P9603Ed

Uefi causes fans to go into overdrive. Just idling causes the things thermals to spiral out of control. This cannot be normal. I got it from hidevo, and got the Fuji/LM thermal upgrade and this behavior cannot be in anyway shape or form normal. So the hell is going on with this laptop? Is it faulty?

*P960ED
Phone keyboard caused typos.

 

Ya must be the paste try undervolt with throttlestop first run occt show us the temps

 

I have no OS installed currently and after feeling under the laptop with it off it's the single fan side that is hot as ****. I have tg liquid metal and fujipoly but still it sounds like a jet engine.

 

Sounds like it's not making proper contact. I'd call them.

 

Sadly they're not in over the weekends so I've sent an email.

Edit: I should also say that the powerbrick is also similarly getting way too damned hot for not doing anything. Like it feels as hot as under the CPU on the chassis(which reads ~70*C+) and to me it shouldn't be getting _that_ hot unless I'm stress testing the system, the power draw shouldn't be high enough for that to happen.

 

When you had the os on there did you open monitoring applications for the CPU and gpu to see what was going on?

 

I had just my sysreccd that I was using to run badblocks on the nvme m.2 ssds and it was reaching ~85*C with total cpu utilization at 4% when I finally shut the thing off.

Edit:I've yet to install any OS onto anything.

I always run badblocks through a complete pass(4 patterns tested) and then do a smart long test, then do 4 more patterns across the drive, and another smart long test before I trust the drive. If it's removeable media it's just the 2 passes of badblocks. Basically I write/read to the entire drive 8 times using 4 patterns before saying it's safe to use.

 

That's kind of redant now with flash, plus if data is that important any drive can fail at any time and backups are way more important that the drive the data is on.

DOS environments and their ilk can keep the CPU fully spooled up which wont help temps.

Speak to your reseller if you cant get an OS on it.

 

It's not a DOS enviroment that I was utilizing, it was a gentoo GNU/Linux live cd. And people can say trust flash all that they want but I always do this test with all media that's RW. Memorycards, flashdrives, harddrives, and now SSDs. I want to catch infinitile death by artificially putting a load on it rather than waiting for it to die 3mo down the road.

I know the curve for HDDs show that quite a few drives die early in life then it smooths out and then towards the end of the lifespan more start dying off. I want to make sure that the medium works before I install anything to it. I may be a senile old man or something but I refuse to spend my time installing an OS and ricing it up to only have it fail 3mo down the road. I am still using the same SSHD for 3yrs right now so my system(at least to me) seems to work. I also do long/extended SMART tests every 6mo, and short tests every 3mo on all of my active drives. And backup ones are just every 6mo a long/extended SMART test is ran along with scrubbing the data to make sure that the hashes match up.

 

The mechanics of flash are very different meaning failure tends to occur straight away or some time down the road.

For instance when you say patterns across the drive you have no idea where what you write is put, the controller is the only chip that can determine the position of data, it will not match the OS.

A simple regular backup would be a lot less work and wearing on the drives.

An enterprise setting wont scrub drives and test them regularly as it causes premature death, there are just redundancies to handle any issues that do arise. You can basically do the same thing at home.