Risks witch clocking

[Abaregi]

12

 

[hclarkjr]

i would leave well enough alone, it is his computer and if you happen to do something wrong and fry something.. guess who is buying new parts for him??

 

[alanwest09872]

IF they overcloced them what would stop you from purchasing the more expensive one. There all about the same price to manufactor but if you spend that extra cash on the faster one then they make more money. Why would they oc a cheaper proccessor that would just cause you to buy the cheaper one and they would get less money.

there are many other reasons but what it always come down to is the allmight dollar

 

[Drsignguy]

Originally posted by: Abaregi

Please help me clock my friends CPU He won't let me basing his argument that if it wasn't dangerous Intel would sell them at higher speeds.

Have Him start reading AnAndTech Forums/cpu and Overclocking preferably. He then just might understand.....possably

 

[VirtualLarry]

Free performance is free performance. If your friend saw a $5 bill sitting on the sidewalk, would he take it?

 

[Abaregi]

Originally posted by: VirtualLarry
Free performance is free performance. If your friend saw a $5 bill sitting on the sidewalk, would he take it?

Yes, but thats not the problem, He doesn't see at as risk free.
If it was i could clock it, but he is afraid that his cpu will fail.

I'm trying to convince him that there are no risks, he has a good cooler, he bought a new one over the stock because the stock was to noisy, and he bought a TRUE...

He keeps saying that if it was unsafe to overclock Intel would sell them overclocked and earn more cash.. Unless there was something unsafe with it.

 

[Idontcare]

Originally posted by: Abaregi

Originally posted by: VirtualLarry
Free performance is free performance. If your friend saw a $5 bill sitting on the sidewalk, would he take it?

Yes, but thats not the problem, He doesn't see at as risk free.
If it was i could clock it, but he is afraid that his cpu will fail.

I'm trying to convince him that there are no risks, he has a good cooler, he bought a new one over the stock because the stock was to noisy, and he bought a TRUE...

He keeps saying that if it was unsafe to overclock Intel would sell them overclocked and earn more cash.. Unless there was something unsafe with it.

It's not risk free, there is a reason overclocking your CPU invalidates your warranty.

My advice - stop trying to drag your friend to the water and forcing him to drink once you get him there. He's not into it. Let it be.

 

[VirtualLarry]

You could always tell your friend that you will buy him a new CPU if it burns out from overclocking.

 

[Concillian]

1) overclocking at stock voltage is almost completely no-risk Intel does do it, they sell multiple versions of exactly the same chip. This is to make more money off of those willing to pay for more performance. By and large, chip companies charge what they can get and not necessarily what something is worth in terms of manufacturing cost. They have thin margins on low end and high margins on high end
2) it takes some testing to find the 'edge'. Time = money. It's a lot easier to leave plenty of margin so your failure rate is excessively low.
3) Stock Intel heatsink designs are not conducive to the voltage that overclockers use. If you have a significantly better way of getting heat out of the chip, higher voltage & speed becomes safer. Intel designs for easy profit, and right now there is no reason to push their heatsinks beyond basic and cheap, because they're in the lead. Take a look at AMD heatsinks on the 140W Phenom in comparison, it's close to aftermarket quality. You typically see this from the leader / trailer. When AMD was leader and intel was trailer, Intel was the one pushing the heat / voltage envelope on their high end CPUs.

But if you're friend is not into overclocking, don't force him.

 

[dguy6789]

Originally posted by: Abaregi
Hi, i'm trying to convince a friend to clock his e4500 and he keeps saying that if it was good to clock them Intel would have done it. I find it hard to overcome this argument so i'm here for help

What both Intel and AMD do is test chips for working at the speed of their flagship models, and once they have enough for however many they want to be produced out of that batch, they just take the rest of the batch and test them for lower processor models. There is no physical difference between a top of the line chip and a mid range one.

Suppose there is a batch of 1000 cpus. They test one after the other until they have enough that reach 3.2Ghz or whatever their flagship model is. Then they take the rest and only test them up to 2.66Ghz or whatever they model they need to produce. This means that plenty of the processors that are limited to lower speeds can go up to the higher speeds without issue.

 

[Idontcare]

Originally posted by: VirtualLarry
You could always tell your friend that you will buy him a new CPU if it burns out from overclocking.

Best advice so far for the OP.

Assume 100% of the liability and then your friend is assured there is zero risk in doing an overclock.

If you aren't willing to assume 100% liability then you should feel guilty for trying to convince your friend it is risk-free (as you aren't convinced yourself in this case).

 

[bharatwaja]

Overclocking is a disease, especially here, AT is a hot zone.... Ask your friend to read the CPUs and Overclocking forum here at AT for a few days and he'll also get infected....

Like IDontCare points out, if you are not ready to assume complete liability then it sounds like you aren't buying your own pitch, how would your friend?

Besides, why are you so keen on OCing your friend's CPU? you've told him the advantages, if he's not interested, its his loss, let him be.... After all, its for ppl like your friend that intel sells higher specced CPUs at a higher price....

 

[CTho9305]

Originally posted by: Idontcare

Originally posted by: VirtualLarry
You could always tell your friend that you will buy him a new CPU if it burns out from overclocking.

Best advice so far for the OP.

Assume 100% of the liability and then your friend is assured there is zero risk in doing an overclock.

If you aren't willing to assume 100% liability then you should feel guilty for trying to convince your friend it is risk-free (as you aren't convinced yourself in this case).

How about dataloss? Losing my data would be worse than damaging my hardware

 

[Drsignguy]

I would say "back it up" anyway no matter if he Oc's or not. Besides, convincing someone that has no interest in over clocking would be a tough task, isn't worth it to me. He/she should read and attempt on thier own terms before having someone try to convince on what ( they) should or should not be doing. Just common sense I guess.

I also agree with bharatwaja, it can be a disease of sorts.

 

[Idontcare]

Originally posted by: CTho9305

Originally posted by: Idontcare

Originally posted by: VirtualLarry
You could always tell your friend that you will buy him a new CPU if it burns out from overclocking.

Best advice so far for the OP.

Assume 100% of the liability and then your friend is assured there is zero risk in doing an overclock.

If you aren't willing to assume 100% liability then you should feel guilty for trying to convince your friend it is risk-free (as you aren't convinced yourself in this case).

How about dataloss? Losing my data would be worse than damaging my hardware

You'll note I was keen to use the phrase "100% liability".

I chose those words to imply the OP would be taking on more than just insuring the CPU...which again shouldn't be something he should shy away from if he really believes it is risk-free. (data-loss, but not hardware loss, would not count as a risk-free scenario in my book)

I do think data-loss is under-rated for failed overclocking though. It is bar-none my number one concern when operating an overclocked system.

I don't run my systems at their peak stable OC with this in mind. I find the limit of stability and then back off the clocks around 10% to add further margin against the systemic issues that I would not detect in stability tests because of their non-fatal nature (making detection difficult).

 

[Denithor]

I have one personal rule I always follow when overclocking -- don't push the speed beyond what the chip will do on stock voltage. I just find the max stable speed at stock vcore and then back it off to the nearest "round" number (my e6400 would do 3.13GHz on stock volts, I backed it down to 3GHz and ran it like that for nearly a full year with complete stability--no issues whatsoever).

 

[FalseChristian]

Originally posted by: Denithor
I have one personal rule I always follow when overclocking -- don't push the speed beyond what the chip will do on stock voltage. I just find the max stable speed at stock vcore and then back it off to the nearest "round" number (my e6400 would do 3.13GHz on stock volts, I backed it down to 3GHz and ran it like that for nearly a full year with complete stability--no issues whatsoever).

That's find and dandy but noone here is going to get their E8400 to 4GHz without using alot more voltage. I'm using 1.425v for my 3.825GHz E8400 and is completely stable. You just got to keep the temps down when overclocking to the max.

 

[emilyek]

Sneak into his BIOS and change the memory multiplier and FSB when he's not looking.

When he notices that his machine is running faster, tell him you overclocked his rig and begin laughing manically (rig some squibs to go off as you do this) and scream "FEEL THE BURN!" at the top of your lungs.

Or not. Maybe that's not the best way to go about it.

 

[VirtualLarry]

You could always buy a new computer for your friend first, and then declare the first one worthless, and then ask him if he wants to overclock the first one.