So I got my new 13 in with the 155H. Which I love BTW!
The one thing that has completely confused me right now is why there appears to be a CRAZY amount of heat generated while using it. Even as I write this post on it, there just seems to be more heat being generated that I would expect for nothing more than a browser currently running.
Basic specs:
155H
32GB memory
2TB NVMe
Linux Mint 22
Kinda curious if anyone else has had issues. I mean this could be a Mint thing, kernel thing, motherboard thing, etc, etc It just seems way weird that this is happening.
Congrats on your new Framework Laptop 13 with the 155H processor. The fans spin up and do a great job of flushing the heat out even at low RPM. I have not been doing much on my 13 and 16 and occasionally put the back of my hand by the exhaust and it is really warm.
I feel the engineering team did a great job pushing the heat out at low RPM instead of making the machine wait until it gets ridiculously hot before barely turning the fans on.
Modern electronics can take much more heat than they did years ago. Though, it always stands that heat is one of fastest ways to degrade electronics over a long period of time.
What do you mean by a crazy amount of heat? Are the sensors showing that the CPU is getting very hot or are you assuming it is very hot because the fan gets very loud?
Also Linux Mint is not an officially supported distro, so if you are using it, you will have to check for drivers and other updates by yourself. I would recommend seeing if you can get a version of Linux Mint with a higher kernel version as Core Ultra Processors are pretty new and old kernels may have some support issues.
If it’s got crazy heat it’s doing something. Try checking load via top and battery discharge rate via powertop. Also battery charging tends to create heat.
In comparison to my other 6 laptops, with the exclusion of my Razer for gaming, this is the first one that kicked fans on for just having Chrome open. (Yes, there are definitely jokes there )
As far as the generated amount of heat, the fans do kick on as they should. For what the laptop was doing tho, they shouldn’t have needed to. I’ll see if I can measure the heat today just for fun.
Outside of this issue last night, yesterday during the day, I had it up and running testing Ansible playbooks for well into 6-7 hours without this issue. And yes, a terminal vs Chrome is a difference in load.
I’m actually going to try and experiment and install Win 11, with full Framework drivers, and see if I get a different result. Even tho I’m a Mint guy, I could try Ubuntu Cinnamon for a bit too if needed.
As far as the kernel, I was told the 6.11, when it comes out, may help a bit. I can try that once it’s out too.
From what I’m gathering, it sounds like this may be unique to me. That’s fine of course but I wanted to see if other folks observed the same behavior or not.
This could also be due to the fan curve set in the Framework EC. Unfortunately, there isn’t a first party option provided by Framework to change it. The only tool available is currently fw-fanctrl.
Also if you are willing to sacrifice a little performance for a quieter laptop, I would recommend changing your performance profile to power efficiency. This usually makes the laptop not turbo as high, reducing heat and fan noise.
OK, so I wiped the drive and installed Windows 10 because Windows 11 blue screened every time I tried to install it.
I’m writing this on the laptop now and I did a rough check of temp and the top side of the bottom of the chassis is registering around 100 F degrees. That seems absurd for just a browser open (with 1 extension for BitWarden) and nothing else. The system is pretty much at idle right now.
Based on the fact that I am getting the same result with 2 different OSs, I’m thinking I want to write into Framework Support to get their thoughts.
40C for the chassis is overly hot, but Windows 10 is not a supported operating system. Did you install the driver package when running Windows 10? The driver installer will not be able to install all the drivers as some depend on Windows 11.
What kind of blue screen are you getting when installing Windows 11 and what method are you using to install it (e.g. Official Windows USB from Media Install Tool, Rufus from FW guide (better because the Official one requires Wi-Fi but there are no Wi-Fi drivers))?
I did download the driver pack for 11 and it installed fine. There were 18 items and it went to the end.
With 11, I would get this:
SYSTEM_THREAD_EXCEPTION_NOT_HANDLED
What failed: Netwtw10.sys
Based on what you’re saying about using the media tool or Rufus, I’m 98% sure I used Rufus. I did try using a USB-C NIC for both. Win 11 still puked, Win 10 used it fine.
I mean I have Win 10 on it now. I have Core Temp up and the current average temp is about 108F right now. I’m typing up this response via my phone so the laptop’s actually doing nothing right now.
EDIT: I’ll recreate the Win 11 USB drive with the MCT and see if I get different results. If that still pukes, I’m going to try Ubuntu Cinnamon (24.04) as Ubuntu is supposed to be an officially supported OS.
This is kind of giving red flags for a faulty cooling system, so if the issues still persist in Ubuntu or Windows 11, I would definitely file a support ticket to get a replacement mainboard.
On the other hand, if you are willing to risk voiding your warranty (not sure if you will for certain) or Framework Support allows you (try to confirm with them first), you can try repasting the CPU yourself.
What is your ambient temp? This isnt an OS issue. You need to actually quantify load and temperature. Install hwinfo, leave machine with nothing open, check for 100% idle in task manager performance tab, check if your battery is charging, check power profile and screenshot your temperatures.
So I have gone through a bunch of tests. As mentioned, I had Linux Mint, Windows 10, Windows 11 and currently Ubuntu Cinnamon on the laptop.
It appears that any version of Ubuntu (which Mint is based on) and Windows 10 gives me the issue. Windows 11 seemed to be the most consistent with regards to heat generation. It created next to none.
I even went as far as installing Steam and playing #Drive Rally for about an hour to see if it would get crazy hot like it was, and it didn’t at all.
So I can go back and try basic, default Ubuntu (instead of the Cinnamon edition) and see where that goes but from what I can currently determine, Windows 11 appears to be the only OS for me that isn’t making the thing cook.
EDIT: The laptop has not been docked to anything. When I was testing all this I was on battery. I only charged it when I was not using it as I knew that would generate more heat. I even ran Furmark in both Win11 and Ubuntu Cinnamon and got 2 different results when it came to heat – Win11 ran cooler.
Unfortunately there is very little to go on without any data. What do you mean by heat in this case? Chassis temperature? Core temperatures?
Higher discharge rate = more energy being used = more heat being generated. You can’t get more heat without using more energy. So if you check your battery discharge rate and it’s the same then the only difference could be how much heat is being moved out of the chassis via the fan. Is the fan behavior different in Windows 11? Does Windows 11 have Variable Refresh Rate turned on by default but not Linux?
When this started I opened a case with Framework Support. They have been SUPER nice and cool (no puns intended here ) with the troubleshooting. After some back and forth and me getting all the info they needed, they did determine the mainboard was bad and have agreed to replace it.
They asked me to use Windows 11 and run Cinebench while having HWInfo running. I did videos for them so they could see what was happening. Based on HWInfo, the temps on some cores hit 101 C during the tests. That’s a damn hot machine. I also ran 4-5 tests for them in various scenarios.
As @Elliot_Lu mentioned, I could reapply thermal paste but this is a new board so I was reluctant in doing so. Putting on new paste could just be a bandaid for, what has now been determined as, a bad board.
Just to not leave anyone hanging, @parawizard you had a lot of good info as well, thank you for that. When I was running Cinebench for Support, HWInfo was showing somewhere between 45-49 C. When Cinebench started, the tems rose to mid 90s as well as 2 cores hitting that 101 I mentioned. The throttling of the board went to somewhere around 12% from what I remember.
You also have a good thought on the battery tho. After the board replacement, if I still notice some oddities, I have 2 other FW batteries I can try to see if that may also be having an impact.
Ultimately I am hoping it’s just a mainboard issue only. These are brand new boards, brand new processors and a unique form factor. There are always going to be issues with a new rollout like this, ask any video game developer.
IMO, Framework values their customers based on my experience with them. That’s part of the reason I bought a whole second unit instead of just upgrading the 11th gen i5 I have. These 2 probably/hopefully won’t be the last ones either.
I have not been doing much on my 13 and 16 and occasionally put the back of my hand by the exhaust and it is really warm. 
did a great job pushing the heat out at low RPM instead of making the machine wait until it gets ridiculously hot before barely turning the fans on. 
Though, it always stands that heat is one of fastest ways to degrade electronics over a long period of time. 
)
.
) with the troubleshooting. After some back and forth and me getting all the info they needed, they did determine the mainboard was bad and have agreed to replace it.