I bought a different “Hive” on Mercari that included the cable, and finally updated my BIOS yesterday
The fan noise got worse because I resetted everything to default settings, and while researching that, I found a blogpost entitled “ASUS, your BIOS fan controls are atrocious”. And a comment on it informed me that I can actually just increase the obscure “Fan Step Up” BIOS setting to solve the issue where the fans briefly spin up to 75% just because I opened a new browser tab or something. Why did I put up with that for 3 years…
Anyone have an app they can vouch for to monitor hardware temps on Windows? My 6800 XT and 7700X run really hard when I am playing games, and I just experienced a situation where my two PCs connected to my desk powered down (didn’t really feel like a crash) unexpectedly. And I’m trying to deduce why this happened.
Afterburner if you want the more surface level stuff (though you can track/expose more, though not to the extent of HWInfo)
RTSS to throw it up on screen while doing things (games, though if you’re not careful it’ll show itself on any program rendering out a 3D surface, like some game launchers). RTSS is in the Afterburner install package, though you can hook it to HWInfo for real time monitoring
the AMD Adrenaline software should also be able to monitor and/or log GPU and CPU since you have an oops all AMD build, but it’s going to be a bit more limited on the CPU side
Furmark can put a load on the GPU and CPU-Z’s bench has a looping mode if you want to load the CPU
RTSS by default exposes GPU and CPU temps and load/utilization, though you can set more up and also customize the layout for it (by default it’s a kind of garish vertical block extending down from the top-left, a little time in the RTSS configurator and you can make it look however you want)
oh, also the new Steam counters are good, but they don’t go as hard as what the above can do (IIRC Steam’s readout can’t show power statistics, but does show real FPS if you have frame gen on)
I mucked about with mine to have it mimic the Mangohud config Valve setup for the Deck:
a 7700x will basically park itself at 95C and sit there (in practice it throttles a lot but in functionally undetectable ways like moving core 8 back and forth from 3ghz to 5ghz) unless you have a massive heatsink or are using eco mode, which I recommend (basically it should be a 65w chip and the defaults are not for 65w for no good reason, there’s a few good ways to dial that in). it should also probably have a -10mV all-core undervolt for a start… I have one as well and I like it but the stock settings are needlessly power hungry
if your PC caused the shutdown of other PCs, it’s not temps, it’s the power at the wall/UPS being overloaded (which is bad and tends to kill PSUs relatively quickly if it happens more than once). most components throttle pretty well these days and most people wrongly think temps in the 80s are bad — they’re fine with almost all modern components — but you should never exceed the mid-80C range unless a particular part (like the 7700x) is notorious for running that hot already
in general you shouldn’t ever have to keep an eye on temps using like a gaming overlay once you have a good idea of whether they’re safe, you can just run HWinfo in the background and tab over to it while you’re ironing that out. this is maybe the OCD talking but I have a massive massive bias for stuff like this where I want very badly to gather enough data up front that I can set it and forget it, and once I have switched past the data gathering stage (which is intense and fun), I do not ever want to look at it because it’s massively distracting if I do
The two PCs powering down at once is the really notable/unusual thing in this story. If the two PCs were plugged into the same power strip, throw it out and get a new one. The surge protector on the power strip starting to fail might explain it.
The power strip is definitely the thing to be suspicious about here. It’s an allegedly very sturdy one. But I have a lot of equipment plugged into it for work.
Losing my current job (as it looks like I will be pretty soon) just gained another positive. I can’t wait to get rid of all this hardware I only use for work…
one outlet is usually only something like 1800W (they’re usually rated for like 120 V * 15A in north america) and if you have two really hungry PCs you can definitely have problems.
This is absolutely the problem then. The PC I have for work is some suped up alienware with a 5080 and a suitably powerful CPU inside of it. I am just surprised because it wasn’t running anything significant at the time. Oh well.
yeah I’m hyper aware of stuff like this because even though I’m running a 4090 and a 7700x in like 450w total, I am plugged into a UPS with a nowhere near an 1800W limit and getting constrained unexpectedly there will kill a PSU
so help me god if I see or feel the frametime dip at any point I must quantify it with numbers and tear apart the task manager to find out what is committing crimes against my frames
also if you are ever trying to calculate the power consumption of other components in your machine a good rule of thumb is that if it doesn’t have its own fan it probably can’t dissipate more than 5w. so unless you have a couple actively cooled NVMe drives and a high end motherboard chipset like me, you’re almost certainly not over 25w for everything else combined, and even if you do, you are maybe glancing at 50
I have battery backup and surge protection at the whole-house (circuit breaker panel) level, partly so I don’t have to worry about that sort of thing. The battery backup was expensive and complicated to install, but the surge protection was surprisingly cheap and easy. It made me wonder why that’s usually done at the power strip level.
Digital Foundry keeps suggesting going on eBay, buying a used full-size Dell Optiplex, and slapping a cheap card (Intel Arc or whatever is still affordable) as a solution.
The main thing they warn is that you gotta make sure the case is actually gonna fit a GPU, and that the PSU will have enough juice to power it.
I guess it won’t be long before old business PCs get snapped up and harvested for parts, but it’s not a bad idea I guess.
i feel happy with myself for hoarding a few old computers/parts now. god willing nothing shits out on me, but if so, i’ve got an old motherboard with an 8th gen i3 on it, plus an old thinkpad t460, if i really get desperate and one just can’t buy computers anymore
I’ve almost got enough spare parts to build a second computer, roll my own Steam Machine, but the things I need (new processor, new SSD, new mobo) are way more expensive than they were a year ago, just like everything else
just do what I did and buy a kit of memory whose price explodes to be 5 times what you paid and never bother to return it and justify some work related nonsense to build out a new thing
Though I have full backup hardware of my kits. But I guess the core issue doesn’t depends on hardware but rather in re-think the existence of the Internet. After a few years later, if we wanna peace in our mind, we have to change our lifestyles. Our daily digital life still relies on continuous updates of security and tech bubbles will full of our life.
However, we shouldn’t continue to treat computers as an ultimate single workstation with office, online shopping, steam or etc. PC with windows 11 or cloud PC will become a printer or fridge in house only for regular work stuffs, and your gaming PCs will isolated and disconnect from the network. This game PC only turn on when we trying to play a game, and then we find that its lifespan has increased by more than double…
The earlier you prepare for a offline gaming PC, the sooner you may be able to rid yourself of this sense of powerlessness.
