All hardware acts up. How does yours?

As a system builder, I have rarely come across a fully built system that does not act up in some way, sooner or later. E.g. my Deskmini X300 occasionally (~1 in 20) boots up into the UEFI BIOS screen, from day 1. An older system with a TV card froze hard when watching TV after a random time. Some LED lit keyboard had a production fault where one key would not be lit after some use (a known problem caused by faulty solder at exactly one location on the PCB).

What are your stories? Any remedy found?
 
Keyboards: I've got a couple of Logitech K485 that have done the LED thing on a few keys. Annoying but easy to work around.
CPUs: I don't want to jinx anything but an intel NUC and a midtower on a GigaByte motherboard have been pretty solid for a long time. The desktop had a failure on internal RealTek ethernet device, but easy workaround with a PCIe Intel card.
 
I bought a pre-built system from Micro Center some years ago that mostly works well, though it did have intermittent problems like locking up every so often. I presumed it was a software issue.
A few weeks ago, I got tired of crashes so I decided to see if I could fix the drive activity light that never worked.
It turns out, they assembled the system wrong and had the reset button feeding voltage into ground.
Reset and the drive activity light work now and the onboard Ethernet is much faster. I wish I would have checked that much sooner.

As for remedies, probably don't buy consumer hardware. Buying second-hand professional HPs has worked well for me in the past. They're built with higher quality parts because its cheaper for them to use more expensive parts than to have to send technicians to service them in the field. In my area, a lot of government systems see very light use and are required to be replaced every three or four years, so there are a lot available and relatively recent.
 
As for remedies, probably don't buy consumer hardware. Buying second-hand professional HPs has worked well for me in the past.
This is an interesting observation (matches my experience). Consumer HW (especially tech stuff) is built on the edge. Power Supply? Barely adequate. I've had some stuff (Dell) where adding a disk drive overloaded the PSU. New PSU fixed it all, but kind of limits everything if you need a new PSU to add a single card.
 
This is an interesting observation (matches my experience). Consumer HW (especially tech stuff) is built on the edge. Power Supply? Barely adequate. I've had some stuff (Dell) where adding a disk drive overloaded the PSU. New PSU fixed it all, but kind of limits everything if you need a new PSU to add a single card.
PSU is the most important thing to think about first (I think), and you should take it with a power reserve. Mine is Corsair AX1600i.
 
fwiw, I have never had a hardware failure. Unless you want to include the mouse that went bad on me just a couple of months ago. Or the keyboard I spilled coffee on. The computer I bought in 1998 or so in my basement turned on and ran just fine last year when I tried it. Even the system I tried to break once wouldn't (that's a different story).
 
I've had a couple of spinning rust hard drives fail unrecoverably. Well, you might have been able to recover the data if you went to a specialist and took the platters out and read them on their special equipment. But not recoverable by normal means. I had one cheap chinese SSD that I replaced the tiny HDD in an X41 with, and that failed after about 6 months. But yes, by and large, most hardware I've had has been super-reliable. I have run all my home systems off UPS's for as long as I can remember, I think that helps.

I guess going back in time a bit, fans and big complicated heatsinks getting clogged up with dust used to be a problem. And there's always heatsink paste drying out. But these are very minor problems.

I don't think I've ever had a stick of dram that failed, not once.
 
I have a Carbon X1 under Linux where the wifi hangs once in a while. Unfortunately in Linux you can't reset the wifi hardware without reboot the way you can under FreeBSD netif restart.

Otherwise things are pretty stable.
 
My laptop sometimes freezes at POST; I downgraded the BIOS to before-MFG (along with XHCI + other firmware) and ME disabled with probably some on-chip version higher than the BIOS hookups so it could be anything, but everything's fine once it boots :p (no solutions yet but I toggle random BIOS settings hoping to stumble on something)
 
I have run all my home systems off UPS's for as long as I can remember, I think that helps.
I also always use a UPS
[lanin@freebsd15srv1 ~]$ upsc pw9130
battery.charge: 100
battery.charger.type: ABM
battery.runtime: 1541
battery.type: PbAc
device.mfr: EATON Powerware
device.model: 9130 1000VA-T
device.serial: PL342A5660
device.type: ups
driver.debug: 0
driver.flag.allow_killpower: 0
driver.name: usbhid-ups
driver.parameter.interrupt_pipe_no_events_tolerance: -1
driver.parameter.pollfreq: 30
driver.parameter.pollinterval: 2
driver.parameter.port: auto
driver.parameter.synchronous: auto
driver.state: updateinfo
driver.version: 2.8.4
driver.version.data: MGE HID 1.55
driver.version.internal: 0.67
driver.version.usb: libusb-1.0.0 (API: 0x01000102)
input.frequency: 50.0
input.transfer.high: 276
input.transfer.low: 140
input.voltage: 221.0
input.voltage.nominal: 220
outlet.1.delay.shutdown: 65535
outlet.1.delay.start: 0
outlet.1.desc: PowerShare Outlet 1
outlet.1.id: 2
outlet.1.status: on
outlet.1.switchable: yes
outlet.2.delay.shutdown: 65535
outlet.2.delay.start: 0
outlet.2.desc: PowerShare Outlet 2
outlet.2.id: 3
outlet.2.status: on
outlet.2.switchable: yes
output.current: 1.50
output.frequency: 50.0
output.frequency.nominal: 50
output.voltage: 220.0
output.voltage.nominal: 220
ups.beeper.status: enabled
ups.delay.shutdown: 20
ups.delay.start: 30
ups.firmware: 0130
ups.load: 33
ups.load.high: 102
ups.mfr: EATON Powerware
ups.model: 9130 1000VA-T
ups.power: 324
ups.power.nominal: 1000
ups.productid: ffff
ups.realpower: 296
ups.serial: PL342A5660
ups.status: OL CHRG
ups.temperature: 26.9
ups.test.result: Done and passed
ups.timer.shutdown: -1
ups.timer.start: -1
ups.vendorid: 0463
[lanin@freebsd15srv1 ~]$ 5~
 
As a system builder, I have rarely come across a fully built system that does not act up in some way, sooner or later.

My first PC dated May 1986 is still happily working and it has never ever had a fault.
Last quirky computer I had was still in 20th century...

Maybe the issue are the various peripherals and stuff people use, like the TV card you mention...its driver can and probably does freeze the system.
 
My oldest, daily-use PC, from 2001 (I upgraded it in 2003), freezes when cold. It runs rock-solid after at least an hour of operation, but you can't even boot OS when the machine is cold. The fastest way to bring it to operation after a night's sleep is: turn it on, enter BIOS, wait at least 10 minutes, then reboot (because it will likely freeze). Then boot system into single-user mode, run a script that produces 100% on CPU, and then keep it there for at least 30 minutes. After this, you cau use it for days, or even months.

I can do all this without turning on the display. Motherboard is at fault. The problem may be age-related, but it began during an overvoltage in 2011, that damaged power supply, GPU, and one RAM module.

Also, my other computer doesn't display anything on its PCI-32 graphics card, unless I insert a PCI-Express GPU and play with appropriate options in BIOS. So, it's impossible to use it without a PCIe card, at least temporarily.

Every time i had AMD based system, i had unpredictable and unsolvable issues.

I've only had one AMD PC. But it left me with such unfond memories that I've never tried an AMD computer again. Exception is AMD thin client I use as a router - so far, no issues, aside from random freezes on startup, and fact that it won't boot without a keyboard attached.

you might have been able to recover the data if you went to a specialist and took the platters out and read them on their special equipment.

There's no way to read platters outside the hard drive, and likely never will. The only reason to remove the platters is motor failure, but specialist always inserts them into another HDD to read them.
You insert flash chips into specialized equipment.
 
Buy components, build your own. Most of my computers over the years have been self-built. And only decommissioned because they got too old (5-10 years) and couldn't keep up any more, not because something broke.

Or the keyboard I spilled coffee on.
Never had that happen. Did have a cat pee in my keyboard though. Not just a little bit, it was completely soaked. Rinsed it under tap water, let it dry, worked again but pressing keys was somewhat "crunchy". Still ended up replacing it.
 
Apart from a ~fifteen years old Netgear switch, which needed to be powered off and on again every couple of months, everything, including myself assembled towers run almost rock solid stable for many years. If it cannot be rated fully rock solid then it's because of me: Either I messed up something on software basis, did some very stupid things with the OS, or once my cooler was full of too much dust. But else my hardware purred like a little kitten sleeping deeply and most satisfied.
occasionally (~1 in 20) boots up into the UEFI BIOS screen
That's a lot. I'd find that inacceptable. But as others mentioned here, the cause can come from several sources.
Not failing while high loads, but when booting, power and cooling problems (dust, badly assembled cooler) can be sidelined. I'd first check the connections of the drive(s) for loose connections: broken SATA cable, [carefully!] check if the connections are clean (a small dust particle or a tiny hair somewhere in the plug or socket can cause trouble). Next I'd try another drive.
 
What are your stories? Any remedy found?
Many years ago I bought an entry level Lenovo Notebook. Then a similar model for a mine friend. Both of them suffered of static charge accumulation. Every 6-12 months of use, they acted in strange ways: mine had boot problems, the other keyboard problems.

I had to remove the battery pack, short-circuiting various parts of the computer (e.g. USB, network interface, battery plug), pressing and releasing the ON button for some time.
 
Both of them suffered of static charge accumulation. Every 6-12 months of use, they acted in strange ways: mine had boot problems, the other keyboard problems.

I had to remove the battery pack, short-circuiting various parts of the computer (e.g. USB, network interface, battery plug), pressing and releasing the ON button for some time.
That's not how static charge works.
 
There's no way to read platters outside the hard drive, and likely never will. The only reason to remove the platters is motor failure, but specialist always inserts them into another HDD to read them.
This is the kind of thing I meant by 'special equipment'. Yes, sure, you have to have them spinning and a flying head to read them, so yes, they transplant them into another drive. Obviously a platter can't be read standalone, and you don't try to take the platter out yourself, that's what you pay the lab for, and/or to replace the electronics if that has failed.

1776249814528.png
 
That's not how static charge works.
It was a rather common problem. https://www.instructables.com/How-to-Drain-Static-Electricity-From-Your-Lenovo-T/

"If you're encountering issues with your Lenovo ThinkPad, such as random glitches, unresponsive keyboard or trackpad inputs, or slow loading times, this guide can help you safely* restore your laptop to optimal performance."

Maybe they call "static charge" but it was some other electrical problem. In any case, the solution worked.
 
It's a lot easier with the older style thinkpads with removable battery pack... just take battery out and hold down the power button for 30 secs or so. I don't like the new thin design where you have to take everything apart just to take the battery out. I don't think it's correct to call it static charge though... what you are actually draining is the residual charge on the capacitors in the machine... including on-chip capacitance.
 
Maybe they call "static charge" but it was some other electrical problem.
It's, at most, some capacitive charge you're discharging by holding down the power button (or short-circuiting various power connections). Static charge doesn't build up inside electrical components, that's impossible.

You know what static discharge is? Me petting my cats on a summer day (dry air) and get some sparks.
 
Buy components, build your own. Most of my computers over the years have been self-built. And only decommissioned because they got too old (5-10 years) and couldn't keep up any more, not because something broke.

Apart from aforementioned Olivetti machine bought around 1990, none of my computers have been OEM. All were built by hand. I had couple of 486s before first Pentium in 1997, and then I moved through P2s and Celerons onto P4 around 2002. Core Quad in 2008, i7 in 2015 and Xeon in 2023.

The P4 base I gave away to a colleague before entering the phase of reacquiring old gear, sadly. But it worked in 2018ish when we powered on the board. The Core Quad config is sitting below me, together with the Intel SSD bought in 2009/10 for it. The Xeon is in the same board i7 used to be; right now this is my most enduring motherboard (MSI X99S MPOWER) which has been daily driving for 11 years straight.

The only quirk I have on this computer is the situational nVidia interrupt storm stuff that affects the USB if the latter is doing high throughput.
 
Back
Top