Framework 13 Restart/Crash When Gaming. Should I Return This Laptop?

Item: Framework 13

OS: Windows 11

Cpu: 7840u

Ram: Kingston FURY Impact 32GB (2x16GB) 5600MT/s DDR5 CL40

Storage: SK Hynix Gold P31 2TB PCIe NVMe Gen3 M.2 2280 Internal SSD

Charger: original framework charger or Dell P2422HE USB c

I have Framework 13 with 7840u that will constantly/consistently restart when I put Windows on “best performance” mode during gaming. It will just turn black and restart on its own, without BSOD or minidump.

The game could be in the pause menu too, it doesn’t have to be in game to reproduce the problem. The games I play are usually gta 5 and overwatch, not too demanding games. Everything would work fine if I put it in “balanced” mode. However, balanced mode on benchmarks will achieve lower than average score usually topped on 28w and best performance will achieve an average or higher score on 30w(ex: 11k vs 12-13k in cinebench r23), sometimes spiking to 35w with 70-86c temp. Games don’t usually crashed during the initial 35w spike, its always during the stable 30w. Inside the event viewer, it has event ID 125(see below) showed up 4 times then Even ID 41.

Things I have tested:

  1. Installed AMD auto detect tool to install chipset drive and main driver, then revert back to original framework bundle drivers. With DDU involved in removing the main driver.
  2. Tested with free version of memtest86(4 tests only) on each sticks then both sticks together. Tested with memtest86+ with all 10 tests with 8 passes on both sticks. All passed with 0 errors.

With 1 stick of ram on each slot stops from restarting, but performance in game drops significantly, performing worse than balanced mode.

  1. Without the charger plugged in, it doesn’t seem to crash, or it will take longer to crash but I don’t have that much time to test since it usually crashes under 30 minutes.

  2. Changed 80% battery limit back to 100% still crash

When changed from 100% to 80% while the battery is 100% then plug in the charger before booting. Even plugged in, the battery goes down while gaming, it stops at 80%, the game will not crash even after the battery is depleted to 80%.

  1. When launching overwatch(usually after a restart), CPU usage will skyrockets to around 90%, it takes around 5-15mins for it to be normal, during this time framerate drops significantly.

  2. Change the charging port around. Tried using HDMI, USB C, and daisy chain. Still restart.

  3. Gaming in best performance mode over 60fps feels super choppy unless limited to 60fps, but this won’t happen in balanced mode.

  4. Stress test with 3d mark using firesticks(direct 11), time spy, and speed way won’t crash in best performance mode. Scores are all expected/average.

  5. Laptop will crash/restart on its own, but in gta 5 if I pause it and alt-tab out, it is the fastest way to reproduce the problem. In overwatch, it can take longer to happen. Even limit to 60fps the laptop will still crash but it takes hours.

  6. Sometimes is not a simple blackout and restart. It could have an audio glitch “TIS” sound plus a visual GPU dying for a couple of seconds. But this is much rarer than just blackout + restart.

My questions/concerns:

  1. What might be the issue here and what other troubleshooting steps should I take?
  2. Does this seem like a hardware issue?
  3. My laptop was shipped on 10/09 so I don’t have much time to decide if I want to return it. I want to keep this laptop but if it somehow is a hardware issue I don’t know if is worth the RMA process.

Is this normal behavior? Does anyone have similar issue?

That is a lot to unpack. You have done some extensive testing and my amateur opinion is uncertain if it is hardware or firmware related.

Sounds like you do a decent amount of gameplay overall and generally using more of the full potential of what the laptop has to offer. As portability may not be a larger factor (usually can not game for long on laptops); other than the added cost, would a Framework Laptop 16 not be better suited for you?

Granted the platform is not as mature as the Framework Laptop 13; though it is built with some added performance that the 13 will never achieve.

If the portability is a primary factor; an alternative route would be to return the AMD Framework Laptop and if you still wanted to stay in the Framework family go with the Intel Core Ultra. Uses the same memory, same SSD, and Intel just has orders of magnitude more people dedicated to ensuring drivers and support work for their hardware.

I love what AMD brings to the table, though it comes with still ironing out things that Intel and Apple can squash in a weeks time vs. sometimes months.

In your specific case it could be a newer firmware issue that needs to be addressed, hardware fault of the mainboard, thermal issue, voltage regulator, so many things. If you had the patience to diagnose and get to the root of the problem, just know it might take much longer than you anticipate.

If it is going to make you happier overall to not have to figure out the details; return your device and either get a different Framework product or dive into one of the many performance/gaming laptops that are on the market.

You have given it a fair shake and done lots of testing; I would hate to see this be the thing that sours your satisfaction with Framework and others that would read about your issues trying to resolve it.

2 Likes

Seems to be happening under sustained load and not peak load, and well within normal operating temperatures.

Your event id 125 followed by a critical event 41 means there was a kernel power problem.

Look under /Windows/Minidump to see if you have a log there. Investigating it can tell you what component specifically had a problem. My guess is it is the GPU, but I have no idea really.

If it is your GPU, I would suggest uninstalling the graphics driver, and reinstalling the one bundled with the Framework driver bundle.

Basically, if there is a minidump log there it can tell us the specific problem, and we can go from there. Drivers are usually to blame for such error codes.

2 Likes

Unfortunately, there is no mini dumps, it’s not a bsod, Its kinda like a smooth restart. I tried with framework bundle, then latest amd driver with other chipset driver then back to fw bundle. I think it might be motherboard after research with similar problem, some people found out its CPU and motherboard not handling voltage or power well

Ok, I would absolutely get in touch with customer support.

Hi thank you for your thoughtful reply. My biggest concern is whether I should deal with support or return and get the same one. I think you bring up some good points. I usually travel in college so I typically require a lighter laptop around campus. I thought about fw16 or other brands. I know laptop like MacBook which almost every single one of my classmates uses one and asus g14 can probably fit in my use case better but I like the modular design. Kinda voting with my wallet here.

Because I have seen people game on this machine. I think if the thermal and performance are within range, there shouldn’t be any reason why 7840u inside fw13 cannot handle it. I am not looking to game on AAA games, just light gaming. The laptop crashed in the pause menu but not during intense load. Which is really strange to me.

Would you recommend going through the support system, or should I consider returning it and ordering the same one? Because it also has other issues like antenna is not working correctly. Webcams crash laptop (this is likely a software issue)

The support system will get you a RMA and new product if necessary, as well. But I think it is important to give them a chance to see if this is a software issue or hardware.

1 Like

You should strongly consider returning it. If you still have issues and you’re past 30 days since purchase you’re going to be stuck troubleshooting with many various steps. You won’t get the option to return it again.

1 Like

I see, that is what I am worried about, I heard some people take months to troubleshoot. I just hope they won’t deduct much on the shipping fee. Thank you for your input!

If you are willing to spend the time troubleshooting certain issues (this doesn’t mean all the time, it usually works fine all the time), then you can consider keeping it, but if you want guaranteed stability, I would still recommend going with a brand like Lenovo, Dell, or HP.

All companies have their bad batches, but Framework seems to still have a lot of ironing out to do when it comes to quality control, so many times bad mainboard and other components slip out and are shipped to consumers.

1 Like

My opinion (and it’s likely an unpopular opinion on this forum given the fan base) is to return it if you’re still within the return period…and it does somewhat depends on the general mindset of returning a faulty product in your region / country / culture.

Justifications:

  1. Consider if this was any other product (electronics or not), if it’s faulty out of the box, you would return it and exchange for a working unit.
  2. Opening support ticket would just lengthen your time with a non-working unit, increases Framework Support workload…dragged out for everyone.
  3. You can re-enter into the Framework QA lottery (To be precise: Every purchase of any brand is a QA lottery…some more so than others…even when buying a house).
  4. Depending on your region, your next (hopefully) working unit (be it a Framework laptop or otherwise), would be working on your desk ‘sooner’ than if you were going through troubleshooting / issue identification / remediation determination / ship out of replacement (of laptop or part)…etc.
  5. The return period is a return period, treated differently from the rest of the in-warranty period for a reason. Use it. There’s a switch-over of power dynamic between you and the manufacture once the return period is over. If you don’t leverage this while you can…then you’ve essentially subjected yourself to a “purchase without a return policy” agreement, which no one in their right mind will go with at the time of purchase.
  6. If it’s not faulty, and it’s giving you this much friction just to ‘use it’…then it’s not a good match. This is with the assumption that you know what you’re doing (and it sounds like you have done all the usual good / best practices in an attempt to resolve the issue), or you’re missing certain pieces of information / details. However, given the wealth of knowledge on the community, and no one has been able to provide you with additional resolution details. It’s likely to say your unit is acting abnormally.
  7. You shouldn’t have to suffer (lose your time and / or mind) over a newly purchased product regardless if it’s QA or product/component design (processor, RAM module, capacitor, charging brick…etc) or compatibility issue or otherwise. AMD and Framework…and every other OEM in-between, ought to provide you with a product that can easily be up-and running.
3 Likes

Just to be clear, my question right now is not about my willingness to troubleshoot or switch brands, it’s that I have the option to return and get a new one, exchange the exact same spec. or go through troubleshooting with support that may still eventually have to swap my main board.

After hearing many people telling me this is not a normal behavior I decided to go through support for a bit just in case there is something I missed. but if the problem cannot be solved before 30 days return period then I will consider if I want to “exchange” or not.

You are likely going to have to pick between returning it now or troubleshooting and getting a replacement. By the time you are getting a replacement mainboard, the 30 day return period will likely have passed, which is why I sent that post.

If you are willing to deal with the occasional problems then keep it, otherwise return it now before it’s too late.

1 Like

I got fed up with Dells and Lenovos blowing up or have insane drivers that puts the computer to sleep for Windows.

I need the 16. I much prefer numpad, and I need lots of USB for hardware dev. Also SSD. High-refresh-rate screen is just a bonus.

Maybe use HwInfo to inspect temperature for other components? Under such (30W) sustained, not only can the GPU and CPU get hot, but RAM and other things (power delivery, etc) as well. If any of them trip, this sort can happen.

i have a similar issue to this (also amd but on linux). contacted support. still doing roundtrips but they suspect it might be hardware-related