Should Overclocking Be Automatic?

[VaporX]

Okay a bit of clarification here for the topic. The question is should a DIYer presume to overclock with every build given that ALL CPUs (at least all I have seen) can get at least a 10% to 15% boost? With this in mind should it be a given that a DIYer will do this basic, no voltage bump overclock all the time?

[WhenKittensATK]

I think it really falls down to what you plan on doing with the computer. If it's not going to be used often for CPU intensive work, then it may not even be a noticeable difference.

[ivangela]

That's all I've done as I only want to max my folding output.

I figure if I don't increase my voltage then I won't shorten my CPU lifespan.

 

Soooo, yes bas ic OC should be doen by all!

[Speedway]

I think it really falls down to what you plan on doing with the computer. If it's not going to be used often for CPU intensive work, then it may not even be a noticeable difference.

Depends on the applications and use! If it's not needed, then why do it?

 

I figure if I don't increase my voltage then I won't shorten my CPU lifespan.

If my volteage is higher than yours, but my temps are lower, do you think that means your cpu will last longer than mine? Because, that may or may not be true The cooling solution plays an important role in lifespan as well!

[hierovision]

I vote no. Increasing frequency still raises the power consumption even if the voltage is left stock. ALL DIY builds should be set at stock unless requested otherwise.

[gotdamojo06]

Totally depends on what the customer is looking for.

 

I did vote yes because your every day user will see that the CPU is at 3.5GHz and when comparing it to a CPU at the next store it is only at 3.2GHz, they will come back and buy yours

[Drdeath]

Depends on the applications and use! If it's not needed, then why do it?

 

 

If my volteage is higher than yours, but my temps are lower, do you think that means your cpu will last longer than mine? Because, that may or may not be true The cooling solution plays an important role in lifespan as well!

 

 

LOL.....And have we tested lifespan at various temps. The difference will be minimal if they're kept under manufacturer specs. Most will have pitched there rig way before the cup dies. I like the I am with stupid too. Just don't see who it us directed at....

Edited May 16, 2011 by Drdeath

[ivangela]

 

If my volteage is higher than yours, but my temps are lower, do you think that means your cpu will last longer than mine? Because, that may or may not be true The cooling solution plays an important role in lifespan as well!

 

Good point. I don't know the answer to that. Intuitively I would say you might be right but I would need to really test that to agree.

 

I vote no. Increasing frequency still raises the power consumption even if the voltage is left stock. ALL DIY builds should be set at stock unless requested otherwise.

 

So what if my power consumption is higher. Are you worried about money. Does a 3.4 frequency burn that much more power than a 3.0 that I would notice or care about the extra $?

 

LOL.....And have we tested lifespan at various temps. The difference will be minimal if they're kept under manufacturer specs. Most will have pitched there rig way before the cup dies.

 

Have you tested lifespan at various temps under manufacturer specs? Do you know the dif is minimal? Having taken too many stats classes, I have to question that assumption and not having built 30 or more of the exact same rig and run them identically for the same amount of time or until death and then run the appropriate statistical analysis I couldn't give you intelligent answer.

Also, with the modern speed of processors, I might actually hang on to a functional rig WAY longer than I used to because now they remain much more useful for much longer if i just want to check email or surf or watch movies, etc, so I don't even agree that I would pitch my rig before the CPU dies

[HatersGonnaHate]

Well there are other issues at play here, but I did vote yes.

 

It should be considered whether the processor is only compatible with FSB overclocking or if Multiplier is an option. With FSB it is then worth discussing whether the memory will be affected by the overclock; which a beginner should be mindful of.

 

If the MoBo has an overclocking utility that increases the Multiplier does it also tweak other settings?

 

I think everyone should experience overclocking for themselves, and I do recommened it. But, when is it the right time to learn?

 

 

[Guest_Jim_*]

I won't vote, but I'm thinking no because there is a reason the manufacturer has a specific stock speed, because that is the speed they can expect with very high confidence (I'd guess they go for 99% confidence) that all the units will run without error. If they could push farther without error, wouldn't they so they could offer "better" parts, and charge more? I would at least. The point I'm trying to make is that there is a reason for the stock speed being where they are, so always overclocking the parts may not be a good idea. Upon request, of course, but not just because you're building it.

[El_Capitan]

I say, yes. Even with the stock cooler and stock voltages, you can gain the 10% - 15% overclock without going over the recommended max temps during load. At least from my experience with the Phenom II x3 720 BE from 3.2GHz to 3.4GHz and the Phenom II x6 1090T from 3.2GHz to 3.6GHz, and the i7 920/930/950 from their stock frequencies to 3.4GHz. It just seems silly not getting the extra speed you can out of it.

 

You can also get a lot more than a 10% - 15% overclock by adding a good aftermarket cooler. From that standpoint, I overclock from three assurances (for peace of mind):

1. Under maximum allowed voltages and under maximum temps.

2. Over maximum allowed voltages and under maximum temps.

3. Over maximum allowed voltages and over maximum temps.

 

Obviously #1 would be the safest overclock if you're going to overclock with higher voltages. I normally do #2, but never ever do #3.

 

Now, the reason why I say yes is, that almost all CPU's are capable of this. The reason manufacturers state the clock speed to default is so it's guaranteed to work at those tolerance levels. It's like running a 100m race where you need a qualifying time of 10.5 seconds - those that don't make the the cut are out. Yes, you can run faster, but it means nothing until the real race (in this case, overclocking). There are multiple reasons manufacturers would specify a default clock speed:

1. Marketing - bad advertising to say their CPU is rated at 3.2GHz (3.5GHz possible from stock voltage overclock) up to 3.6GHz Turbospeed (3.9GHz possible from stock voltage overclock)

2. They can't individually test how much each CPU can overclock to on default voltages, it would be too time consuming

3. Every motherboard and memory might limit the potential overclock of he CPU under optimal conditions. You might get a better overclock with a different motherboard and a better set of memory

 

At any rate, those that can, should. Those that don't are just missing out. Any DIY'er should automatically see what their CPU is capable of overclocking to at stock voltages.

[wevsspot]

I voted no - mainly because the moniker DIY sometimes seems to take on the incorrect meaning that a DIYer is also competent to overclock their hardware without tearing something up. I think that it's taken for granted that anyone that can build a computer can also safely overclock a computer. We see it here all the time - new member pops in and his/her first post is "hey I just built my new computer and it has xxx, xxx, xxx in it and I want my i7 930 to clock up to 5.0Ghz using the stock cooler - please give me the settings........................) I know that I'm exaggerating a little bit there, but you get my point.

 

I probably would have voted yes if there had been some caveat or clarification as to when a DIYer is ready for overclocking. Do they know the importance of proper cooling, safe voltage ranges and even the need for a good to great power supply? I think there are too many variables for a simple and straight "yes" or "no" answer to the poll question.