Framework Laptop 13 Intel Core Ultra Series 1 BIOS 3.04 Release

The left expansion slot nearest the audio port no longer works after performing this update.

Edit: seems to be working now after disconnecting the battery for a few minutes and reconnecting. weird.

Windows warns me about running this update (3.03) and says it is potentially harmful to my computer. In the box that pops up with the warning it says that the distributor is unknown. Can this be fixed?

is your PC battery at 100%?
i read somewhere that 100% battery causes an issue with bios updates on ubuntu and unbuntu based systems

It’s running in standalone mode so no battery but I was able to get a replacement from support.

(NEVERMIND ON THIS. I swapped the power one more time back to the side it was on when I originally flashed, flashed again, and it updated this time. Doesn’t seem like it should be this complex… :frowning: )

After doing this from Windows (running all 4 steps sequentially, no failures observed), I see two different versions of the PD BIOS when I go into the bios after the update. PD Controller 1 shows 0.0.08 while PD Controller 2 shows 0.0.0A.

I tried switching the power to the other side and running the bios update (part 4) again, but it stays the same. It shows a checkmark on the flash screen by PD2 but just leaves PD1 alone. What can I do to fix this?

1 Like

I hit this too when updating using the UEFI method.

The only way I was able to resolve this was to install Windows and use the Windows updater to reinstall Step4.

I’m also seeing some weird behavior that might be BIOS related.

OS: Manjaro Linux
Kernel: 6.13.0-1-MANJARO

  1. When docked to my Caldigit TS4, I am unable to see the GRUB menu. Both my primary (built-in) and secondary (external) displays remain black, I get no output though both monitors are ā€œonā€ (i.e. not in standby). I can tell that the machine is just sitting at the grub menu because if I hit enter, the machine starts booting normally. If I reboot without being plugged in to the dock, I can reach the GRUB menu no problem. I don’t remember this being a problem before, though I also don’t remember ever needing to break into the GRUB menu, so this might be a red herring.
  2. Also when docked (though only sometimes), plymouthd never quits which holds up several systemd units from completing… I stumbled on this because the KDE Power and Battery menu started reporting that power profiles were not available. For fun, I reinstalled Manjaro using the latest ISO and the issue remains. However, I also reinstalled Manjaro from the ISO I used back in August when I got this board and I can’t seem to replicate this issue there.


I am setting up my brand new Framework 13 with Core Ultra 165H CPU and before I install a system I wanted to upgrade the BIOS. I am following the method described under ā€œLinux/Other/UEFI Shell updateā€ on the bios page.

It started the upgrade process and rebooted several times. However it is now doing the same thing over and over again, showing a progress bar and Retimer Port01: 270 -> Retimer Port01: 270. Then it displays ā€œUpdate Complete!!!ā€ reboots and does the same thing again.

When I remove the USB Stick after the ā€œUpdate Complete!!!ā€ message and go to the BIOS, I see that it updated to Version 3.03 but not 3.04.

How do I update to 3.04 now?

I decided to continue with the OS installation and set up Arch Linux. From there I used fwupdate to update the firmware, and this time it worked flawlessly.

I tried 3 times to update my BIOS but nothing changes. What am I doing wrong? I downloaded and ran the Windows zip file. During the update it said it updated 18 files and a restart popped up once, but I didn’t notice the restart. I did restart manually after the 18 files updated and it instructed me to. The BIOS version and date cames back the same each time.

OS Name Microsoft Windows 11 Home
Version 10.0.26100 Build 26100
System Model Laptop 13 (Intel Core Ultra Series 1)
BIOS Version/Date INSYDE Corp. 03.01, 7/8/2024

Hello with Windows method, I’m stuck on step 2.

It’s giving me:
Update Failed!!!
EFI ERROR CODE: 07
Firmware update failed!!!

Step 1 ran just fine.
I am now on bios 03.03 (was on 03.01 before step 1).

I tried to plug the power cable in another port.
I tried to manually suspend BitLocker.
I always unplug all extensions cards except for the one being used for power.

Currently my firmware versions are (as seen in device manager):

  • CF
  • 85D
  • CF
  • 303

My DIY hardware is:

  • Samsung SSD 990 PRO 4TB
  • Crucial 2x32Go 5600MHz (CT2K32G56C46S5)

Is there a way to overwrite or remove ā€œ.wupd-efi version 1.4ā€ from the bootstrap process? I attempted to update the BIOS using the Software store on Ubuntu (when it was offered), which was apparently a mistake because the updater is in beta, then part of the process got stuck so I cancelled it, and then whole update process failed.

So, at first, when I rebooted, I would see

.wupd-efi version 1.4
WARNING: No updates to process, exiting in 10 seconds…
start_image() returned Invalid Parameter, falling back to default loader

But, after a couple of hours of searching, I managed to find the location of this bootloader image. I removed it from the boot path, so that the image couldn’t be found. Now, it can’t find the image and the screen shows this

Failed to open \EFI\ubuntu\fwupdx64.efi - Not Found
Failed to load image [ ]: Not Found
start_image() returned Not Found, falling back to default loader

It still takes an additional 5 seconds for the ā€œFrameworkā€ splash screen to go away, but it’s a lot faster than the 15 seconds it took previously. My question is, how do I get rid of this seeming ā€œjunk dataā€, so that it gets straight to the OS?

Have you tried updating via the EFI method?

I’m pretty sure I tried both. I did both the EFI method and used the Windows BIOS update tool, because I dual-boot, just to make sure all of the necessary EFI updates were applied correctly and neither of them were corrupted by a partial install. Granted, I did the Windows method after I learned that the EFI shell update is in beta, but still.

Unfortunately, because I was only one revision behind, I’m already on the latest version of the update tool. I don’t know how to delete or reconfigure the fwupd tool, so that it doesn’t spawn at startup. I’ve tried disabling the fwupdate daemon. I tried to configure the bootstrap process in grub (which doesn’t work, because it loads before grub) and installing an additional bootloader just compounds the process. It doesn’t overwrite whatever bootloader fwupd is using. It also doesn’t show up in the Boot Manager.

My suspicion is that it has embedded itself somewhere in the core of the bootloader process, but I don’t understand Linux enough, to meddle with the bootloader without risking breaking my system.

Sorry, I screwed up. Core Ultra’s fan curve is fine, 3.04 is fine. Apparently Core Ultra’s power management reacts differently than earlier generation when it sees weaker PD power source.

My monitor technically provides enough power through PD. But now it appears, depending on the brightness and peripherals attached to the monitor, it might be encroaching into PD’s power quota.

On 12th Gen, performance isn’t constrained when connected to low wattage power source, system will just pull from both the adapter and battery. (Verified with an 18W charger) And since my monitor’s PD output is likely just a little short under very high load situation, nothing was noticeable.

On Core Ultra, it seems like the system proactively limits the processor’s peak power consumption when an insufficient PD source is detected. Subsequently, there’s no reason for the fan to max out. This led me to erroneously assume that the obviously slow fan is leading to thermal throttling and low performance. Attaching an additional adapter appears to resolve this.

I still prefer Core Ultra to behave more like 12th Gen under this specific scenario, but that’s more of a tuning decision that I would personally done differently, than a ā€˜bug’ or issue. Thanks to Framework support’s suggestion of using HWiNFO. Finally put two and two together after closely monitoring HWiNFO’s readout while putting the system under different load.





It seems like we could benefit from a more aggressive fan curve in the next BIOS version?

I’m using/used 11th Gen and 12th Gen extensively and happy with their thermal. (Except 12th Gen 3.09 Beta but that’s already being fixed)

Recently I began also using Core Ultra on 3.04, it seems like both 155H and 125H could use a more responsive fan curve?

Core Ultra seems to be more quiet under moderate sustained load, which is nice. But once the laptop is saturated with heat from that sustained moderate load, and suddenly transition to a full-core 100% load, the fan really takes its sweet time to speed up. Temperature rise, and the reduction in clock-speed and performance is pretty noticeable.

(For example: running a not too busy VMWare VM for 30 minutes, suddenly shut off the VM and hit the CPU with a 100% all-core load in the host. The fan will likely hesitate to speed up past a moderate level, and heat will continue to saturate the laptop lowering the frequency.)

As a result, it always feels like there’s a lot of pent-up heat with Core Ultra boards installed.

11th and 12th Gen are definitely louder, but the fan always jumps up to 100% under heavy load, never once I felt it’s building pent-up heat.

(Windows 11. My Core Ultra arrived with 3.04, therefore unsure about fan behavior with earlier BIOS version)