Hardware Decay (real or not?)

8lack8rain

Limp Gawd
Joined
Nov 11, 2005
Messages
372
Has anyone heard of hardware loosing "speed" over time? I spoke with an IT at a major college in Chicago about this. Here we have someone that is an expert. But it just doesn't sound right to me. So as your system gets older it looses speed he tells me.

I didn't argue with him because I'm older and my time is more valuable. If I want answers I know where to look. But then later I thought, what they call it? How do you google that :)? I'm referring to it as decay, but could be wrong.

Someone else brought this to my attention, previously. They were not very adept with computers so I dismissed it as a rationalization to buy a new laptop/ victim of advertising. Then I heard it a second time I thought maybe there is something there.

My first thought is that any system has a cpu clock that requires a certain voltage... You don't loose voltage, so how can you loose MHZ?. Hehe, I suppose if your hardrive fails it could slow things down, but I was mainly interested in the cpu/bus/video/memory. Does anyone know about this?
 
All hardware has a life-cycle. Typically though, it doesn't decay, it just dies. Having said that, hardware designers often build redundant paths in critical designs paths to increase the parts longevity. These redundant paths are often longer and can make it seem like the hardware is running slower.

I know there are some designers here that can add more, so I will let them correct me if I am wrong. I am not a chip designer, but work closely with them.
 
Windows rot? :p

I am certianly no expert here, but I really don't think as far as cpu/gpu ram goes there is any realy degredation of performance/speed as the hardware ages. I would think things such as caps may not perform as well with age due to leakage, but as far as chip based hardware, I don't think so.

Of course, I could be completely wrong, but I am posting this from a 4 year old P4 Northwood based AGP machine, that runs just as well as it did the day I first fired it up.
 
I think there really is no decay. I have machines that are very old and still perform like day one.

That being said, things can seem slower I think because of some simple reasons:

1) If you go back to that older machine after getting a new one or using other peoples, it will seem slower. Human mind is easily tricked and we cant do anything about that.

2) Many times parts will have a failure after many years that may make the machine slower. Its not a decay, but a failure that has effected the performance of the whole machine.



The only thing that there is decay in IMO is data. It has been shown that over LONG periods of time, data can corrupt (in essence, decay) making it unusable. As far as parts though, it is just failure or 100% as it always was from my experience with items.
 
I'm fairly certain that computer hardware doesn't just get slower over time. I mean, my 486 plays doom2 just as well as it did ten years ago.
 
A chip either works 100% or it doesn't, at a certain frequency.

Sounds like a stupid point to make, but it won't slow down unless you tell it to. Theoretically, the maximum operating frequency (overclock) can decrease if electromigration does its work in the critical path.
 
Talonz said:
Theoretically, the maximum operating frequency (overclock) can decrease if electromigration does its work in the critical path.
I have absolutely no idea what you just said.
 
8lack8rain said:
Has anyone heard of hardware loosing "speed" over time? I spoke with an IT at a major college in Chicago about this. Here we have someone that is an expert. But it just doesn't sound right to me. So as your system gets older it looses speed he tells me.

Sure it does. If you're a normal person who never reinstalls Windows on a clean disk, never uninstalls anything, and happily runs anything other people give them.

That kinda hardware definitely loses speed over time. The root cause is that the owner knows bugger all about computer software maintenance.

Hardware wise, it'll keep operating to spec until it dies, at which stage it will die completely, or give so many errors that it might as well be dead.
 
I agree with the general opinion here... hardware either dies, or it works. However, I have a rather peculiar situation with my PC... it boots up fine (normally), but the CPU speed seems different every day. Its so fucked up that it cant even boot windows, yet it'll run linux pretty much perfectly. Its an athlon 2800+ XP (socket A), but the BIOS pretty much always (though not ALWAYS), detects it as 1250MHz. However, once Linux boots and gets a hold, the CPU clocks up to 2.08MHz. What is THAT all about?

Its been weird ever since it was struck by lightning (via the network port) (even though the mobo and psu have been changed... they were TOAST).
 
Back
Top