it, but I haven’t check it out. 

 

I hope you can find your answer, I am sorry I can can’t explain well enough all the technical and programming stuff. 

 

All the Best. 
 

 

[giltheone]

Thank you for your lengthy reply, and it's good to know that the XMG bios is stable. That gives me a bit more confidence to replace what my notebook has with it.

However, a lot of of what you said I already knew. I do have a EEPROM programmer (the CH341A and the software to use it with another computer, plus some decent clips to avoid having to solder/desolder anything), and I know how to back things up and reflash that way. I would rather not do it like that --- I don't have the space to do it, and I have too many cats that get in the way whenever I open up anything electronic. What can I tell you, I like cats, but they can be overly curious at times, and more than once they have interfered with stuff I'm doing --- writing the above post didn't go without a hitch or two. I just don't have a good working space.

(By the way, if you can share your original firmware files, so that I can compare with what I dumped from my notebook, I would really appreciate it.)

On the software side, I also know how to get into the EFI shell to run commands there. That's not what I'm having trouble with. And, yes, the X170KM-G EC firmware is flashed separately from the UEFI-BIOS. That's not something I'm really too worried about.

What I am asking is about what's inside the specific *.efi files that I mention. The firmware packages that I have contain these files and they are run from inside a "flashme.nsh", which is also contained in the firmware package, after the flash command (the "fpt" command in the *.nsh files I quote below does the flashing, and it's also a *.efi executable contained in the package: fpt.efi is the whole name). Here, let me quote you the content a couple of those *.nsh files. This one is for the vanilla Clevo BIOS:
 

@echo                                                     
@echo off                                                 
set BIOSROM X170KM.08  
if '%1' == '' then                                       
goto flash                                                
endif                                                     
set BIOSROM %1                                           
:flash                                                    
# Check the file exist or not.                            
if not exist %BIOSROM% then                             
 echo The %BIOSROM% doesn't exist.                      
 goto end                                                 
endif                                                     
fpt -f %BIOSROM% -a 0 -l 0x1000000 -y                    
:end                                                      
X170KM08.efi      

And here's the flashme.nsh for one of the XMG BIOSes:

@echo                                                     
@echo off                                                 
set BIOSROM X170KM.09RTR7-G2  
if '%1' == '' then                                       
goto flash                                                
endif                                                     
set BIOSROM %1                                           
:flash                                                    
# Check the file exist or not.                            
if not exist %BIOSROM% then                             
 echo The %BIOSROM% doesn't exist.                      
 goto end                                                 
endif                                                     
fpt -f %BIOSROM% -a 0 -l 0x1000000 -y                    
:end                                                      
X170KM09RTR7-G2_xmg.efi   

As you can see, both have an executable *.efi file at the end. I want to know what that does. I want to know its purpose. By using a HEX editor, I can see they are different between brands, and even between different versions of the firmware packages that XMG has for their X170KM-G notebooks. So it's not simply that I want to be able to backup things before I flash something else, I also want to understand what's happening when I run the flashme.nsh batch file.

So, if anyone can enlighten me, I would really be grateful.