Computer Issues

Lanx said:

like i said... low power

but...

please take a pic of this second powerchord first and post it at least

Click to expand...

Lanx said:

man how old is that video card? i don't remember seeing supplemental power in a long while.

Click to expand...

Apologies for the confusion, but the plug is on the Motherboard for that connector. Not the actual video card. My memories error.

But I think I found the issue within precision when I was just messing around with it. For some reason the 650 Watt Power Supply is reading as 550. Im not sure if its failing or what, but at the time this PC was being built - I had scraps to choose from for power supplies. Im guessing either I was just fooled into false labeling, the part is failing cause its an obscure brand, or maybe I missed a switch in the Bios or something? If its not the third, then I guess Ill just buy a new PSU for her since theyre basically all back in stock now.

Zindan said:

So would this image, from New Worlds fps counter, be showing the FPS + Frame timing?

View attachment 364118

Click to expand...

i havent played the game, so this is just a guess... but I'd think that 28.8ms is your ping to the server, no? does the number vary significantly?

edit: just re-read your follow up post, maybe it is frame time

Daidraco said:

Interesting that there is already another issue about New World before I even asked anyone to delve into mine. My g/f's computer is experiencing a "nvlddmkn" error 13 and 14, randomly during New World. The screen will go completely black, or just outright freeze and she has to end process it to close it out. There are a few band-aids she can change to extend the time it takes to get the error to reproduce, but if she cranks up the resolution, uses full screen and puts everything on highest - it crashes within 5-10 minutes. She doesnt experience this error in any other game, so Im unsure what the issue is here. If anyone here has experience with this issue, feel free to ask me for what information you need for me to help you diagnose the culprit.

Click to expand...

I've had a somewhat similar issue for years on my 980 GTX. while gaming, my screen would go black, but i would continue to hear in-game sounds. then my monitor would throw up a "HDMI No Signal" error, and it was not recoverable without hard rebooting the whole system.

after reading forums and feedback from hundreds of similar reports over the years, my guess was something wrong with the power delivery to the card, or the power processing on the card itself. I was running a 1000w 80+platinum PSU, and I changed rails and made sure I was not pigtailing a single line into the card's two power inputs.

Still, the only fix that worked for me was to use MSI Afterburner to reduce the Core Clock and Memory Clock, stepping each down by -25mhz, until I was able to game stably.


not entirely sure if this would fix your GF's issue, but maybe worth a shot until the new PC comes in?

Nemesis said:

I've had a somewhat similar issue for years on my 980 GTX. while gaming, my screen would go black, but i would continue to hear in-game sounds. then my monitor would throw up a "HDMI No Signal" error, and it was not recoverable without hard rebooting the whole system.

after reading forums and feedback from hundreds of similar reports over the years, my guess was something wrong with the power delivery to the card, or the power processing on the card itself. I was running a 1000w 80+platinum PSU, and I changed rails and made sure I was not pigtailing a single line into the card's two power inputs.

Still, the only fix that worked for me was to use MSI Afterburner to reduce the Core Clock and Memory Clock, stepping each down by -25mhz, until I was able to game stably.
View attachment 364520

not entirely sure if this would fix your GF's issue, but maybe worth a shot until the new PC comes in?

Click to expand...

Ya, I neutered the card a bit and the game plays fine now. The games she plays, you cant tell the difference. Her monitor is only a 120 and Ive never allowed the FPS to go higher. Still need to buy a new PSU, but at least she has a bandaid for now.

Well, finally had to undervolt my 5900x. It had worked fine out of the box, with package temps right around 40c while idle and 50ish during most games, spiking to around 60-62 on CPU intense games.

Motherboard pushes through an update and after restart my temps spike to 70c. Seemed a little weird, but it typically spiked right after boot to around 65c anyways, idling back down to 40ish after 30 seconds or so. Well, after about 30 minutes, the package temps still aren't moving and hovering around 70-75c while idle. Fans are all working fine(now boosted because of my temp curves), AIO is still pumping, etc. Can't figure out wtf is going on.

So, I uninstall Armory Crate after reading up on what a piece of shit it is. Still doesn't fix it. So, I download AMD Ryzen Master to look some more and notice that my clock speeds aren't really changing. Pretty much consistently sitting at 4200-4400mHz, even while idle. Prior, most cores typically sat around 3600-3700mHz, with small spikes to 38-3900 during web browsing and such. Went to the BIOS and undervolted. Now temps are sitting at 1-2c within where they were before and performance seems fine/the same so far. Fingers crossed..

Kirun said:

the package temps still aren't moving and hovering around 70-75c while idle.

Click to expand...

That's a red flag in my mind. The CPU has to be doing some real work to get that hot, (unless your cooler stopped cooling) so regardless of what monitoring software is telling you about clock speeds and voltages, what power is actually being consumed? My old 2700X for e.g. idles at < 30 watts and peaks at ~ 125 when fully pegged.
You may have solved the problem, by "fixing" the wrong thing. Nobody should have to fool around with under-volting on a stock system, unless there's something really funky about your setup that you forgot to mention.

Argarth said:

That's a red flag in my mind. The CPU has to be doing some real work to get that hot, (unless your cooler stopped cooling) so regardless of what monitoring software is telling you about clock speeds and voltages, what power is actually being consumed? My old 2700X for e.g. idles at < 30 watts and peaks at ~ 125 when fully pegged.
You may have solved the problem, by "fixing" the wrong thing. Nobody should have to fool around with under-volting on a stock system, unless there's something really funky about your setup that you forgot to mention.

Click to expand...

It was sitting at around 80-90w being consumed while idle/at desktop. Now it has been sitting around 40-50w.

If the cooler stopped cooling I'd have approached Tj Max pretty quick. It was idling at 70-75c and then bouncing between the same while playing a game. I've had a CPU lose cooling before and it didn't take long before temp warnings started to happen and it reset.

Kirun said:

It was sitting at around 80-90w being consumed while idle/at desktop. Now it has been sitting around 40-50w.

Click to expand...

Yeah, but 80~90W at idle is just stupid, assuming it really IS idling (i.e. all cores floating ~ 0% and total CPU usage ~ 1% max like mine does)
There has to be something else afoot. No reason you should need to be playing with voltage at all, especially the way Ryzen CPUs fully manage themselves by default.

Not trying to diagnose your problem, but if it was me, I'd be taking note of all my important BIOS settings and doing a reset. Sounds like that "forced" MB upgrade screwed something up.

Argarth said:

Yeah, but 80~90W at idle is just stupid, assuming it really IS idling (i.e. all cores floating ~ 0% and total CPU usage ~ 1% max like mine does)
There has to be something else afoot. No reason you should need to be playing with voltage at all, especially the way Ryzen CPUs fully manage themselves by default.

Not trying to diagnose your problem, but if it was me, I'd be taking note of all my important BIOS settings and doing a reset. Sounds like that "forced" MB upgrade screwed something up.

Click to expand...

It was definitely "idling". CPU usage was right around 2-3%, most of which is my live wallpaper.

I went through and checked all BIOS settings, believe me. Odd thing is that while in BIOS, it had my temps around 51-55c(so, I started thinking that maybe HWMonitor was just misreporting them). So, totally reinstalled HWMonitor and it didn't really change until I undervolted. BIOS now matches HWMonitor as well, meaning that on idle now I'm around 40c(lowest recorded in HW monitor is 39c), and power consumption is around 49-55w after leaving it on desktop for 30 minutes. Lowest wattage was 39w.

I appreciate the troubleshooting walkthroughs though. I agree that it's odd as hell and I do think it was something that ROG Armory Crate pushed that jacked shit up.

Lanx said:

do you still have temp problems? were you controlling it with qfan bios or aisuite? was this before or after update? you have a aio? you sure you plugged the pump power to the cpuheader?

Click to expand...

Temp problems are totally gone now. CPU is performing how it was - 40-43c while idle and highest temp recorded was 71c after 5 hours of Pathfinder: Wrath of the Righteous at 1440p, max settings. Processor speeds are 3.6kmhz at idle and PBO to around 4.9kmhz.

Temp problems started after a MB update pushed from ROG Armory Crate. Shortly after restart is when I noticed the temp issues. It was almost like it was "stuck" in PBO, where idle wattage was around 80-90, processor speeds around 4.0-4.4k, etc. While idle it was 70-75c and while gaming it hovered around the same, with max temp recorded at 78c according to HWMonitor.

I was controlling fans with AISuite 3, but I uninstalled it along with ROG Armory Crate, thinking they were part of the issues. I use FanControl now, but also have a QFan profile setup in BIOS as a last resort. PWM fans/fan profile is performing as expected, monitoring it with HWMonitor - fans start to ramp up as temps increase. 71c was the highest recorded temp while gaming, after undervolting.

AIO is definitely functional. Like I said, I've lost CPU cooling before and you can't just sit at desktop without it eventually encroaching on Tj Max and restarting. It's possible that permeation is a factor, but the AIO is only 6 months old at this point - I mounted it properly, etc. HWMonitor shows it working, BIOS shows it working, fans on radiator work, etc. And even if permeation is occurring, there's no way I'd see temp jumps of 30-40c. It'd be a gradual increase of idle/gaming temps.

Lanx said:

are you controlling pump speed or just have it 100%?

ppl say aio pumps should be 100% and not controlled since its only for a decibal decrease and it barely registers

Click to expand...

Pump is on 100% and has been since I built the system.

If it's not happening in BIOS, that means it's software.

Kiki said:

If it's not happening in BIOS, that means it's software.

Click to expand...

Agreed, but temps were still elevated in BIOS, just not to the level that software monitors were displaying. Temps in BIOS were 50-55c, which is still a solid 10-15c higher than normal idling temps used to be.

MusicForFish said:

(you just left this here, I'm sorry)

Click to expand...

Yeah, that was kind of a softball.