Sorry if i post again (this time in a different category) but i need to have some answers, currently the support did not answer yet and i need to decide if i have to return the item (bougth the 20 december 2025, delivered the 02 january 2026) so the 30 days return window is expiring
I have an issue with my brand new Framework Laptop 16
When Gaming or doing something intensive (like a stress test) the system sets the PROCHOT flag to 1 despite the CPU temperature being 85 degrees celsius or less.
I tried to set a TDP limit for the CPU to around 30W but the flag is set also at 60 degrees celsius.
I use a 180W charger and the PROCHOT flag is set after 5-10 minutes.
How can i solve this issue? I already created an issue on the Framework Git for maybe bios related issue.
When the PROCHOT flag is raised the system goes around 25W consumption at 2500 MHz frequency.
To reproduce i can start OCCT and perform a FULL Load test with the discrete GPU and after 5-10 minutes the issue persists.
The system is brand new and uses PTM by default. I don’t know if i have a defective unit or something else
Another thing i notice just now if that when i perform a stress test the integrated GPU (Radeon 780M) thermal throttle itself instantly.
I have reported on PROCHOT being triggered all the time in the 16 for probably a year now. Best response we have received from the firmware developer is to just ignore this flag. I don’t agree with that and it makes troubleshooting issues more difficult. But good luck, it’s been this way since release and I don’t see them taking the issue seriously enough to fix.
So basically i am screwed? I have spent 2.5k Euros for a machine that refuses to fully use the hardware? I can’t even downgrade to older BIOS like 3.05 because i am using the RTX 5070 module
For now, yes. I’m on original hardware, running the 7700s dGPU and on bios 4.03 my CPU is hard limited to 35w when the dGPU is in “off” state. There have been major performance degrading bugs in every bios we have received on this laptop since it was released. I don’t think Framework has this very high up on their priority list. They already got our money and most of us are out of warranty by now.
You can look at the other GitHub issue for this bios. Even with repeatable steps, they still post responses doubting our issues and saying it’s not designed to work that way, etc.
I would recommend you submit a support ticket so they know about it and can work with you to resolve. Then return it if they can’t resolve your issue within the 30 days.
PROCHOT is an overused term.
It can be used by anything that wishes to ask the cpu to slow down.
Causes of prochot:
reached a definded power limit. So asking the cpu to not draw any more power, above the limit.
temperature too high, thus asking the cpu to slow down, and thus produce less heat.
overcurrent protection. Similar to (1).
safety: the fans have failed so slow or power off the cpu.
So, you are probably seeing prochot due to (1).
Using tools such as amdgpu_top can go some way to differentiate the different prochot causes.
Another possible problem is many of the prochot paths are not error corrected, so wrong messages can get through or message can be lost.
You know, I just saw a video yesterday taking about overheating ddr5 and how manufacturers are not accounting for it. Stability issues, lockups, etc that all went away when the ram was heat managed by adding heatsinks and fans (also 3d printed ducting). Definitely something to investigate.
The FW16 fans are triggered by cpu temp. I think they should also be triggeted by ram temp and battery temp.
I have modified my EC firmware so i have an EC command to read battery temp.
I have not looked before, but i might now look at RAM temp vs. Video playback.
For RAM chip cooling, maybe remove any plastic labels from the ram chip, and then add some non-electrical conducting thermal pads so it contacts the midplate metal.
Problem, it might make the keyboard too hot.
I have found that maxing the fans, not only cools the cpu, but also cools the battery, so maybe it would cool the ram a little also. One can force the fans on with ectool.
True, but cooling the cpu a bit might slightly cool other components just because the laptop internal ambient temp then falls. I tested benefit for battery temps this way. Reason i did the battery temps, is because keeping battery temps below 35C while charging/discharging extends battery life. The EC uses the battery temps. If it gets over 45C, the EC stops charging the battery.
I’ve been wonding if there would be any benefits to adding insulation between the interior and the heatsink so that the rest of the laptop isn’t being heated by the cpu excessively.
Unfortunately, it appears that the fan cooling is insufficient. Even with fans at 85% i have issues.
Fans doesn’t blow on the ram sticks so we can actively cool the CPU & GPU but not the rams.
The RAM eventually reaches 80 degrees Celsius (slightly cooler when thermal paste is applied), triggering CPU throttling via the PROCHOT_EXT signal.
Prior to applying thermal paste, OCCT would cause the CPU to overheat within five minutes. With thermal paste, I occasionally manage to sustain operation for 10 to 20 minutes.
Regrettably, this seems to be a design flaw where the RAM modules cannot be adequately cooled so the entire system goes down.
Can you try using ectool to max the fans and see if that helps the ram temps at all. It helps with battery temps so might help ram.
I have different ram chips, so cannot test your particular problem. I have kingston fury, you have crucial.
Another test could be. Remove midplate, use external mouse and keyboard and direct a big fan at the ram chips. It might give you an idea as to how much cooling is actually needed.
I guess another thing to consider.
CPU Tctl is 100C.
RAM chip limit is 80 or 85C.
In the photo i did not manage to catch the flag PROCHOT_EXT flag to 1 (my bad, i did the screenshot in the wrong moment) because the flag oscillates between 0 and 1 but we can see that the CPU is limited at 20W with a frequency of 1253 MHz!!
Another thing was that i set a TDP Limit to 40W to the CPU so not even running at “full speed”.
