Will Overclocking Wreck My Pandora?


WizardStan said:
God Ginrai said:
But since you are not overvolting, you are not stressing the switches to flip them. So yes, they will snap, but I don't see any reason to believe that they will snap any large amount sooner than they would have if you hadn't overclocked.
The reason they "snap" is because they're only good for so many "flips". If they "flip" 50% more per second you're going to see an ultimate 33% drop in life expectancy.

Yes but lets not start to worry the original poster. Overclocking with the built in app isn't likely to shorten the life of the CPU beyond the device's reasonably expected lifetime.
 
Last edited by a moderator:
pelrun said:
Speaking as an EE:

HAHAHAHAHAHAHAHA NO
Speaking as someone trying to explain things in a way that the average person would understand, yes. A lot of analogies are used that don't follow reality, especially when dealing with electrons, to explain how things work. I don't understand why you've chosen this one in particular to laugh about.
If you can come up with a better analogy, a better way of explaining it to the layman, I'd be happy to hear it.
 
Last edited by a moderator:
All analogies break down at some point, but they need to be at least *partially* valid. Not even at it's most basic and abstract does the 'transistor=mechanical switch' analogy line up with reality, and it can't be used to give a worthwhile prediction of failure cause or timeline or preservation measures.

Transistors can fail; heat and high voltage can speed their degradation, not how many transitions they experience (although faster transitions will increase the heat output). And the failure mode isn't usually one of outright breaking; they just move far enough out of tolerance in speed, gain or noise resistance that the rest of the circuit stops functioning as intended at that speed or voltage or temperature. A lot of the time you can underclock or undervolt or improve the cooling when this happens and the chip will start working again.
 
pelrun said:
Not even at it's most basic and abstract does the 'transistor=mechanical switch' analogy line up with reality
Ok then, without using any words that might make someone think of a mechanical switch, explain how a transistor works as you would to someone without an EE degree.
 
Last edited by a moderator:
WizardStan said:
pelrun said:
Not even at it's most basic and abstract does the 'transistor=mechanical switch' analogy line up with reality
Ok then, without using any words that might make someone think of a mechanical switch, explain how a transistor works as you would to someone without an EE degree.

1. A Transistor Transists
2. Transisting is what a Transistor does

- Both comprehensive and accurate!
 
Last edited by a moderator:
Note that I didn't say 'transistor=switch' is completely invalid; at some level a transistor does look like a switch. But only if you're thinking of an 'ideal' switch; once you try and make the comparison with a physical switch and the failure modes associated with it does everything go pear-shaped.
 
Might be easier to say a transistor is more of a valve than a switch. When a small voltage is applied to the middle element of the transistor it allows the main power to flow through or not (think water through a faucet) the valve lets it flow or shuts it off. Although that analogy is mechanical, the transistor does it all with energy and nothing physically moves (apart from the flow of electrons that is)

So at 500mhz at normal voltage you could say standard size flow can be switched on or off at a rate of 500 Million times a second.
Overclocked to say 750Mhz would make the same flow switchable at 750 Million times a second.
Overvolting will do the same speed-wise but allow a higher flow of current. Which for reasons I cannot understand myself somehow will stabilize the increase of the rate of on/off of the flow as it goes higher. (I'm guessing cause you need more power to facilitate the switching process beyond a certain range)

Remember flow here is speed (volts) not Pressure which would be (amps) just so no one says I didn't clarify that and I don't even have a clue what the measure of amperage is through a silicon transistor. Way beyond me. I have a decent understanding of electronics not to say I understand it all. (was just a hobby as a kid, and I wasn't that advanced at all)
 
WizardStan said:
pelrun said:
Not even at it's most basic and abstract does the 'transistor=mechanical switch' analogy line up with reality
Ok then, without using any words that might make someone think of a mechanical switch, explain how a transistor works as you would to someone without an EE degree.
Transistors aren't really like switches at all. Think of them more as a delicate piece of art work. Expose them to the sun, run them hot, they will degrade. Just looking at them more won't wear them out any faster. Yes, they have a finite life, but that is more limited by diffusion and thermal stress than switching frequency.

Edit: that's not to say that overclocking won't have a detrimental effect, it probably causes more heat as the circuits end up fighting each other for more of the time - but this is a 2nd order effect compared with the impact of increasing the voltage.
 
Last edited by a moderator:
Poem58 said:
Might be easier to say a transistor is more of a valve than a switch. When a small voltage is applied to the middle element of the transistor it allows the main power to flow through or not (think water through a faucet) the valve lets it flow or shuts it off. Although that analogy is mechanical, the transistor does it all with energy and nothing physically moves (apart from the flow of electrons that is)

So at 500mhz at normal voltage you could say standard size flow can be switched on or off at a rate of 500 Million times a second.
Overclocked to say 750Mhz would make the same flow switchable at 750 Million times a second.
Overvolting will do the same speed-wise but allow a higher flow of current. Which for reasons I cannot understand myself somehow will stabilize the increase of the rate of on/off of the flow as it goes higher. (I'm guessing cause you need more power to facilitate the switching process beyond a certain range)

Remember flow here is speed (volts) not Pressure which would be (amps) just so no one says I didn't clarify that and I don't even have a clue what the measure of amperage is through a silicon transistor. Way beyond me. I have a decent understanding of electronics not to say I understand it all. (was just a hobby as a kid, and I wasn't that advanced at all)

That's an elegant explanation, however I know little about transistors, so I can't say if your analogy is correct or not. (But if it is, it's very good)

-God Ginrai
 
Last edited by a moderator:
Thanks GG. while I'll never claim to be an expert on anything, things I do know I always try to find a simple way of explaining.

For example a buddy I worked with asked me why "more stuff" fit on a DVD than on a CD. After all they are the same size. So I explained it like this...

I first asked him, if I told you I could fit 700 marbles in one box and 4,700 marbles in the same sized box, what would have to be different about the second set of marbles? He said "They would be smaller" Then I could see when the light bulb switched on in his head and he totally understood it when I told him the the information written was smaller and needed a different laser to read it.
Of course I told him there were differences in the way things were organized etc. but essentially that was why a DVD held more than a CD.

I fully believe that anything technology wise that the average Joe considers to be "over their head" can be explained easily, if you just try to equate it to something everyone knows. It's not always perfect and doesn't always cover the small details, but for someone who has themselves convinced that this stuff is more than they CAN understand, it can really make it easy for them to grasp.
 
Back
Top