How do you acquire your good overclocking CPU?

Page 3 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.

Gillbot

Lifer
Jan 11, 2001
28,830
17
81
You won't get through to anyone. The original poster said this-

"or do you return CPU's until you get a good one?"


Do you RMA 10 working cpu's till you get one that reaches your overclocking expectations? I am not ragging on the thread starter because he didn't say he ever did that type of thing. That type of behaviour is unethical and I can't believe anyone would condone it. The end result is higher prices for everyone.

Exactly. Everyone else is ruining it by believing the overclock hype and they justify their shady actions by twisting and contorting reality.

Point being, you are paying for a chip that does exactly what it was advertized and intended to do at STOCK settings, anything above that is gravy. Be happy with that extra or take the hit yourself because you got suckered into believing you could hit 43GHz with a PPro 200.
 

Hulk

Diamond Member
Oct 9, 1999
5,118
3,660
136
You won't get through to anyone. The original poster said this-

"or do you return CPU's until you get a good one?"


Do you RMA 10 working cpu's till you get one that reaches your overclocking expectations? I am not ragging on the thread starter because he didn't say he ever did that type of thing. That type of behaviour is unethical and I can't believe anyone would condone it. The end result is higher prices for everyone.


No I have never done that and without getting into ethics, it's just not something I'm comfortable doing. I started the thread because I see so many really good Sandy/Ivy/Haswell overclocks and was curious as to how people find these golden chips.
 

Hulk

Diamond Member
Oct 9, 1999
5,118
3,660
136
Would you pay an extra $20 for one with known good overclockability?


Isn't that the point of the "K" series CPU's?

Who knows how Intel is binning the K's? It's a always been a "behind the curtain" type of thing.
 

rtsurfer

Senior member
Oct 14, 2013
733
15
76
Isn't that the point of the "K" series CPU's?

Who knows how Intel is binning the K's? It's a always been a "behind the curtain" type of thing.

Exactly.

This is why i like AMD's 9000 series idea. There CPU's IPC might suck but they are doing a great thing by providing parts that can OC really well.

Intel on the other hand sends highly binned parts to reviewers and then everyone else is just left for luck.
 

PPB

Golden Member
Jul 5, 2013
1,118
168
106
Exactly.

This is why i like AMD's 9000 series idea. There CPU's IPC might suck but they are doing a great thing by providing parts that can OC really well.

Intel on the other hand sends highly binned parts to reviewers and then everyone else is just left for luck.

Well, according to intel, the k parts are already binned. I assume not so highly binned like top laptop, skus, but binned nonetheless
 

SiliconWars

Platinum Member
Dec 29, 2012
2,346
0
0
Do you RMA 10 working cpu's till you get one that reaches your overclocking expectations?

Nobody mentioned RMA's. You can't RMA a chip unless it has a fault, and if you do they will test it before accepting the RMA.

The end result is higher prices for everyone.
Nonsense. The cost of the actual silicon is generally less than $10. The percentage of overclockers is tiny, the percentage of those who want extreme OC's is even smaller, the percentage of those who return chips because of not reaching the OC they wanted is miniscule and the overall cost doesn't even register on any known scale, certainly not to $billion corporations so to blame this for higher prices is incredibly disingenuous.
 

rtsurfer

Senior member
Oct 14, 2013
733
15
76
Well, according to intel, the k parts are already binned. I assume not so highly binned like top laptop, skus, but binned nonetheless

My 4770K barely making to 4.5 tends to disagree.

While all the reviewers like Anand, Linus and Paul from Newegg hit 4.8.

Intel shouldnt be sending their top bins to reviewers and should let it be as random as you or me buying the CPU anywhere.
 

Gillbot

Lifer
Jan 11, 2001
28,830
17
81
My 4770K barely making to 4.5 tends to disagree.

While all the reviewers like Anand, Linus and Paul from Newegg hit 4.8.

Intel shouldnt be sending their top bins to reviewers and should let it be as random as you or me buying the CPU anywhere.

I highly doubt this is the case. Odds are more likey that the "top reviewers" have better overall parts like top tier motherboards, great ram and stable PSU's that allows them to squeak out that extra performance whereas most people try to overclock with whatever junk they have laying around and complain when it doesn't work.

Not to mention the handful that boast about hitting certain numbers are FAR FAR fewer than the average joe who hits a decent modest overclock but doesn't post or brag about it.
 

blackened23

Diamond Member
Jul 26, 2011
8,548
2
0
Exactly.

This is why i like AMD's 9000 series idea. There CPU's IPC might suck but they are doing a great thing by providing parts that can OC really well.

Bad CPU is still bad even if you overclock. Whereas another 100mhz won't matter on the 4770k. You would be nitpicking over a minor 1-3% increase which is ridiculous.

On the other hand, you can't overcome a horrible CPU such as the FX 9000 series even if you OC.
 
Last edited:

TerryMathews

Lifer
Oct 9, 1999
11,464
2
0
I highly doubt this is the case. Odds are more likey that the "top reviewers" have better overall parts like top tier motherboards, great ram and stable PSU's that allows them to squeak out that extra performance whereas most people try to overclock with whatever junk they have laying around and complain when it doesn't work.

Not to mention the handful that boast about hitting certain numbers are FAR FAR fewer than the average joe who hits a decent modest overclock but doesn't post or brag about it.

To be fair though, "golden" in the case of Haswell could be as simple as a hand-assembled CPU + IHS so that there isn't the same gap like there is in most of the production units.
 

Gillbot

Lifer
Jan 11, 2001
28,830
17
81
To be fair though, "golden" in the case of Haswell could be as simple as a hand-assembled CPU + IHS so that there isn't the same gap like there is in most of the production units.

Regardless of the architecture, we go through this debate with every "hot" chip. It was the Celeron 300A that would run 450, Q6600 that would go over 3.6GHz, etc.....

No matter what chip it involves at the time, fact is MOST people don't hit that golden number and those who go through chip after chip to get there are morally questionable if they keep returning them. Fact is, you bought a chip and it works 100% at it's rated specs. PERIOD.
 

Blamblooga

Junior Member
Dec 25, 2013
12
0
0
Regardless of the architecture, we go through this debate with every "hot" chip. It was the Celeron 300A that would run 450, Q6600 that would go over 3.6GHz, etc.....

No matter what chip it involves at the time, fact is MOST people don't hit that golden number and those who go through chip after chip to get there are morally questionable if they keep returning them. Fact is, you bought a chip and it works 100% at it's rated specs. PERIOD.

Stop using the words Moral and Ethic -- You don't know their definitions. :rolleyes:


Vendors know people RMA these chips for these reasons. Don't you think if they were losing a shit load of money, they would have some kind of procedure implemented to check RMA'd CPUs?


Bottom line:

Until you've worked in a RMA department and actually have seen concrete statistics, stop spreading FUD.


It's not immoral, or unethical to RMA a chip that is 100% functional. And one could even argue that people might not try to get 'golden' chips if Intel would stop being cheap with their shitty TIM under their IHS'.

The bad language got you the infraction, and yes it is unethical to RMA a perfectly working chip.
Markfw900
Anandtech Moderator
 
Last edited by a moderator:

TerryMathews

Lifer
Oct 9, 1999
11,464
2
0
Regardless of the architecture, we go through this debate with every "hot" chip. It was the Celeron 300A that would run 450, Q6600 that would go over 3.6GHz, etc.....

No matter what chip it involves at the time, fact is MOST people don't hit that golden number and those who go through chip after chip to get there are morally questionable if they keep returning them. Fact is, you bought a chip and it works 100% at it's rated specs. PERIOD.

What in the world does that have to do with my post? If youre just going to ramble and not address what I said, decorum dictates that you not quote me.

Also, just to rattle your cage: you would agree it is defective if it throttles during IBT at stock, yes?
 

Gillbot

Lifer
Jan 11, 2001
28,830
17
81
Stop using the words Moral and Ethic -- You don't know their definitions. :rolleyes:


Vendors know people RMA these chips for these reasons. Don't you think if they were losing a shit load of money, they would have some kind of procedure implemented to check RMA'd CPUs?


Bottom line:

Until you've worked in a RMA department and actually have seen concrete statistics, stop spreading FUD.


It's not immoral, or unethical to RMA a chip that is 100% functional. And one could even argue that people might not try to get 'golden' chips if Intel would stop being cheap with their shitty TIM under their IHS'.
:rolleyes:

They can assemble them with bubble gum or string and as long as they operate as advertized, you have nothing to complain about nor a valid reason to return it. Just because you don't get more than you pay for is not justification to return it.
What in the world does that have to do with my post? If youre just going to ramble and not address what I said, decorum dictates that you not quote me.

Also, just to rattle your cage: you would agree it is defective if it throttles during IBT at stock, yes?
It has everything to do with your post. If it works as it was intended at stock settings, there is nothing wrong with it. You can speculate all you want about the rest. I stand behind my statements and I can't wait till Karma returns the favor. As I said earlier, I guarantee that most of the people trying to justify their returns to retailers would NEVER accept a return from a person to person sale in FS/FT just because the buyer was unhappy with their purchase because they couldn't overclock it as far as they had hoped. I guarantee that the first words out of their mouth is going to be "It worked exactly as described at stock settings when I sent it" and they wouldn't take it back.
 

TerryMathews

Lifer
Oct 9, 1999
11,464
2
0
It has everything to do with your post. If it works as it was intended at stock settings, there is nothing wrong with it. You can speculate all you want about the rest. I stand behind my statements and I can't wait till Karma returns the favor. As I said earlier, I guarantee that most of the people trying to justify their returns to retailers would NEVER accept a return from a person to person sale in FS/FT just because the buyer was unhappy with their purchase because they couldn't overclock it as far as they had hoped. I guarantee that the first words out of their mouth is going to be "It worked exactly as described at stock settings when I sent it" and they wouldn't take it back.

It had nothing to do with my post and you know it.

Now you're moving your own goalposts again and bringing in private party sales. It's practically impossible to discuss anything with you, your argument flutters around more than a biplane in War Thunder.

I think you need to step back and look at why you're so personally invested in this topic. I can assure you, to Intel or Amazon or Newegg "it's just business".
 

Gillbot

Lifer
Jan 11, 2001
28,830
17
81
It had nothing to do with my post and you know it.

Now you're moving your own goalposts again and bringing in private party sales. It's practically impossible to discuss anything with you, your argument flutters around more than a biplane in War Thunder.

I think you need to step back and look at why you're so personally invested in this topic. I can assure you, to Intel or Amazon or Newegg "it's just business".

I disagree, we are ALL invested in it because in the end it does nothing but drive up prices. Call it what you want, my "goalposts" have not moved. I do know one thing, I'll be sure to watch what you have for sale since you have no issue accepting returns on items if i'm dissatisfied with the overclocks on the items you sell.

EDIT: My point being, you are trying to place the blame on the manufacturer or reseller and are trying to reason it as being acceptable. To me, it is NOT acceptable because they haven't promised you anything extra other than exactly what you've ordered. It's the same way YOU operate with your personal for sale threads. Why should retailers be forced to accept a shitty overclocker in return because you are unhappy? It did what it was supposed to do, it worked as it was intended.
 
Last edited:

SiliconWars

Platinum Member
Dec 29, 2012
2,346
0
0
Drive up prices, seriously? Tell us the maths behind it so we can see just how much this is costing the industry, because clearly you don't believe what I've told you twice.
 

Gillbot

Lifer
Jan 11, 2001
28,830
17
81
Drive up prices, seriously? Tell us the maths behind it so we can see just how much this is costing the industry, because clearly you don't believe what I've told you twice.

If you can't see the business case, I can't help you.

But to make it simple, if a store accepts returns and these people are abusing it returning chip after chip until they get a good one, how is that NOT going to drive up pricing? The retailer has to send another chip which shipping isn't free. They have to either return to the manufacturer or re-test the returned chip, which isn't free. They have to possibly resell the returned chip but they can't sell it as new or at a new price, so their profit margin is smaller or eliminated.

Since you insist "The cost of the actual silicon is generally less than $10" then what pays for all the fab equipment, R&D, programming and research? I guess that's all free too.

Yeah, I don't get it. :rolleyes: Justify it all you want. The world revolves around profit and if you can't see or accept that you are fooling yourself. There's a reason you want to get more for your money but no, you want even MORE. We get it, you can use whatever reasoning you want to justify your actions.
 

SiliconWars

Platinum Member
Dec 29, 2012
2,346
0
0
If you can't see the business case, I can't help you.

But to make it simple, if a store accepts returns and these people are abusing it returning chip after chip until they get a good one, how is that NOT going to drive up pricing?

Who says they are "abusing it"? If you get off your high horse and see that any person doing this is within their rights because of the retailers policy then that's your problem.

The retailer has to send another chip which shipping isn't free.
Since when? I've returned a chip that wasn't what I expected but I didn't get another one sent to me, I got my money back instead. So you can say it cost them the initial shipping and it cost me the shipping as well. At least the shipping companies win out? Oh and the retailers too who get exceptional shipping rates overall because of mass shipping. The only person who really lost out was myself, as I had to pay full shipping rates back.

Of the thousands, or hundreds of thousands of chips this retailer sends out every year, how many do you think are being sent back because of not being good overclockers (or in my case because it didn't unlock to a quad)? Enough to raise their prices? o_O

They have to either return to the manufacturer or re-test the returned chip, which isn't free. They have to possibly resell the returned chip but they can't sell it as new or at a new price, so their profit margin is smaller or eliminated.
Yes and if this happens with one in every 10,000 chips they sell, how much do you think it's costing them or anyone? Enough to drive up prices? o_O

Since you insist "The cost of the actual silicon is generally less than $10" then what pays for all the fab equipment, R&D, programming and research? I guess that's all free too.
Yes amazingly enough these costs are diluted over hundreds of millions of chips. How much do you think a few thousand returns of this nature are costing them? Enough to raise the price of CPU's by more than a cent?

Yeah, I don't get it. :rolleyes: Justify it all you want. The world revolves around profit and if you can't see or accept that you are fooling yourself. There's a reason you want to get more for your money but no, you want even MORE. We get it, you can use whatever reasoning you want to justify your actions.
I can tell you don't get it - you clearly have zero concept of the numbers involved here and have never worked in retail either. The $billions made by chip companies compared to the $thousands "lost" for this doesn't even register with them. At the retail level the cost is still tiny - basically it's only shipping fees which is costing them peanuts anyway.

Some of them might not even repay the delivery cost of sending it back (most here in the UK don't), in which case they almost certainly made a profit on you anyway because whatever they charged you for the initial delivery would have been a lot less than what it cost them. You think it costs Newegg $8 to ship an item to you?
 
Last edited:

Gillbot

Lifer
Jan 11, 2001
28,830
17
81
I don't get it? SUUUUUUUUURRRRRREEEEEEEEEEEE.....

Continue to justify it however you need to so you can sleep at night. We'll have to agree to disagree here.
 

YBS1

Golden Member
May 14, 2000
1,945
129
106
This. You only hear about those 50+ % overclocks because those who achieve that will let the world know they did it. People who only get a 10 - 20% OC won't post about it on overclocking forums.
This is probably 50% of it. Like another post mentioned, 25% of it is probably choosing to use top notch components rather than bargain basement to mid-range stuff. Finally, experience and familiarity with your settings is probably the last 25% of it. That being said, I've had a pretty good go of it with my last two chips, 5+ GHz out of my 3930K and just north of 4.4Ghz on my W3520, so I've become a bit spoiled. Which means if my 4770K (assuming I end up with that as I'm still waiting on the rumored Silverstone 550 watt SFX PSU to become available before buying the rest of my parts for my M1 NCase mini-itx build) fails to get 4.7GHz+ stable I'll probably throw a $h1*fit.
 
Last edited:

TerryMathews

Lifer
Oct 9, 1999
11,464
2
0
I disagree, we are ALL invested in it because in the end it does nothing but drive up prices. Call it what you want, my "goalposts" have not moved. I do know one thing, I'll be sure to watch what you have for sale since you have no issue accepting returns on items if i'm dissatisfied with the overclocks on the items you sell.

EDIT: My point being, you are trying to place the blame on the manufacturer or reseller and are trying to reason it as being acceptable. To me, it is NOT acceptable because they haven't promised you anything extra other than exactly what you've ordered. It's the same way YOU operate with your personal for sale threads. Why should retailers be forced to accept a shitty overclocker in return because you are unhappy? It did what it was supposed to do, it worked as it was intended.

Funny you mention that, I sold a guaranteed overclocked rig not too long ago - SR-2 with 2xL5639s. And I would have gladly taken it back if it didn't meet the overclock I laid out. I believe its folding for TeAM, but I am not sure. I guess we could ask my buyer to comment. The bottom line was I set a return policy like Amazon, and I would have followed it just like Amazon. Not complicated.

Attack attack attack. Again, why so personal?
 

Gillbot

Lifer
Jan 11, 2001
28,830
17
81
Funny you mention that, I sold a guaranteed overclocked rig not too long ago - SR-2 with 2xL5639s. And I would have gladly taken it back if it didn't meet the overclock I laid out. I believe its folding for TeAM, but I am not sure. I guess we could ask my buyer to comment. The bottom line was I set a return policy like Amazon, and I would have followed it just like Amazon. Not complicated.

Attack attack attack. Again, why so personal?

I've not seen newegg nor amazon sell a chip with an overclock guarantee so your example doesn't apply. Who's moving the goal posts now?
 

rtsurfer

Senior member
Oct 14, 2013
733
15
76
I highly doubt this is the case. Odds are more likey that the "top reviewers" have better overall parts like top tier motherboards, great ram and stable PSU's that allows them to squeak out that extra performance whereas most people try to overclock with whatever junk they have laying around and complain when it doesn't work.
.

So I am running junk components..??
That is your pathetic excuse.
I am running what can be considered pretty good components but i will not waste my time mentioning them as you are so great to draw conclusions without knowing what you are talking about.

:rolleyes:

They can assemble them with bubble gum or string and as long as they operate as advertized, you have nothing to complain about nor a valid reason to return it. Just because you don't get more than you pay for is not justification to return it.
And what exactly is advertised..?
That is the problem. Intel says that here you have a CPU with an unlocked multiplier and that is it. They wont even take time to properly Bin those parts.
Which they should as they are advertised as Overclockable parts and charge a premium for it.
So either Intel should stop bubble gum under their IHS or stop advertising them as overclockable parts.