I'm really struggling to figure this out, could you assist me?
I'm really struggling to figure this out, could you assist me?
I’ve faced several individual problems that have intensified over the past week, making it increasingly difficult to find progress. Each time I believe I’m moving forward, things reset and start again. I’m seeking assistance in diagnosing the situation, since my existing hardware doesn’t work well with the new setup for testing or replacement.
Symptoms:
1. My GPU driver sometimes fails to launch at startup, leading to display issues and only one monitor functioning among three.
2. When it boots correctly, it experiences random crashes—ranging from minor glitches to complete black screens or freezing—regardless of what’s running.
3. The Windows installation has become corrupted multiple times in the last two weeks, repeatedly failing the boot diagnostics and requiring a reset. It’s been possible to fix it via the built-in menu at times, but errors persist, sometimes needing a USB boot device (though this seems to have stabilized temporarily).
4. Occasionally, I see only black screens after posting, though the Windows login sound is audible.
What I’ve tried:
1. Swapped my RAM for an older model (same brand if relevant).
2. Increased the RAM voltage slightly to check for stability (the corruption initially pointed to RAM).
3. Disconnected and reinserted all components except the CPU, as I don’t have thermal paste.
4. Installed fresh Windows updates (also changed the installation drive to rule out an SSD issue).
5. Ran memory tests multiple times, but they either fail completely or crash before finishing (mentioning memtest86!).
System Specs:
I5-11400f
Nvidia 3060
32 GB DDR4 RAM
Gigabyte B560 DS3H AC
Corsair RM850E PSU
At this stage, I’m limited in testing options and haven’t narrowed it down enough to justify a purchase. I don’t have integrated graphics for GPU checks, but I suspect the problem isn’t with the GPU itself. It seems the RAM might be central, though I’m unsure if it’s the controller, the motherboard, or the RAM module. Changing the RAM doesn’t seem to help much, and it hasn’t been stable even with XMP enabled—especially after a few years of no issues.
I’ve had some memory instability with XMP profiles and have experienced crashes during long tests (like memtest86). I’m seeking advice or tests that could clarify what’s going on. My situation is frustrating, especially with a child on the way, so I’m cautious but open to suggestions from Tom’s Hardware. Hopefully, someone can help me understand this better before investing further.
It might be related to the graphics card itself. Consider slightly lowering its settings with Msi Afterburner; there are guides available on YouTube. For the video driver issues, it's possible the card wasn't recognized correctly in that setup. Try resetting the contacts, though it's uncertain if that would resolve the problem if the card was always installed properly. Concerning your RAM, what are the XMP values and slot types? With two sticks, they should fit in the gray slots. If your RAM consists of four DIMMs, it could complicate things further—trying to achieve XMP speeds will depend on both the speed and the timing. 3200MT/s is recommended for 11th generation processors.
3200 is the recommended setting for the RAM, currently operating at 2400 at the default level. Two sticks are installed in the matching gray slots. I plan to underclock it this evening to check for any significant changes. The main reason I've been hesitant about the GPU is the recurring Windows boot corruption, which seems linked to a RAM problem according to recent reports. It's also the most costly component to replace, though I haven't checked older card prices lately. Perhaps I'll attempt to use the old 970 module if it fits the slot to help pinpoint the issue. I'm curious about switching to another PCIe slot since they have different speed ratings and I haven't tried that yet.
Another option might work and it's worth testing your old GPU. Anything that leads to system instability could risk data corruption, not just memory but also excessive RAM usage might cause problems. I'd try both single-stick configurations to identify which one might be problematic and possibly affecting XMP settings. Graphical issues are typically related to the graphics card.
Maybe purchasing a RAM package would allow you to get some Gskill.
Are both RAM modules, whether new or used, failing the memtest without XMP support? An overly tight CPU cooler might lead to memory issues. Adjust the tension screws halfway back from their maximum setting.
It could check each RAM stick separately with memtest to identify any issues. Also, a SSD diagnostic might help confirm everything is functioning properly. Following that, consider performing an Intel burn test on your CPU, then run Furmark on the GPU for roughly 20 minutes. These steps may help pinpoint specific components.
I didn't test it on the old setup because the issues kept happening even after swapping components. The new set didn't really fail—it passed a full test once, but it crashed again when I was away during the second run. I plan to slightly loosen the cooler, as it was originally built with an overzealous builder. With the recent cold weather (I moved from California to Montana this summer and have been worried about the temperature effects), maybe the wood floor is too cold. Thanks for your help and suggestions! It's running decently so far, but I'll swap the GPU if something goes wrong. I'll keep monitoring temperatures and see if it gives more insight.
I've run both the M2 SSD and my general SSD using Windows diagnostic tools, and both came back clean. However, trusting the results can be tough given the inconsistent nature of the problems. Sometimes it works fine for a night, but often it needs 4-5 restarts to boot, or last night it just froze completely and required a Windows reset.
It ended up crashing with some artifacts, so I swapped cards and there have been no issues so far. I ran it through 4-5 restarts and didn’t encounter any booting problems. I’ll let you know if anything changes! It wasn’t what I expected, but at least I might have something that works now!
Before discarding the 3060, do you have another machine to try it on? If the replacement graphics card is weaker than the 170W 3060, you might want to switch the RM850e temporarily if the power supply is nearing its limit. The 3060 could still function, but the PSU may be reaching its end.
My system was experiencing crashes in Topaz Video AI. I resolved it by adjusting MSI Afterburner to cap the Gigabyte 3060 at 95% usage. A few months later, after fixing the software, I removed that restriction. I’m uncertain if the crashes were due to the GPU itself, but Afterburner helped stabilize things. I was using the more stable "Studio" driver for the 3060, not the newest "Gaming" version.
How cold is the room at night when you power on the PC? Is it near freezing or even colder? Are there any signs of moisture on the metal parts inside the case?
This site mentions:
https://ramtechno.com/how-your-power-sup...-its-cool/
The ATX guidelines suggest a room temperature between +10°C and +50°C (under certain tests). The EN 60601-1 standard recommends +10°C to +40°C.
Although the spec says the PSU should work best at +10°C, I wouldn’t be overly concerned if the room is slightly cooler, like +5°C or even 0°C. However, I’d be careful not to start any commercial gear at -20°C.
I’ve tested Aerospace systems down to -51°C/-60°F. In extremely cold conditions, frost can form on circuit boards as moisture freezes. Once the temperature rises above freezing, the frost melts and a thin layer of water may coat the boards. It’s usually best to wait a few minutes for the frost to melt and the moisture to evaporate before turning the system on, to prevent short circuits.