Question is it true that a good overclock does not decrease the "health of the processor"?

[Grealish01]

I have to buy a 12600k and my reasoning is this, for 4/5 years I don't overclock at its own undervolt to keep the cpu "healthy" then from the fifth year onwards I overclok. this is because in 4 years I will start university, now I do not need a lot of computing power, but probably already from the second year of computer science and therefore I could do overclocking, what I perhaps question with respect to my hypothesis is that if already since immediately overclocked it would not change anything. I ask you because it is a world that I do not know well, thanks in advance.

 

[Grealish01]

I have to buy a 12600k and my reasoning is this, for 4/5 years I don't overclock at its own undervolt to keep the cpu "healthy" then from the fifth year onwards I overclok. this is because in 4 years I will start university, now I do not need a lot of computing power, but probably already from the second year of computer science and therefore I could do overclocking, what I perhaps question with respect to my hypothesis is that if already since immediately overclocked it would not change anything. I ask you because it is a world that I do not know well, thanks in advance.

another question: what is it that really makes processors and motherboards "age"?

 

[lvt]

Temperature is the killer.

Even if you don't overclock but you keep your CPU at high temp for prolonged period of time, it will hurt the CPU.

 

[Grealish01]

Temperature is the killer.

Even if you don't overclock but you keep your CPU at high temp for prolonged period of time, it will hurt the CPU.

Only temperatures?

 

[lvt]

Only temperatures?

Yes, if you discard all physical damages, temperature is enemy n° 1.

 

[USAFRet]

Only temperatures?

Almost 100% temperature.

Additionally, overclocking has become increasingly less importance in the last few years.

ex:
A couple percent difference in the OC brings your gaming FPS from 150fps to 155fps.
Irrelevant.

Similarly, undervolting.
Why?

Just leave the thing alone.

 

[Grealish01]

Almost 100% temperature.

Additionally, overclocking has become increasingly less importance in the last few years.

ex:
A couple percent difference in the OC brings your gaming FPS from 150fps to 155fps.
Irrelevant.

Similarly, undervolting.
Why?

Just leave the thing alone.

thanks, one last thing, between cpu and motherboard which of the 2 tends to age more / lose performance / break? motherboard?

 

[Grealish01]

Yes, if you discard all physical damages, temperature is enemy n° 1.

because now I need a PC, for three years I will not need much power, in 4 years I will need more power in general. And I don't know whether to get a CPU that is already good now, whether to get a very decent CPU now (which may suit me) and in just 3/4 get the powerful one. Or whether to upgrade with the motherboard

 

[USAFRet]

thanks, one last thing, between cpu and motherboard which of the 2 tends to age more / lose performance / break? motherboard?

Motherboard, most likely.

I've never had a CPU 'die'.

But for the singular device sitting under your desk....there is no definition of "more likely".


Both will be replaced due to performance before they actually die.
Not 'decreasing from original performance', but rather 'can't keep up with current use and software' performance.

 

[USAFRet]

because now I need a PC, for three years I will not need much power, in 4 years I will need more power in general. And I don't know whether to get a CPU that is already good now, whether to get a very decent CPU now (which may suit me) and in just 3/4 get the powerful one. Or whether to upgrade with the motherboard

Build the best system that fits within your budget.

Worry about "4 years from now" in 2026.

 

[Grealish01]

Build the best system that fits within your budget.

Worry about "4 years from now" in 2026.

do you think that with a 12600k in 4 years I will be able to program and finish the university with that CPU? (in 8 years)

 

[USAFRet]

do you think that with a 12600k in 4 years I will be able to program and finish the university with that CPU? (in 8 years)

All depends on what you mean by "program", and what software you'd be using.

Right now, I could resurrect my ancient i5-3570k system (2012), and do pretty much the same programming as any recent system.
Because the main server lives on a whole different system.
I could probably do it from my phone.

Your supposed 12600k system will be just fine.

You're getting lost in the theoretical details, and analysis paralysis.

Leave the OC and undervolt alone.
Buy it, build it, enjoy it.

 

[Grealish01]

All depends on what you mean by "program", and what software you'd be using.

Right now, I could resurrect my ancient i5-3570k system (2012), and do pretty much the same programming as any recent system.
Because the main server lives on a whole different system.
I could probably do it from my phone.

Your supposed 12600k system will be just fine.

You're getting lost in the theoretical details, and analysis paralysis.

Leave the OC and undervolt alone.
Buy it, build it, enjoy it.

All depends on what you mean by "program", and what software you'd be using.

Right now, I could resurrect my ancient i5-3570k system (2012), and do pretty much the same programming as any recent system.
Because the main server lives on a whole different system.
I could probably do it from my phone.

Your supposed 12600k system will be just fine.

You're getting lost in the theoretical details, and analysis paralysis.

Leave the OC and undervolt alone.
Buy it, build it, enjoy it.

thank you very much for these clarifications😃 ... ok, so for 8 years I'm fine😅. very last thing: instead motherboard do you think I will need an upgrade from now to 8 years?

 

[USAFRet]

thank you very much for these clarifications😃 ... ok, so for 8 years I'm fine😅. very last thing: instead motherboard do you think I will need an upgrade from now to 8 years?

You only need an upgrade when the current system no longer meets your needs.

 

[Grealish01]

You only need an upgrade when the current system no longer meets your needs.

but I say, the same speech you made with the cpu also applies to the motherboard? or does the latter tend to age faster?

 

[USAFRet]

but I say, the same speech you made with the cpu also applies to the motherboard? or does the latter tend to age faster?

The CPU and motherboard are pretty much a unit.

Changing the motherboard brings no "performance" gain.
Changing the CPU might.

 

[King_V]

Similarly, undervolting.

Slight digression - wouldn't undervolting (assuming no change in clock frequency) result in less heat, thus possibly slightly extending the lifespan of the CPU? At least in theory?

 

[USAFRet]

Slight digression - wouldn't undervolting (assuming no change in clock frequency) result in less heat, thus possibly slightly extending the lifespan of the CPU? At least in theory?

"in theory", yes.

But a CPU left at stock voltage and clock, that is never overheated, will last.......until it dies. Sometime a few decades from now.

They are rated to run at a max of XXc Keep it below that, and anything less does not really change things.

Changing the lifespan from 30 years to 32 years is irrelevant.
And REALLY irrelevant for a singular instance of a device.

2 CPUs (or other devices)...one on your desk, one on mine...treated exactly the same.
One may die while the other is still going strong.

Or the one on your desk, being undervolted...may still die earlier.

Lifespan numbers really only count in fleetwide stats.
0.0X% died after 20 years, with no undervolt.
0.0Y% died after 20 years with a slight undervolt.

If yours is in that X or Y....oh well. The other 99.0X% had zero issues either way.

Now...if you're messing with different (and maybe insufficient) cooling and overclocks....then maybe delve into undervolting.
Otherwise, leave it alone.

 

[Grealish01]

"in theory", yes.

But a CPU left at stock voltage and clock, that is never overheated, will last.......until it dies. Sometime a few decades from now.

They are rated to run at a max of XXc Keep it below that, and anything less does not really change things.

Changing the lifespan from 30 years to 32 years is irrelevant.
And REALLY irrelevant for a singular instance of a device.

2 CPUs (or other devices)...one on your desk, one on mine...treated exactly the same.
One may die while the other is still going strong.

Or the one on your desk, being undervolted...may still die earlier.

Lifespan numbers really only count in fleetwide stats.
0.0X% died after 20 years, with no undervolt.
0.0Y% died after 20 years with a slight undervolt.

If yours is in that X or Y....oh well. The other 99.0X% had zero issues either way.

Now...if you're messing with different (and maybe insufficient) cooling and overclocks....then maybe delve into undervolting.
Otherwise, leave it alone.

instead of avoiding or lengthening that moment in which the cpu dies, is there a way to ensure that in a very long period of time a CPU does not lose even an iota of performance? (Does the undervolt help in this?)

 

[Lafong]

"Ensure" implies certainty.

I can't imagine such a case, considering randomness and the molecular subtleties of YOUR particular CPU.

 

[USAFRet]

instead of avoiding or lengthening that moment in which the cpu dies, is there a way to ensure that in a very long period of time a CPU does not lose even an iota of performance? (Does the undervolt help in this?)

Assuming no over temp, a CPU does not lose performance over time.

If the particular little chip YOU have were to lose performance over time, it would do that anyway, no matter what you did.
(but it won't)


Again, you're chasing problems that don't exist, for a system you do not yet own.

 

[Unolocogringo]

I just retired an Asus m5a97/ Phennom2 x6 system overclocked @3.6. Old system . That has folded all of its life, 24/7/365 use.
Still running an Asus Z170 and I5 6600 with a 1070 folding . Same thing.
As stated above the MAIN destroyer is HEAT. Next is overvolting. Next is cheap power supplies.
Do not over volt. let it run hot or give it crappy power and systems will last many years. Most past their useful life for the original user. But could still be used for email,internet , coding, app development and other non stressful apps.

 

[ex_bubblehead]

instead of avoiding or lengthening that moment in which the cpu dies, is there a way to ensure that in a very long period of time a CPU does not lose even an iota of performance? (Does the undervolt help in this?)

Please stop. This is a total non issue. Just buy the components, assemble the PC and use it until such time as it no longer serves the purpose for which it was built.