First of all I really do not want to waste my precious hardware by blowing it up. But I have asked this question because I want to know, is there any way in modern time hardware so that it can be damaged just because of wrong/corrupted program/OS etc.

This is because I often tinker with my Linux based OS. Also I often get to know that e.g.
there are software fan control in some computers. Does it mean that if I am going to turn fan off completely, then it will certainly blow up my PC or is there any automatic processor shutdown mechanisms (using inbuilt thermal sensors) in use which can protect computer in overheating.

Also I have once read that by setting higher (unsupported) refresh rate in older CRT monitors blowed them up.

Does such kind of precautions with computer hardware has to be taken care of even in present time in daily uses.

So to generalise my question, "For consumer grade smartphones, tablets, laptops, desktops, and servers only (excluding industrial control systems embedded devices which may have controlling critical tasks hence wrong programming can be obviously fatal) how many possible ways are there so that we can damage them up using bad coding.

I don't think there is any longer a simple answer like there used to be. This is because modern motherboards, excepting many OEMs, allow the user to set parameters like clock speed, fan speed and power management. It is possible on some boards (and with some kernel and power options) to disable safeguards. It is fairly unlikely to do actual hardware "blow ups" but it is possible so look before you leap.

If you flash a bad BIOS/UEFI firmware you can brick the system, and that's probably the worst you can do. Newer systems have some protection against bad flashes, but it is limited.

There is no real way the CPU will overheat because it has methods that will throttle or cut the power before that happens.

I suppose a badly designed monitor might be damaged by bad settings, but it is rare.

I'm not convinced that this is always the case. Before I replaced the laughably pathetic stock cooler on my AMD FX8120 it could easily hit >70C running mprime with no sign of shutting down -- according to AMD this will result in CPU damage with anything from 55C upwards being a bad idea. Admittedly I didn't leave it to see whether it would shut down as I didn't want to have to buy a new CPU but if AMD aren't lying then it seems I could easily have left my CPU to fry itself.
With graphics cards I seem to recall that some open source NVIDIA drivers had problems with fan speed control -- not switching on the fan could, I'm sure, mean the card cooking itself.

There are some acer aspire laptops that let the os, (Vista) take over cooling fan control after post, this results in a dead fan when not using Vista, the machines overheat and I have seen a couple that completely failed due to solder connections overheating before the machine shut itself down. my solution was to solder the red fan wire to the positive usb supply. the fan then runs continuous.
Bad decision on the manufacturer to allow the os to control the fan.

Doing a huge number of writes to a hard drive will eventually wear it out. This is true for both rotary HDs and for SSDs.

It is true that the maximum temperature is rated at 70, but the thermal cutoff is at 90C:
http://www.overclock.net/t/1241993/t...fx-8120-limits
http://www.overclock.net/t/1128821/a...tion-and-guide

Graphics cards do vary, but they do throttle as well.

This doesn't mean you can't damage these by running them hot. If you want to keep your computer for a while, get good coolers to extend its life.

Mostly what I'm saying is that you're unlikely to see the processor burning a hole in the mobo like they used to.

Moved: This thread is more suitable in Linux-Hardware and has been moved accordingly to help your thread/question get the exposure it deserves.

That is why I think if you could get your CPU running at 85C (plausible with my old cooling set-up) for 24 hours or so you could well damage it.
As I understand it the thermal cutoffs are more about preventing single catastrophic failure events so running anything just below the thermal cutout will be damaging it and could prove fatal in the mid to long term?

Hi,

Modern 'SSD' do not have the life issues like older 'SSD'. Modern 'SSD' controller do a great job of control when the system is setup properly. Life for a 'SLC' does exceed a 'MLC' based 'SSD'. Most of the newer 'SSD' will surpass the expected life of a modern system.

Even a modern 'hdd' well last over the long term if the system has been configured properly. Most newer 'HDD' do incorporate a larger cache and do support 'write back' thus providing improved I/O performance. Not all consumer 'hdd' have 'write back' or even larger cache memory on the drive while some will have a 'write through' cache.

@dugan please substantiate this general statement;Or at least show some proof in this comparison. Manufacture data, test benches or what can prove the failure. Most modern SLC & MLC 'SSD' do not have the failure rate of years past. Even newer consumer grade 'hdd' do not have a large write fail rate. Western Digital NAS specific 6Gb/s drives (MTBF 1,000,000 Hours) is a drive that would get a lot of reads or writes per second over the network. I randomly selected a drive test from http://www.tomshardware.com. That would be '41,666.6666667' days for the 'MTBF' for a lifetime. Sure 'MTBF' is using a standardized test thus in the field install the term may be even larger or smaller. Mechanical systems do wear.

For a 'SSD', Patriot Vertex Tom's hardware review Please notice the 'mean time to failure' above.
Looks to me that my tenth generation grandchildren could use my Laptop with a enterprise Patriot 'SSD' installed. That is if they knew how to turn it on let alone connect to the newer power generation system. Any Electrolytic Capacitors in the Laptop would be wasted by then.

A consumer grade patriot 'SSD' would have a 'MTTF' of 2,000,000 hours or 83,333.3333333 days. Long past my lifetime let alone the live of my system.

Most manufactures do provide 'MTBF' & 'MTTF' data for their hardware. Can hardware fail before the stated data? Sure can! Not overall for all equipment use! Environmental conditions will dictate wear or failure.

Thermal cutoff is different from throttling, so I'm sure you could get the temperature that high in the first place. If you could, then yes you could damage it over time.

The two ways I can think of is soft overvolting a component or a bad firmware flash.
Otherwsie it is unlikely

at a security conference recently there was a demonstration for a Apple Mac to catch fire

the CPU was over clocked and the fan was disabled. I think this was a remote attack ?
but but a edit to the firmware might have been needed

Greetz
Technically some of the example listed here are not hardware "blowing up" but rather software failure. A prime example is bios flashing. There is nothing a person can do from the keyboard that will actually destroy or even damage the physical chip. In fact, it will still be possible to restore the bios under the right conditions.

I'm not certain how modern UEFI handles the Boot Block, and some interim BIOS boards were effectively locked out from it's use because of lacking of the old busses that were supported. It used to be that if you had a floppy drive and an ISA slot, which were supported within the boot block, you could get back up to redo the flash.

Interim boards required having a similar chipset board so that one could hot-swap the BIOS chip in order to flash it. I have actually done this many times. IIRC modern UEFI boards have a complete backup instead of a traditional Boot Block but afaik all BIOS chips of any type have ways to restore the code.

There was a time when there was so little user config'd code in BIOS that it was impossible to do damage to hardware from the keyboard with the single exception of Vertical Refresh Rate on displays. Voltages, for example, were set by hardware DIP switches if available at all. Over time features were added so that users could set parameters by software, at first mostly on Enthusiast or Performance boards supporting Overclocking and After-Market coolers. Later it became obvious that more control meant more flexibility to handle more kinds of hardware so these features became commonplace on most standalone boards but are still uncommon on OEM boards who assume their users are ignorant.

Back in the day when phone modems were common, I witnessed Black Hat hackers referring to "smoking modems" by altering ping parameters to "many vikings in longboats" but the smoke was largely imaginary. All that resulted was buffer overruns or angry ISPs.

Today it is still nearly impossible to damage hardware provided voltages and thermal parameters are properly set in BIOS. However, afaik, Vertical Refresh is still one possible exception.

No not that either. LCD monitors don't care and even newer CRT would shut down on an out of range signal