Install Steam
login
|
language
简体中文 (Simplified Chinese)
繁體中文 (Traditional Chinese)
日本語 (Japanese)
한국어 (Korean)
ไทย (Thai)
Български (Bulgarian)
Čeština (Czech)
Dansk (Danish)
Deutsch (German)
Español - España (Spanish - Spain)
Español - Latinoamérica (Spanish - Latin America)
Ελληνικά (Greek)
Français (French)
Italiano (Italian)
Bahasa Indonesia (Indonesian)
Magyar (Hungarian)
Nederlands (Dutch)
Norsk (Norwegian)
Polski (Polish)
Português (Portuguese - Portugal)
Português - Brasil (Portuguese - Brazil)
Română (Romanian)
Русский (Russian)
Suomi (Finnish)
Svenska (Swedish)
Türkçe (Turkish)
Tiếng Việt (Vietnamese)
Українська (Ukrainian)
Report a translation problem
Have you tried re-seating the RAM?
Any debug LEDs on the MB?
If you can boot into Windows finally, you can run some benchmarks and see how your gpu handles them. Then, it's possible to compare your score to others online.
https://www.passmark.com/
https://benchmark.unigine.com/superposition
https://benchmark.unigine.com/heaven
Remove any frame-limiting and/or monitoring software before running benchmarks.
Here's an article--there are many others like it too:
https://www.wikihow.com/Tell-if-My-Gpu-Is-Dying
But after I took out the GPU (the CPU had integrated graphics) it booted fine, so either
1) GPU was the problem because it worked without it, or
2) PSU was the problem, because it worked with the lower power required without the GPU.
Anyone's guess. I don't have the old PC any more - gave it ppl who re-purpose them for those less fortunate.
This issue specifically reminds me of the 16-pin ATX cable issue on my MB. Have you attempted to re-seat this thing? I had an issue where this was a culprit. I seen fixes where its a simple turning off monitor, discharging all static, replugging in the monitor.
If/ When you get it to run, do the SFC Scan and DISM CheckHealth commands in Admin PowerShell or Win Terminal.
Also may as well check on BIOS updates and the Chipset Driver.
Tried reseating the ram, though back into their original slots. Worth sliding them 1 slot along each maybe? I believe there was a soild amber light during the stuck boot up process. I'll look again next time I restart this and pay closer attention to what is happening.
@Plat - Updated OP with more info. Thanks for the other links and suggestions too. I'll have a look at that.
@Wynters - Yes, I have tried the GPU in 2 different PCIe slots so far. No luck. I have not checked the cables you mentioned on the Mobo.
Thanks for all the replies and suggestions etc. Appreciate it.
If you can make it to the desktop you can try debug mode in Nvidia control panel. It will lower your performance to Nvidia reference values and disable any factory made overclock that shipped with your 2080ti. If debug mode is stable then that means your GPU clocks are not. This is technically bad and you can choose to continue to use it at reference or replace it since the chips are actually wearing out. The problem is debug mode doesn't stick, so you'll have to enable it on every boot (if you can make it boot!).
If you do end up getting a GPU, make sure you look out for promotions. Right now you can get starfield premium edition for free with select GPUs and that's a $100 game. This will add value to your purchase.
Power it off, disconnect from socket, take out cmos battery and gpu.
Press power button a couple times like this while disconnected. Leave it like that for like a minute.
Put cmos battery back in, plug into socket. Turn back on without gpu.
Them turn it back off by power button or unplug from wall or whatever and disconnect from wall again.
put gpu back in, plug back into wall and power back on. See if anything changes.
This might do nothing but sometimes it can do the trick.
Worth a try, as well as other suggestions from above comments.
Before you start taking things apart, though, you should identify which area that debug light indicates. That'd narrow it down a bit and avoid unnecessary tinkering.
https://nvidia.custhelp.com/app/answers/detail/a_id/5396/~/geforce-rtx-40-series-%26-power-specifications
I really didn't know you could do this without the GPU. I'm still waiting for the issue to happen again and I think I'll take the GPU out and try it on a friend's as suggested. The PC shipped with that PSU from Scan/UK. You think it's maybe not enough?
Yes, I did make it to bios. Other times I even got to the login screen, entered my password, back into desktop for a moment then back to black screen and 100% GPU fans.
I did the 10 second method you suggested too. I end up in the same loop for 2-3 attempts before I get back to desktop and resume as normal as if things are all good again. At this point, the PC performs perfectly and no further issues or black screen/100% fans. I guess if it keeps up, I could try the battery.
Today, there have been no issues. I paid attention to the light sequence and it boots with a brief white light, green light, then straight to desktop. I'm 90% positive it was an amber light when stuck in the booting loop and 100% fans.
Thanks, I shall add this to my troubleshoot list. Not that I'm a tech guy at all, but how hard can be. Appreciated.
Thanks Plat. That has saved me another potential problem if I need to replace the GPU.
As for now, and after moving the RAM one slot over each, I've been able to boot up as normal and had no issue so far. I'm reluctant to mark this thread as solved just yet, but fingers crossed.
Thanks again for all the tips and advice folks.
The position of the debug light is relevant, not its color. Next time you're getting these issues, check at which one out of those 4(?) LEDs it gets stuck.
The current 650W PSU is enough, if not malfunctioning.
If you'll change the GPU, ask again. There are huge differences in power consumption between the high-end and mid-range GPUs.
Although... I imagine you'd want to go 'up' from a 2080Ti, so 850W+ might be necessary.
Most (if not all) of the last two generations of mid-range cards with lower power requirements would be a down-grade for you, IMO.
correction, get a 1000w for a 4090, had a friend buy a prebuilt with a 4090 and the company put in a generic non modular 850w psu, the pc would crash every time the person played a game, i came over and replaced the psu with a corsair 1000w and his pc stopped crashing.
what was causing it? it was the power spikes pushing way past what that low power generic psu could handle.
that being said, always go higher than what is stated, as whats stated is always the minimum.
--------------
edit: the thing is, when it comes to electricity, giving it only the minimum is always going to end with issues, unless you run everything (pc and games) at balanced or lower and lowest settings in your games, then and only then will you be left with head room for any electrical inconsistencies, spikes, ect..
anywho and it should be common sense by now, but the psu quality and its wattage, are the one thing you dont skimp on when it comes to buying or building a pc, who cares if its overboard, at least you know you arent riding the line power wise and that you have enough power for additions to the pc as well, as well as wont have to worry about your hardware causing your psu to fry and burn down everything.