Is intel underclocking it's CPUs so they can release higher clock models in the future without any difficulty?

[Hajpoj]

So is this the case, or is there some other reason these chips are underclocked?

 

[myocardia]

Hah, that's a definite possibility. But, Intel cpu's have always been very, very overclockable. I mean, around 2000, people were running 1.6 & 1.7 Ghz P4's @ 2.5-2.7 Ghz. And hey, quite awhile before that, people, myself included, were running 366 Mhz Celerons @ 550 Mhz, and others were running 300 Mhz Celerons @ 450. Then, about 2001 or 2002, alot of people were running 2.4 Ghz P4's @ 3.2-3.5 Ghz. So, it really isn't that new, to tell you the truth.

 

[F1shF4t]

Thats a possibility, why release something faster when u already have an advantage, release it when needed and make more money off it.
Also another reason could be is the power consumption levels. I mean i doubt core 2 duos would consume anywhere close to 65 watts at 3+ ghz

 

[dguy6789]

There is no question that Intel is currently holding back on how fast they make their Core 2 Duos.

 

[SergeC]

They are definitely holding it back - and as someone else already stated - it's to keep the price/preformance crown.

 

[imported_AndyT]

The bulk of all processor sales are in the mid and low range, with the higher clocked and extreme versions being high pofit, but low volume. As time rolls on, the CPU's get refined and normally will run faster. If Intel binned all processors to be their fastest models, the shelves would be full and sales would be low. This is why we get some great overclockers marked as low speed chips. For Intel to be top dog, they have to sell us some bargains that will kill the competition. Good for us!

They without a doubt have much faster chip speeds than they are selling at this time. They will pace the releases to be faster as they go and this means that even the top binned CPU's today might still have tons of overclocking headroom.

 

[Cheex]

I agree. I think Intel will release a faster Extreme processor (maybe something like 266x13 = 3.46GHz). I think they might do that nearer to when they launch Penryn. It will keep the C2Ds alive by further reducing their price and introducing 45nm at the current Extreme and Quad prices.

 

[Idontcare]

Considering that majority of folks report hitting their overclocks by increasing voltage and pimping out their case with a Tuniq 120 HSF I don't think you can make this claim.

If Intel took on more product lifetime headaches necessary to validate chips are higher operating voltages and temperatures then sure they could start overclocking their product line and selling higher speed bins to the masses. Likewise if they opted to bundle their chips with Vapochill LS they could sell really high binned parts.

Is a X6800 an underclocked chip? Not from mwhat I can tell. My QX6700 sure the heck is not an underclocked chip. To use it 24x7 at 3.33GHz I've had to use a vapochill LS to cool it while over-volting to 1.5Vcore.

 

[Furen]

I think Intel is underclocking its CPUs to hit its own TDP targets. Many Core 2s will hit 3.2GHz without too much trouble but their "TDP" is closer to 100W than 65W at that speed. The same thing can be said about the quad-core parts. This is usually the case with Intel. TDPs are what held back Prescott and what held back Yonah, Dothan, etc.

 

[f4phantom2500]

It's due to the fundamental design that they use when designing their processors. I'm sure they design them like this intentionally so that they have a lot of overhead so that like if AMD comes out with a killer processor they can crank up the clock speeds and have a somewhat competitive product.

 

[nyker96]

well i think celeron 300A was very Ocable but netbursts aren't too OCable just due to heat problems. AMD of course X2/A64 both are OCable but that XP and older line are soso. I think it's also depend on supporting chipset stability and component as well. for C2D I think they just trying to release chips at a nice low TDP level that's all. If you OC say above 3 all the C2Ds probably need better than stock HSF.

 

[HurleyBird]

If Intel could safely release faster chips they would (seems like they've learned from the 1GHz P3 mistake. They would be stupid not to, especially with the price war going on. The reason why Intel cannot release a faster product is best described by Scientia on his blog:

You don't understand. Clock tolerance is not the same as thermal tolerance. Chips can be tested for stable clocking but not for thermal tolerance on bulk silicon. The only way to establish thermal tolerance is to run the chip until it experiences thermal cascade failure (which can damage the chip). So, instead of testing to find thermal tolerance they test to find measured power draw (which is equal to TDP) and then rate with margins.

This is common in engineering BTW, you can't test every beam to failure so you test some and establish a normal failure limit and then set your maximum at 67% of that. They basically do the same thing with bulk silicon chips. AMD chips are different because SOI does not experience thermal cascade failure.

 

[Mallet]

Originally posted by: Dark Cupcake
Thats a possibility, why release something faster when u already have an advantage, release it when needed and make more money off it.
Also another reason could be is the power consumption levels. I mean i doubt core 2 duos would consume anywhere close to 65 watts at 3+ ghz

the power consumption point hit the nail on the head. With the netburst architecture they got so much negative publicity about power consumption and over heating that they made this a priority

 

[Accord99]

Originally posted by: HurleyBird
If Intel could safely release faster chips they would (seems like they've learned from the 1GHz P3 mistake. They would be stupid not to, especially with the price war going on. The reason why Intel cannot release a faster product is best described by Scientia on his blog:

Except that the Core 2 Duos have a massive amount of thermal headroom before the built-in throttling mechanisms begin to kick in.

 

[lookin4dlz]

This is a really stupid post, I can't believe the conspiracy theories people come up with.

 

[imported_AndyT]

Originally posted by: lookin4dlz
This is a really stupid post, I can't believe the conspiracy theories people come up with.

Nice to see you have an opinion. That's what forums are for. What do you know about CPU's and binning that we dont?

 

[HurleyBird]

Originally posted by: Accord99

Originally posted by: HurleyBird
If Intel could safely release faster chips they would (seems like they've learned from the 1GHz P3 mistake. They would be stupid not to, especially with the price war going on. The reason why Intel cannot release a faster product is best described by Scientia on his blog:

Except that the Core 2 Duos have a massive amount of thermal headroom before the built-in throttling mechanisms begin to kick in.

Think before you post. If Intel could ship their entire product line 200 or 400 MHz faster don't you think they would? With the price war going on, the argument that Intel is "sitting on" faster products isn't very convincing at all. If it were, instead of cutting prices Intel could release faster products that bring the cost of the rest of the product line down (like they did during the Pentium 4 Northwood ramp), thus saving margins from being trashed. This isn't happening, so there are only two possible explanations: Intel can't safely release faster products (likely), or the people running Intel are stupid (unlikely). If you can think of a third alternative let me know, but I think you should read Scientia's post again: Clock tolerance is not the same as thermal tolerance. Because AMD uses SOI they can clock their chips high enough so that the top end chips can OC no more than 200MHz in stock conditions. Because Intel uses bulk silicon and can't test every chip to the point of thermal cascade failure they need to be a lot more conservative with their clocking. 400 - 600MHz OCs being common at stock for high end products does not mean that Intel can *safely* release higher clocked parts.

 

[Accord99]

Originally posted by: HurleyBird
Think before you post. If Intel could ship their entire product line 200 or 400 MHz faster don't you think they would? With the price war going on, the argument that Intel is "sitting on" faster products isn't very convincing at all. If it were, instead of cutting prices Intel could release faster products that bring the cost of the rest of the product line down (like they did during the Pentium 4 Northwood ramp), thus saving margins from being trashed. This isn't happening, so there are only two possible explanations: Intel can't safely release faster products (likely), or the people running Intel are stupid (unlikely). If you can think of a third alternative let me know, but I think you should read Scientia's post again: Clock tolerance is not the same as thermal tolerance. Because AMD uses SOI they can clock their chips high enough so that the top end chips can OC no more than 200MHz in stock conditions. Because Intel uses bulk silicon and can't test every chip to the point of thermal cascade failure they need to be a lot more conservative with their clocking. 400 - 600MHz OCs being common at stock for high end products does not mean that Intel can *safely* release higher clocked parts.

If it's temperature, why can Intel release a QX6700 which has a significantly higher power consumption? Even spread out over two separate dies, the core temperatures under load is higher than on the X6800.

http://www.hwupgrade.com/articles/cpu/1...he-first-quad-core-amd-platform_6.html

With a decisive performance advantage, what difference does it make if Intel releases a new speed grade. Existing grades go down one price level but most people will still buy the products from $180 to $300. And really, has the price war done much to Core 2 Duo prices since launch?

 

[aka1nas]

Power consumption is not the same as Die temperature. They can get away with the higher TDP in the Quad-cores as they are pretty much a show-off part on the consumer side at this point. Both Intel and AMD are trying stay within their TDP designations as it's something they can market on still now that it's getting hard to scale clockspeed.

 

[HurleyBird]

Originally posted by: Accord99
With a decisive performance advantage, what difference does it make if Intel releases a new speed grade. Existing grades go down one price level but most people will still buy the products from $180 to $300. And really, has the price war done much to Core 2 Duo prices since launch?

Prices on the Core 2's started out at an unusually low level for a next-gen part. If Intel would have launched with some more higher end parts with higher clocks and higher prices it would both help their margins and pressure AMD, or, if there's as much headroom as you think there is they could have bumped the entire product line by 400MHz and absolutely crushed AMDs offerings at every price point. Right now AMDs lower priced offerings are still somewhat competitive with lower priced C2Ds (not counting overclocking) so it makes no sense for Intel not to bump frequencies if they could.

 

[JEDIYoda]

Originally posted by: lookin4dlz
This is a really stupid post, I can't believe the conspiracy theories people come up with.

exactly...lol.....could be an ex amd fanboy..lol

 

[postmortemIA]

TDP definitely matters for big OEMs. Obviously chip that produces less heat results in cheaper system (smaller heatsinks, less fans, more design options). I think they know it can run on higher speed, but the drawbacks are big at the moment.

 

[HurleyBird]

Originally posted by: JEDIYoda

Originally posted by: lookin4dlz
This is a really stupid post, I can't believe the conspiracy theories people come up with.

exactly...lol.....could be an ex amd fanboy..lol

Right... I'm assuming that's an insult hurled at me. If so, I've already detailed why it would be stupid of Intel to underclock their entire product line, so try to actually refute that instead of making garbage posts.

 

[Skotty]

I believe the C2Ds are currently clocked lower than they could be. They are the first of a new generation that far exceeds the previous, but I think Intel likes to have more incremental improvements. While Intel spends the next however many months or years working on the next leap of architecture, their current line of processors can continue to "advance" by slowly clocking them up to what they are really capable of. I'm sure the reality is a little more complicated than that, but in general, that's how I think it works.

 

[HurleyBird]

But they *haven't* been slowly clocking up and we *haven't* even seen a substantial increase in over clocking headroom on the top end parts.