Rumour: Bulldozer 50% Faster than Core i7 and Phenom II.

[Rifter]

Right, so you want to prove that overclocking does damage by linking to an article demonstrating how over-volting does damage? You don't need to over-volt to get a 300MHz increase, and a 1-2C increase in CPU temps won't degrade a chip. At this point, if you're gonna argue about temps, room temperature will make more of a difference than the OC. A mild OC on stock voltage won't do anything to a chip.

If it's machines in a home office or home business you're gonna support yourself instead of an enterprise, there's no problem with a mild OC.

But leave it to you to spread paranoia and BS.

maybe you should read the article jackass

"Too many people believe overclocking is "safe" as long as they don't increase their processor core voltage - not true. Frequency increases drive higher load temperatures, which reduces useful life."

 

[Lonbjerg]

What it says. He's always spreading BS. If you're paranoid over a chip increasing in temps by 1-2C and being degraded in any significant matter by a 300MHz OC, you're a moron. The amount of degradation over-volting with OCing causes is an order of a magnitude higher than what a mild OC with no over-volting will.

You are still unable to grasp the term "enterprise" and how it affects IT and stuff like uptime / data integrity / TOC...stuck in your enthusiast mindset...and you point the finger at others?

Mindboggling...to say the least.

TL;DR
You ignorant in one field and you run of and display your ignorance in yet another field and claim others have it wrong

 

[AtenRa]

1-2c will have zero to minor effect in the CPU degradation without OverVolting and 10% OC over the default frequency will be within the manufacturer limits, after all they (Both AMD and Intel) OC there CPUs with turbo.

Yes i know turbo dont run all the time but not because of a degradation effect but for power efficiency.

I will like here to say that even if you could OC your CPU 300MHz or 10-15% with 1000% stable, no one in enterprise will do it.

Home and small office users may do it but no admin or enterprise technician will OC any PC or Workstation or even an Office client for a lot of reason but not of CPU degradation.

 

[GaiaHunter]

But since many people overclocking buy better coolers the temperature actually drop.

I went from 2 cores to 4 core, 3.2->3.6 without any voltage increase and my temperatures dropped some 10-15ºC from 2 stock cores -> 4 OCed cores.

Still for business OC is a no go (well some PCs aren't actually doing anything important).

 

[Rifter]

What it says. He's always spreading BS. If you're paranoid over a chip increasing in temps by 1-2C and being degraded in any significant matter by a 300MHz OC, you're a moron. The amount of degradation over-volting with OCing causes is an order of a magnitude higher than what a mild OC with no over-volting will.

I have provided links, from our own webiste to backup my statements. All you have done is resort to personal attacks and zero data to backup your claims. Also i never said a 1-2c difference would effect the chip long term i just said temp is just as much of a factor as voltage. I think its obvious who is spreading BS.

 

[LOL_Wut_Axel]

You are still unable to grasp the term "enterprise" and how it affects IT and stuff like uptime / data integrity / TOC...stuck in your enthusiast mindset...and you point the finger at others?

Mindboggling...to say the least.

TL;DR
You ignorant in one field and you run of and display your ignorance in yet another field and claim others have it wrong

You are still unable to grasp the term "reading comprehension" and how it affects the stupidity you just wrote. I am not arguing for mild OCing in enterprise; I'm arguing for home and home business users. Now learn to read before you spew crap.

 

[LOL_Wut_Axel]

I have provided links, from our own webiste to backup my statements. All you have done is resort to personal attacks and zero data to backup your claims. Also i never said a 1-2c difference would effect the chip long term i just said temp is just as much of a factor as voltage. I think its obvious who is spreading BS.

You provided "proof" for an argument that was never made. I was arguing over mild OCs with no over-volting. The article was arguing over how over-volting degrades CPUs and by how much and in what time-frame. If you're not over-volting the article does not apply. Are you gonna be arguing now over the chip degrading from the mild OC in 20 years? Get a clue of the context of my arguments and get back to me.

 

[Rifter]

You provided "proof" for an argument that was never made. I was arguing over mild OCs with no over-volting. The article was arguing over how over-volting degrades CPUs and by how much and in what time-frame. If you're not over-volting the article does not apply. Are you gonna be arguing now over the chip degrading from the mild OC in 20 years? Get a clue of the context of my arguments and get back to me.

do you know how to read? this line is in the article:

"Too many people believe overclocking is "safe" as long as they don't increase their processor core voltage - not true. Frequency increases drive higher load temperatures, which reduces useful life."

How are you able to read english and not think this is talking about when you DO NOT over volt?

Voltage is not the only thing which effects lifespan.

 

[LOL_Wut_Axel]

do you know how to read? this line is in the article:

"Too many people believe overclocking is "safe" as long as they don't increase their processor core voltage - not true. Frequency increases drive higher load temperatures, which reduces useful life."

How are you able to read english and not think this is talking about when you DO NOT over volt?

Voltage is not the only thing which effects lifespan.

Which is a point I already countered. A 300MHz OC will result in 1-2C increase of full load temperatures. Going from 65C to 67C on a stock Core i5 2500 cooler at full load will theoretically decrease its lifespan. But that's the problem. The argument they're making is theoretical and not practical. In the real world, the CPU will probably last more than 20 years. Room temperature will make more difference in temps than the OC will.

And if you're so concerned about temperatures, you can just buy a cheap air cooler like a Corsair A50 and temps will drop by more than 10C, thus meaning the CPU with the original clocks + stock cooler should theoretically last less.

 

[Rifter]

Which is a point I already countered. A 300MHz OC will result in 1-2C increase of full load temperatures. Going from 65C to 67C on a stock Core i5 2500 cooler at full load will theoretically decrease its lifespan. But that's the problem. The argument they're making is theoretical and not practical. In the real world, the CPU will probably last more than 20 years. Room temperature will make more difference in temps than the OC will.

And if you're so concerned about temperatures, you can just buy a cheap air cooler like a Corsair A50 and temps will drop by more than 10C, thus meaning the CPU with the original clocks + stock cooler should theoretically last less.

All i was ever saying is that temps and voltage will both degrade a CPU. You seem to think only voltage will degrade a CPU.

I provided an article that agrees that it is not "safe" to overclock even without voltage increase.

Why would you do something which is unsafe to a machine you rely on to make money?

Im all for overclocking on gaming/personal computers. I've overclocked every CPU/GPU i have ever owned. I have personally burned out CPU's/Mobos over the years pushing the limits so i know first hand how unsafe of a practice it is, and would never do it on a machine i relied on to make money. Either on a large corporate scale or even at home.

 

[podspi]

Speaking of OCing, the rumor that I heard (from where I can't remember?) is that the respin was to improve clocks, but not stock clocks, but overclocking head room.


Here's hoping that's true...


Given the recent Tom's normalized benchmarks, I think it is pretty clear that if BD can get near Nehalem IPC BD should be a winner. It is going to be very interesting seeing BD benchmarks, because I expect it is going to lose horribly in some synthetic benchmarks, given BD's peak maintainable throughput (PER CORE) is lower than even K10... Average IPC should increase though. And if CodeAnalyst on my machine is any indication, IPC has a lot of room to grow.

 

[Rifter]

Speaking of OCing, the rumor that I heard (from where I can't remember?) is that the respin was to improve clocks, but not stock clocks, but overclocking head room.


Here's hoping that's true...


Given the recent Tom's normalized benchmarks, I think it is pretty clear that if BD can get near Nehalem IPC BD should be a winner. It is going to be very interesting seeing BD benchmarks, because I expect it is going to lose horribly in some synthetic benchmarks, given BD's peak maintainable throughput (PER CORE) is lower than even K10... Average IPC should increase though. And if CodeAnalyst on my machine is any indication, IPC has a lot of room to grow.

I doubt it, they do not need to improve OC headroom as they only gurantee stock clocks. It would be a waste of money for them to respin and not increase stock clocks. If they increase stock clocks then at least they could charge more for there higher performaning parts and recoups costs.

 

[LOL_Wut_Axel]

All i was ever saying is that temps and voltage will both degrade a CPU. You seem to think only voltage will degrade a CPU.

I provided an article that agrees that it is not "safe" to overclock even without voltage increase.

Why would you do something which is unsafe to a machine you rely on to make money?

Im all for overclocking on gaming/personal computers. I've overclocked every CPU/GPU i have ever owned. I have personally burned out CPU's/Mobos over the years pushing the limits so i know first hand how unsafe of a practice it is, and would never do it on a machine i relied on to make money. Either on a large corporate scale or even at home.

No, I said that over-volting with over-clocking is an order of a magnitude more degrading to a CPU than just a mild over-clock.

And the article never says that mild OCing is not safe. It just says it can degrade the chip theoretically by the way of higher temps, which are more affected in this case by room temperature anyway. A Core i5 2500 at 3.6GHz with 67C full load temp and one at 3.3GHz with 65C full load temp should both last over 20 years. If you're worried about it, buy a cheap air cooler. That will decrease temps by more than 10C and may make you feel better about the chip lasting 40 years instead of 30.

I would do this to a machine which I rely on because it would increase efficiency and because the theoretical degradation of the CPU will be way after it has to be replaced, if there's any. I'm certainly not gonna worry about the chip needing higher voltage to run in 30 years, if it needs it at all.

As for your personal experience, that's why the article says that you should make mild over-volts if you're an enthusiast that wants his CPU to last a good time before it degrades. If you do this, you may have the chance of running the machine that way for five years or more before you have to push higher voltage through it. If you don't know when to stop, that's your problem.

I'd heavily advise people that administer and service the computers themselves in their home or home business to do a mild OC with no additional voltage if it's a recent CPU. It will increase efficiency by a good amount with no side effects, apart from very slightly higher temps.

 

[Phynaz]

But leave it to you to spread paranoia and BS.

Anyway, you should try to stop getting this thread derailed further, especially since what you're arguing over I've already addressed in the thread.

If you're paranoid over a chip increasing in temps by 1-2C and being degraded in any significant matter by a 300MHz OC, you're a moron.

You are still unable to grasp the term "reading comprehension" and how it affects the stupidity you just wrote.

Get a clue of the context of my arguments and get back to me.

Do you find this approach works to get people to listen to and understand your viewpoint?

 

[bridito]

I'd heavily advise people that administer and service the computers themselves in their home or home business to do a mild OC with no additional voltage if it's a recent CPU. It will increase efficiency by a good amount with no side effects, apart from very slightly higher temps.

First of all I'd like to apologize for starting this firestorm as I was only expressing my personal view about what I do or don't do to the computer that pays my rent. However, as for convincing "home business" people to OC:

1) I assure you that at least 95% and possibly 99% of all home business people think OC is Orange County and wouldn't know how to set up an overclock if you paid them.

2) That same percentage would look upon the prospect of doing that with fear, anxiety, and very possibly panic.

No one is to be blamed for any of this as we all have our own perspectives. As a non gamer non OCer primarily business user, OCing is simply not an option. I'm not going to void my warranty and I'm not even going to take the 0.00000000000000001% chance that I'm going to fry something one nanosecond before it was going to go south by itself anyway. My computer makes me earn a three figure per hour rate when I'm consulting (not full time... I'm not rich unfortunately) and if I want faster performance I'll pay for the box stock version of it and leave it that way. I'm NOT saying that people who OC are wrong but they are in the overwhelming minority (as measured among all installed PCs on the planet) and I can't see any scenario that would change that and increase the current "bare minority" status. Again, that's just IMHO.

 

[AtenRa]

OverClocking is not dangerous, ignorance and overzealous/stupidity is dangerous.

Every processor has its own specifications, if you dont overcome that you are fine to go, overcome the manufacturer guide lines and you are in trouble.

All current processors from both AMD and Intel will not sweat with a 10% OC with a stock voltage and cooling.

 

[LOL_Wut_Axel]

First of all I'd like to apologize for starting this firestorm as I was only expressing my personal view about what I do or don't do to the computer that pays my rent. However, as for convincing "home business" people to OC:

1) I assure you that at least 95% and possibly 99% of all home business people think OC is Orange County and wouldn't know how to set up an overclock if you paid them.

2) That same percentage would look upon the prospect of doing that with fear, anxiety, and very possibly panic.

No one is to be blamed for any of this as we all have our own perspectives. As a non gamer non OCer primarily business user, OCing is simply not an option. I'm not going to void my warranty and I'm not even going to take the 0.00000000000000001% chance that I'm going to fry something one nanosecond before it was going to go south by itself anyway. My computer makes me earn a three figure per hour rate when I'm consulting (not full time... I'm not rich unfortunately) and if I want faster performance I'll pay for the box stock version of it and leave it that way. I'm NOT saying that people who OC are wrong but they are in the overwhelming minority (as measured among all installed PCs on the planet) and I can't see any scenario that would change that and increase the current "bare minority" status. Again, that's just IMHO.

CPUs don't really die except if on a very particular occasion a power surge occurs and the PSU doesn't do its job or if you short out the motherboard and it takes the CPU with it. If the CPU dies not because it was OCed (which it won't) they'll replace it.

But from what I'm reading you're just kinda worried about non-issues like something frying. The CPU is only running 5W higher; nothing's gonna fry. If the CPU dies while OCed but not because it was OCed they'll replace it. Buying the "somewhat higher-end" version these days is just paying to get what would be equivalent to a small OC. Notice how the " somewhat higher-end" versions run at higher wattage than the ones with slightly lower clock speeds and the same wattage as if you took the slightly slower one and OCed it to the same clocks. In reality only the extremely high-end parts are higher binned than the lower-end ones. For example, from a Phenom II X4 955 you get an average OC of 4-4.2GHz, while on a Phenom II X4 980 you get an OC of 4.3-4.5GHz. Obviously I'm not arguing over doing heavy OCs like this for a machine you'll use for work; it's just to say that the lower clocked chips are simply there not because of worst binning but to fill the market in lower price points.

I don't know what to say. Perhaps sorry that you're so worried about a non-issue? But anyway, it's your choice. I'm just saying you'll get better efficiency out of a mild OC.

 

[LOL_Wut_Axel]

Do you find this approach works to get people to listen to and understand your viewpoint?

Style versus substance. You fail at making good counterarguments, hence why you'd mention this.

It's like making a 500 word paragraph saying facts and then a moron like you saying "oh, you called X stupid. Your argument is invalid." What are you, 5-years-old? Focus on the facts.


Personal attacks are not tolerated here. This is your only warning. Next time, infractions and a vacation come your way.
Markfw900
Anandtech Moderator

 

[bridito]

OverClocking is not dangerous, ignorance and overzealous/stupidity is dangerous.

I have a client who is a prominent doctor - specialist. He likely doesn't make a million bucks a year but I can assure you he comes close. He has been unable to read the .docx I sent him. I asked him what version of Word he had. He said he didn't have Word at all, he said he had Office (without realizing Word is in there). So I asked him what Office version he had. He said he didn't know and how would he find out? I told him how and he said 2003. I asked him if he had the Compatibility package installed. He said what? I had him go to the Microsoft site. He couldn't find it. I sent him the URL link. He couldn't install it. I finally told him to just go to the nearest Best Buy and purchase Office 2010. He came back with Windows 7 Ultimate.

Now... You want me to convince this guy to OC what exactly?

It's just so easy to assume that everyone on the planet is busily discussing the pros and cons of VCCSA and Vdrop. I have news for all. Most people are just as likely to think that there are dilithium crystals in their PCs as silicon.

 

[bridito]

CPUs don't really die except if on a very particular occasion a power surge occurs and the PSU doesn't do its job or if you short out the motherboard and it takes the CPU with it. If the CPU dies not because it was OCed (which it won't) they'll replace it.

But from what I'm reading you're just kinda worried about non-issues like something frying. The CPU is only running 5W higher; nothing's gonna fry. If the CPU dies while OCed but not because it was OCed they'll replace it. Buying the "somewhat higher-end" version these days is just paying to get what would be equivalent to a small OC. Notice how the " somewhat higher-end" versions run at higher wattage than the ones with slightly lower clock speeds and the same wattage as if you took the slightly slower one and OCed it to the same clocks. In reality only the extremely high-end parts are higher binned than the lower-end ones. For example, from a Phenom II X4 955 you get an average OC of 4-4.2GHz, while on a Phenom II X4 980 you get an OC of 4.3-4.5GHz. Obviously I'm not arguing over doing heavy OCs like this for a machine you'll use for work; it's just to say that the lower clocked chips are simply there not because of worst binning but to fill the market in lower price points.

I don't know what to say. Perhaps sorry that you're so worried about a non-issue? But anyway, it's your choice. I'm just saying you'll get better efficiency out of a mild OC.

There is nothing wrong with homosexuality. It is a lifestyle choice which I fully support. I have homosexual friends who are just wonderful. It's great for the people who love to make love to their own gender, I support them, I will defend them against their detractors, and I will do everything in my power to ensure that they are never discriminated against.

But I aint gonna jump in the sack with a guy.

And I aint gonna OC my business PC.

And all the discussions of the benefit by all the OCers in the world are not going to convince me otherwise. ()

 

[LOL_Wut_Axel]

There is nothing wrong with homosexuality. It is a lifestyle choice which I fully support. I have homosexual friends who are just wonderful. It's great for the people who love to make love to their own gender, I support them, I will defend them against their detractors, and I will do everything in my power to ensure that they are never discriminated against.

But I aint gonna jump in the sack with a guy.

And I aint gonna OC my business PC.

And all the discussions of the benefit by all the OCers in the world are not going to convince me otherwise. ()

Yay for being narrow minded. Alright, off into ignore.

 

[bridito]

Yay for being narrow minded. Alright, off into ignore.

With all due respect and appreciation for your perspective could you please tell me why I have to be pressed into adopting your viewpoint and insulted when I do not? You are making arguments which are no more compelling than those made by my homosexual friends in that I really should "try it". I'm not going to engage in either of these pursuits. Homosexuality because I like chicks ONLY... and OCing because I have not seen any evidence presented by you or the countless other OCers in the well over a decade since I've first engaged in these types of OC vs. No OC debates that it does not shorten service life. If OCers were so convinced that it was a non-issue why has no one ever (to the best of my knowledge) set up this experiment:

Purchase three identical computers

Leave one box stock

Set up one at a MILD OC

Set up one at a KILLER OC but still on air

Set each up at the same day to run Folding (or some other high steady load) 24/7

Keep them running for however many years it will take until they die.

Publish the results

Until then it's just your word that it's safe. Well, I don't know you, and although I respect your opinions, that does not mean that I'm going to clone them for myself.

 

[Phynaz]

Frankly I'm not worried about service life.

As mentioned earlier, silent corruption is the real killer. Bad data in a business? Lights out.

 

[bridito]

Frankly I'm not worried about service life.

As mentioned earlier, silent corruption is the real killer. Bad data in a business? Lights out.

Agreed. Amend that to service life and data corruption.

A test could be set up in the same way with the 3 PCs. Run the heck out of them and then check to see what data has been corrupted.

 

[Cogman]

With all due respect and appreciation for your perspective could you please tell me why I have to be pressed into adopting your viewpoint and insulted when I do not? You are making arguments which are no more compelling than those made by my homosexual friends in that I really should "try it". I'm not going to engage in either of these pursuits. Homosexuality because I like chicks ONLY... and OCing because I have not seen any evidence presented by you or the countless other OCers in the well over a decade since I've first engaged in these types of OC vs. No OC debates that it does not shorten service life. If OCers were so convinced that it was a non-issue why has no one ever (to the best of my knowledge) set up this experiment:

Purchase three identical computers

Leave one box stock

Set up one at a MILD OC

Set up one at a KILLER OC but still on air

Set each up at the same day to run Folding (or some other high steady load) 24/7

Keep them running for however many years it will take until they die.

Publish the results

Until then it's just your word that it's safe. Well, I don't know you, and although I respect your opinions, that does not mean that I'm going to clone them for myself.

You can't draw any scientific conclusions from just 3 machines. In fact, you would need several 1000 machines before you can even think of drawing a conclusion. This is why there are no published results. Nobody is going to spend the millions it takes to verify that a machine being run out of spec has comparable service life.

That being said, there are several known physical phenomena that happen when you overclock. You are putting more stress on 1 or more component (depending on how you overclock) and thus shorting the life of those components.

The fact is, a CPU pushed to its limits will eventually slowly degrade and be unable to maintain its high clock, especially if it is being overvolted. This is a known phenomena that is pretty easy to observe. That being said, it usually takes a couple of months to maybe even a year before the system degrades into an unstable state. By the time it gets to the point of being unstable at stock takes several years (by which time, most OCers have moved on to new stuff).

The service life for a CPU is ridiculously long. I have never heard of a non-overclocked CPU going bad, and it is pretty rare for even an overclocked CPU to go bad. You are looking at 10 years, at least, for your CPU to operate without a hitch unoverclocked and something like 5 years overclocked.

By that time, usually something else in the system has failed (hard drives being the most common points of failure).