Home
Creators
Search Recent Random
Posts
Search DMs Popular Hash Lookup Tags Random
Importer
Import FAQ
Register Favorites Keys Review DMs Logout
Community
Telegram ThePornDude
Home Creators Posts Import Register Favorites Logout

Hardware Monday | The first and hopefully last one. (Patreon)

Published:
2023-08-14 09:58:43
Imported:
2026-06
Tags:

Content

Hey everyone, I'm back from my vacation and ready to start working again... is what I'd like to say. The reality is that my hardware is acting up in a major way once again. Let's talk about it.Some of you might remember that I mentioned a big hardware failure I had a few weeks ago, but it was "resolved". The reality is I never fully resolved it, I just got my PC to boot again, but with frequent crashes and only one of my two GPUs working. But because I didn't want to risk anything before getting the last update out, I decided to not deal with any of that before my vacation and instead tackle the problem once I'm back home. And as you can see in the picture the problem persists. I started working on it on Friday and have been going at it ever since Going through one possible solution after the other, trying to narrow down what the actual problem is, but I still don't have the slightest clue. Right now I'm in the process of reinstalling the NVMe drivers for my SSDs which, for some reason, decided to not get recognized as bootable devices anymore, don't ask me why, and launching in CSM will simply not post at all, I'm guessing because of the GPUs. In general the 4090 has no signal no matter what I do and I'm actually using an older mainboard that a technician declared broken, because my new board won't post either and gives me a constent loop of error codes regarding the CPU, which works perfectly fine on the older board. But on the older board I can't use multiple GPUs because it will, for some unknown reason, stop powering the top GPU whenever I boot. And because this isn't even the dumbest thing, no matter how much I mucked around wtih BIOS settings and drivers to get my dual GPUs running, the only fix is to play Monster Hunter World for about 30 minutes and the second GPU will power on magically and works again, until I reboot the system. I guess it's not very surprising that there is very little advice to be found on this very particular situation. And as you might expect that's not even all of it.So... I'll be honest. I don't know when this will be resolved. I'm trying out everything I know and I think I'll find a solution at some point, but for now this simply means I can't work on the game. I have many possible fixes yet to try, but there is a limit to how long I can work on my broken machine without going absolutely crazy. Most of the time I can confidentally say that I'm a pretty calm guy, but when it comes to fixing the hardware my entire work is based on, I have the shortest fuse imaginable. I get really angry. Reaaaaaaaly angry. I hope this won't lead to any big delays but this time I can't promise anything. I'll start working on the game again the day this problem is resolved. but I can't say if I'll find the solution today, tomorrow or at the end of the week. I'm confident that I can find it though.And that is all. I'll go back to talking about the game again when this whole episode is behind me, you just need to give me a bit of time. Until then I wish everyone a great week and please wish me luck : D

Files

Comments

RMZ

Without knowing the exact specs of the machine, some thoughts: - Make sure you have way more than enough power coming from the power supply, like 1200W+. 30 and 40 series can be spicy with power spikes - You might not have enough PCI-E lanes left to power 2 full graphics cards + various NVME and possibly SATA devices and USB controllers, so consider getting a CPU + board that can handle more lanes or at least look at the manual how those lanes are assigned and see if you can leave the ones coming directly from the CPU reserved for the GPUS and leave the rest to the chipset - Full DDU nuke on all the display drivers in safe mode works wonders sometimes

395games

PSU is definitely fine with 1600W, I'm using a Threadripper processor that supports up to 128 4.0 PCIe lanes so those aren't a problem and I tried DDU so, sooo many times it's embarrassing : D The crazy thing is that the system worked fine for a year, then a random black screen and since then the GPU has been acting up, I tried fixing it and now everything is crazy. But I'm making progress. I just got the system to boot for the first time since Friday, with only one GPU, not all of my storage devices installed and on the old board, but at least I saw windows again, that's a bit of hope for me.

RMZ

Mh yeah that sounds fine on paper, but it booting without the storage installed is sus. Threadripper can be bitchy too but if it worked before it should keep working if you didn't touch the BIOS firmware or anything else. GPU might also just be defunct. Can always happen under constant heavy loads

395games

I don't think the storage is the reason why it boots up now. I just removed it to have a blank slate while testing and not run the risk of damaging it. And I tested all my GPUs in another setup and confirmed that they are working just fine. As I said, many mysteries to unravel here : D

RMZ

Well, good luck then. Wouldn't be the first computer to mysteriously be a bitch and be thrown out the window haha

Michael Chiarcos

The only advice I can give is what they did to us in our final exam in Computer Electronics. Start unplugging devices that are 2ndary and just run with the most basic format possible. When that is stable, add one item at a time and wait for it to crash. When it does you've either found your cause or if you swap that item with another and still see a crash, you've discovered that the problem is more fundamental, like the board can only handle so much before it gives out. Keep track of the variables you are manipulating and cross your fingers that logic will solve your problem

TallestLegend

Wish I wasn't homeless I would have a good pc someday.

Nick

Good luck with your hardware issues! Dealing with technical errors are the worst especially when the obvious fixes don't work. The one thing that comes to mind is that a Windows update could have broken something. I've had various things break upon updating windows although not to the extent here. But I have had Bluetooth crap out on me, plus a separate time I had my pc randomly shutting off after getting an update. If your pc was working fine for an extended period and then stop working without changing any of the hardware, then it seems likely that could have something to do with either a component breaking or possibly a software update causing issues or conflicts.

Nick

Have you inspected the PCI expresses slots where the GPUs are inserted? Maybe dust got inside? You could try using compressed air from a compressed air canister meant for electronics in order to remove any dust that might be in the PCI slots as well as removing dust from inside of the rest of the case while you're there.