Hard crash at random times

[ThetaCoder]

Quick Update.

While it still happens at random and without warning, I did experience 3 crashes in under 30 minutes yesterday(3/30/2025) while in Revanents Toll. 2 were within 3 minutes of each other as I was looking at what items could be turned in for Purple Scrips, and the 3rd was while working with one of my retainers.

[DarkArvan]

I do realize that Intel current gen CPUs had an issue with their microcode, and while it mainly affected i9 chips, i7s' was also affected. The BIOS installed in august had the microcode update to correct this. I did find out there was one further release of the F7 BIOS that I did not have, but it was included in the F8 BIOS release that was installed in my diagnostic steps.

Does not matter, if your CPU is already damaged, it can NOT be repaired by the BIOS update. Since your old BIOS was dated 2024, it might alreay be to late.
Check a CPU swap for testing.

[ThetaCoder]

The CPU has effectively been ruled out from no other issues in other programs, and from testing I have been doping with Intel.

And if the CPU had some damage, the update would prevent any more from the same source from happening. As I have said, if even 1 other program was experiencing issues or crashes, or even causing event viewer entries to occur I would engage Intel for a CPU swap. But the issue is only with FF14, and even people with AMD setups are having this issue (or the resulting error boxes and symptoms) so it appears more likely to be with the game than the hardware.

[ThetaCoder]

Ok, more updates, and still no reply or response from the mods or anything.

The game continues to crash every day, with this morning being the worst of the lot. In the span of my normal Saturday morning routine, which I had to cut short, over the course of 90-100 minutes, the game crashed 11 times. ELEVEN!

This was mainly a log in, run MSQ, cycle retainers, do custom deliveries, do allied quests, and my weekly gold saucer roundup. 11 crashes, no error screens, and of course, no errors in event viewer.

[GoodEater]

Apologies if you already mentioned, have you tried a complete OS re-image or tried running a memory test? *Update: I would try rolling back the Nvidia drivers to version 566.36. Been seeing many complaints about the 570x nvidia drivers causing crashes and blue screens across the RTX 4 and 5 series.

[Demon_Giri]

Your situation actually sounds a lot like mine, and I notice you have some similar hardware that I did in my recent build. While you may not suspect any issue, please try a memtest86.

I also just did a thorough writeup of various research and my own personal troubleshooting here: https://www.reddit.com/r/ffxiv/comme...t=share_button

Hope that helps.

[Demon_Giri]

Also in the same vein and maybe worth asking: what is the exact model of your RAM? Even if memtest comes back clean, I do wonder if something with your memory isn't playing nice.

Do you know if your specific ram pack is on your mobo's QVL?

Knowing what frequency the BIOS is actually running the RAM at is also relevant in this thought process.

I know it seems counter-intuitive to suspect any hardware, as it's only FFXIV that seems to have an issue, but I think there's room for both to be the problem. I had a similar experience and only found the problem with some heavy digging, and may have gotten lucky that I even found it in the first place.

If you're interested, I can probably aid you in setting some very minor undervolting to your CPU, the recommended power limits by Intel, and checking the RAM support.

The last two 14700Ks I've built with, I applied a -.05v offset and applied power limits 253w to short term turbo and 125w to long term turbo and have found this to not only be very stable but it also slightly increases performance while drawing less power.

I don't mean to give you any suggestions here that will annoy you or that you may think are completely irrelevant and missing the mark, I just have personally found with my recent experiences that this seems to come from a place of extremely minor hardware instability that is almost otherwise entirely undetectable and hard to find. SHOULD it be a problem? Hard to say. Would be nice if the game was a bit more reliable and didn't crash at the sheer thought of something sneezing, but there's not a lot we as the players can do about that. We can only do what we can.

Cheers and good luck

[ThetaCoder]

I'll go through the post and see.

I did run a Memtest 86 and it was ok, but I'll do it again. I did run a OCCT RAM and full system stress for 1 hour each, and no errors.

I am running Corsair DDR5 6000 RAM (PC5 48000), with a CL36 CAS Latency and rated at 1.4V. All 4 sticks are of the same type ( all bought at same time, and verified matched by their stickers), and yes, they are on the QVL for the mobo. Exact part num is CMH32GX5M2E6000C36

I am open to suggestions, it's just that since the game was running fine since the PC came online last June, I am less likely to believe it's a hardware issue, but yes it is possible.

Presently the only thing I have changed in the BIOS is altered the fan curves to ramp up sooner, and have not touched any of the voltage or frequency settings. I am a tad old school, and am from the days when over/underclocking was only used to shorten the life of your hardware (yes I know things are a bit different now).

[Demon_Giri]

All four sticks? The model you posted is a 2 x 32GB pack. Did you mean both sticks or are you actually running four?

If you are actually running four, while the stickers/models may match, being from separate packs can still cause problems, and also a lot of newer motherboards aren't as stable running four at modern frequencies.

I realize you may have just misspoken, so ignore my comment if that's the case. Just trying to cut down on potential back and forth.

[ThetaCoder]

When I bought the ram I bought 2 kits, so 4 sticks in total. Bought at the same time, to avoid the problems of possibly different chip batches, different chip providers, etc.