• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Intel locks all overclocking features on 915/925 ??

D

Double Trouble

Elite Member
According to Toms Hardware, Intel has decided to lock down all the overclocking features of the new 915x/925x chipsets. I don't trust THW as a source for much of anything, but if that's true it would certainly be a good reason for enthusiasts to stay away from those chipsets.

After seeing the .....ahem.... somewhat less than exciting results of AT testing on the new chipsets, this would be another blow.

<edit> Anand and his crew have done some testing and added their own take to keep us informed so we don't have to rely on rumors. Thanks!
 
Well, you have never been able to overclock an intel board. It doesn't seem to have bothered anyone yet.
 
You missed the point of his post. He's not saying that Intel won't allow o/cing on their boards, he means they are locking out o/c features in the chipset itself. IE, it won't matter what board you buy, you won't be able to overclock.
 
While this sucks for ocers, it doesn't seem like SiS will be following suit, so there will be excellent alternatives that are just as stable and probably as good as or better performing in comparison to these Intel chipsets. Plus they'll overclock more than even the 20% max rumored with mobo maker hacks the Intel chipsets will allow.
 
Well, some mfg's have already gotten around it to the tune of 15-20% overclocks.

I am sure it will all come out in the wash, like it usually does with these panicky early stories. 😀
 
What's Intel afraid of? Oh yeah, that the smart consumers will just buy the cheaper chip and o/c it past one of their overpriced high end chips. We all know that Intel is basically planning to make one high end chip and clock some lower than others just to sell them at a cheaper price *cough* 2.4C *cough*.
 
Originally posted by: Arsynic
We all know that Intel is basically planning to make one high end chip and clock some lower than others just to sell them at a cheaper price *cough* 2.4C *cough*.
Click to expand...
That is what all chip manufacturers have always done. You try to make the chips the best you can. However, since the manufacturing process isn't uniform, some chips cannot handle the speed that others do. Thus the company tests them to see what speed they can safely run, and marks them accordingly. That is just standard business, always has been and always will be.

Sometimes demand outpaces supply (for example you do too good of a job at making the high end chips, and you just mark a high end chip with a lower price to meet the demand). But is that really a problem? Doing too good of a manufacturing job?

It isn't limited to Intel.
 
AMD locked the A64 754's. So a 3400 is just an overclocked 3200 for a huge chunk of more change.

But I wouldnt want to over clock the 925x anyway. Not with such a high FSB as 1.4Ghz.
 
Originally posted by: LTC8K6
Well, some mfg's have already gotten around it to the tune of 15-20% overclocks.

I am sure it will all come out in the wash, like it usually does with these panicky early stories. 😀
Click to expand...

See the detail at Tom's Hardware.

The real question is how much margin in thermal stability will Intel allow in their corresponding CPU. Up to P4, it is easy to get 25% and ocassionally, 50%. Celeron 300 and the Lockwood P4-1.6A are two of the 50% overclocker that come to mind. More importantly, they are destruction proof with the built-in thermal protection circuitry.

I stopped buying AMD after the 200MHz CPU purchase (years ago) because I cannot get any margin out of them.

I am amazed that Asus and Gigabyte devised the bypass scheme even before the first product was released.
 
I stopped buying AMD after the 200MHz CPU purchase (years ago) because I cannot get any margin out of them.
Click to expand...

This without ignoring that the AMD can do more instructions per clock than the Intel chips?
 
Well I'd wait for mobos to show up before coming to a conclusion. I ecall similar discussions in the past concerning AMD locking Multipliers and other changes. They turned out to be unfounded.
 
Originally posted by: Arsynic
What's Intel afraid of? Oh yeah, that the smart consumers will just buy the cheaper chip and o/c it past one of their overpriced high end chips. We all know that Intel is basically planning to make one high end chip and clock some lower than others just to sell them at a cheaper price *cough* 2.4C *cough*.
Click to expand...

no, they're more concerned about middlemen remarking their chips to higher speeds and then selling these remarked chips for higher margins.

will this clock locking prevent underclocking as well?
 
Originally posted by: Arsynic
What's Intel afraid of? Oh yeah, that the smart consumers will just buy the cheaper chip and o/c it past one of their overpriced high end chips. We all know that Intel is basically planning to make one high end chip and clock some lower than others just to sell them at a cheaper price *cough* 2.4C *cough*.
Click to expand...


I love blanket idiotic statements like this. Sigh.

1) Intel (Or AMD for that matter) couldn't care less what the "smart" consumer does after buying their product. They are happy the consumer bought it in the first place. They ARE concerned about what some of the less reputable manufactures do when they buy a slower chip and then over clock it and sell it as a fast chip for more money. Why do they care? Because the manufacturer ends up taking the hit when the thing keeps locking up and wont run right. EIther a true financial hit or just a reputation hit with that customer who doesn't even KNOW they have an OCed chip. That is where multiplier locking came from and that is were this latest system is coming from. It sucks to be us, but they are NOT targeting the computer geeks of the world.

2) Intel has pretty much ALWAYS made one chip (in a given line for the most part) and sold the speed bins at different prices. The fact that you can get 95% of the die to work fairly consistently is amazing. The fact that they are not all able to run at the same speed is completely understandable. Supply vs. demand leads to the price spread between speed bins. (and yes, marketing DOES get to play too. Unfortunately...) Try this split. You can manufacture 100k die. Out of these you get 10k 3.4 ghz parts, 20k 3.2 ghz parts, and 50k 3.0 ghz parts and the rest are below this speed. So this 80k die is what you can sell in this time period. Then marketing gets to step in. They decide what price to put on the 3.4 ghz parts to sell them all in the alloted time. The same for the 3.2 and 3.0 parts. This is the normal approach. But what happens when your ability to manufactury fast parts out paces the demand? Then you are in this senario: You manufacture 100k die. Of those, 50k are 3.4 ghz parts, 25k are 3.2 parts and 20 k are 3.0 parts. Now you have 95k parts to sell this period (yields are up!). Marketing steps in and gives you the "bad" news: At the lowest price they are willing to sell the top of the line part for, there is only a demand for 30k parts. That gives you 20k too many 3.4 ghz parts. But they could sure use another 15k 3.2 and 5k 3.0 parts... So they get a lower multiplier and they get sold as the slower (and cheaper) 3.2 and 3.0 parts. Even thought they fully test out as 3.4 parts. And this almost always happens when they don't quite have enough yield on the next speed bin part (3.6 ghz in this example) to market them. Once you improve bin yields a bit more, then you end up in the 15k 3.6, 30k 3.4, 40k 3.2 and 15k 3.0 space and life is good in marketing once more...

In short, there is a whole lot more going on than some evil plan to keep us all using the slowest, highest price part possible.
 
Originally posted by: thermite88
Originally posted by: LTC8K6
Well, some mfg's have already gotten around it to the tune of 15-20% overclocks.

I am sure it will all come out in the wash, like it usually does with these panicky early stories. 😀
Click to expand...

See the detail at Tom's Hardware.

The real question is how much margin in thermal stability will Intel allow in their corresponding CPU. Up to P4, it is easy to get 25% and ocassionally, 50%. Celeron 300 and the Lockwood P4-1.6A are two of the 50% overclocker that come to mind. More importantly, they are destruction proof with the built-in thermal protection circuitry.

I stopped buying AMD after the 200MHz CPU purchase (years ago) because I cannot get any margin out of them.

I am amazed that Asus and Gigabyte devised the bypass scheme even before the first product was released.
Click to expand...
"Destruction-proof?" I guess you haven't heard of SNDS 😉

But yeah, wait for shipping products. It could be Tom's Hardware spreading a little FUD on their sandwich to attract more ad views, or a mindgame like the "sample" Raptors with the dismal performance, followed by the shipping ones that turned out to be better. Where people would've been upset at a 20% OC, now maybe they'll be gloating about how they thwarted the mighty Intel. Eh? 😉
 
Thanks to Anand and crew for stepping in and shedding some light on the issue!

For years the best 'value' has been AMD, but if you wanted the best performance, you had to go Intel. Then, AMD caught up in terms of performance, but chipsets (mostly VIA) still pushed many people to opt for a 'safer' and better tested Intel chipset/cpu.

Now we've reached the point where the AMD chipsets (nforce, via) are every bit as good as the Intel ones, and the AMD procesors are every bit the equal of the Intel ones. From my perspective, adding a lock feature like this is not going to hurt Intel in the grand scheme (enthusiasts are just a fraction of the buyers). However, as an enthusiast I see this as just another nudge in the direction of getting AMD instead of Intel on my next system.....

In fact, the new chipsets, the new processor and even the BTX standard from Intel are *all* very dissapointing. Yeah, there's some solid new technology that will be part of the standards for years to come, but there's just nothing really compelling there.

<end editorial> 🙂
 
Originally posted by: dullard
Originally posted by: Arsynic
We all know that Intel is basically planning to make one high end chip and clock some lower than others just to sell them at a cheaper price *cough* 2.4C *cough*.
Click to expand...
That is what all chip manufacturers have always done. You try to make the chips the best you can. However, since the manufacturing process isn't uniform, some chips cannot handle the speed that others do. Thus the company tests them to see what speed they can safely run, and marks them accordingly. That is just standard business, always has been and always will be.

Sometimes demand outpaces supply (for example you do too good of a job at making the high end chips, and you just mark a high end chip with a lower price to meet the demand). But is that really a problem? Doing too good of a manufacturing job?

It isn't limited to Intel.
Click to expand...
Dude, I'm not an AMD fanboy, I've built Pentium and Athlon based systems, so there's no favoritism. I'm not blindly bashing a company. I know AMD does it too, they are just trying to make the extra cash.
 
Dude, I'm not an AMD fanboy, I've built Pentium and Athlon based systems, so there's no favoritism. I'm not blindly bashing a company. I know AMD does it too, they are just trying to make the extra cash.
Click to expand...

He wasn't calling you a fan boy. He was just simply stating how the manufacture of CPU's work for all manufacturers. No need to get defensive.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top