Intel's Response For Q3 '07: Yorkfield

[Carfax]

This is what the K8L will face, according to VR-Zone:


Yorkfield Extreme Edition based on the 45nm Penry core architecture will meet heads-on with AMD Altair based on the 65nm K8L core in Q3 2007. Both are quad core offerings but AMD touted theirs to be a native quad-core with dedicated 512KB L2 cache for each core and an additional 2MB L3 cache for the 4 cores to share. Intel has different plans for Yorkfield. Instead of sharing L2 cache for all the cores, the L2 cache size is further increased. Due to its advanced 45nm process technology, Yorkfield XE is able to pack a total of 12MB L2 cache (2 x 6MB L2) and still achieving a much smaller die size. The clock speed is significantly higher at 3.46-3.74GHz too.

And Intel can pull it off too.. Their 45nm process will be top notch, and the Core architecture doesn't consume much power due to it's incredible efficiency.. It will also have SSE4, the most sigificant extension to Intel's SIMD since SSE2..

I hope the K8L is a badass mofo, because it will be facing an uphill battle next year..

Interesting times ahead folks....interesting times indeed!

http://resources.vr-zone.com//newspics/Oct06/04/Yorkfield.GIF

Source (http://vr-zone.com/?i=4109)

[Carfax]

I hope the game developers get cracking on making new engines that can handle these quad core monsters.

With all of this power, it would be a shame for any of it to go to waste.

Lets hope Alan Wake is the first of many.

[[XC] hipno650]

:slobber: :slobber: :slobber: :slobber: i wonder how this thing would oc:slobber: :slobber: :slobber: i am abit dissapointed tho seeing that it won't share the l2:( because if it did then:eek: :eek: :eek: :eek: :eek: :eek: i might buy conroe and wait for a shared l2 version of this:slobber:

[nn_step]

yawn, more like a pocket monster...
basic math is Increase FP performance and add Cache liberally...
hmm almost expected.. Funny fact that most people miss is that K8 still beats Conroe in interger performance...

[Carfax]

yawn, more like a pocket monster...
basic math is Increase FP performance and add Cache liberally...
hmm almost expected.. Funny fact that most people miss is that K8 still beats Conroe in interger performance...

Are you kidding? Conroe slaughters the K8 in integer performance!

The K8L might be better than Yorkfield clock for clock in FP, but due to Yorkfield's clockspeed advantage, I think Yorkfield may have the upper hand overall.


i am abit dissapointed tho seeing that it won't share the l2

The cache is still shared, just not between all four cores. The reason is probably because it's difficult to get four cores to share cache effectively at the same time.

[Poodle]

Seems like an act of desperation to me. Intel must know that the Kenty sucks badly (Hallo Prescott!) We are seeing the Pentium D rutin again on the new quads from Intel. The non native thing seems to have a bad effect on power consumption. When Intel adds cores the heat multplies with the number of cores added. When Amd adds cores it stays in the same Tdp value as before the core addition.

And the k8L has some new advanced power feats. that will increase performance/w even more. :clap:


And then Intel have that fsb 1333 to worry about (it sucks). Amd will win easy at least on the server plattform, due to memory and bandwidth performance.

[accord99]

Seems like an act of desperation to me. Intel must know that the Kenty sucks badly (Hallo Prescott!) We are seeing the Pentium D rutin again on the new quads from Intel. The non native thing seems to have a bad effect on power consumption. When Intel adds cores the heat multplies with the number of cores added. When Amd adds cores it stays in the same Tdp value as before the core addition.
That's because Intel's TDP right now are much lower than AMD's. Even with the doubling, Kentsfield's power consumption is roughly the same as a FX-62.

[.sentinel]

This is what the K8L will face, according to VR-Zone:



And Intel can pull it off too.. Their 45nm process will be top notch, and the Core architecture doesn't consume much power due to it's incredible efficiency.. It will also have SSE4, the most sigificant extension to Intel's SIMD since SSE2..

I hope the K8L is a badass mofo, because it will be facing an uphill battle next year..

Interesting times ahead folks....interesting times indeed!

Source (http://vr-zone.com/?i=4109)
You are a walking Intel advertisement. SSE4 does nothing and at 3.74 with a large heatdump I would like to see that on their stock cooler.

[nn_step]

Are you kidding? Conroe slaughters the K8 in integer performance!

The K8L might be better than Yorkfield clock for clock in FP, but due to Yorkfield's clockspeed advantage, I think Yorkfield may have the upper hand overall.

funny considering that K8 beats conroe at the same clock speed in
SM Cryptography ( 8.2% faster) and SM STREAM (4.3% faster) which If I remember correctly are both mainly integer programs that use lots and lots of memory

[accord99]

funny considering that K8 beats conroe at the same clock speed in
SM Cryptography ( 8.2% faster) and SM STREAM (4.3% faster) which If I remember correctly are both mainly integer programs that use lots and lots of memory
Funny considering a C2E scores 18.5 on the single-threaded SpecInt2006 benchmark while a 3GHz Opteron only scores 13.3. Cryptography is dependent on a few instructions and Stream is a memory test.

[adamant415]

I do not see yorkfield as an act of desperation. AMD is the company with their backs against the wall needing to prove something. K8L is getting delayed, and it will be interesting to see what makes it to the market first. Draft N wireless or K8L. Intel has stepped up their game and looks like they will be following their plans for a new architecture every two years.

AMD needs to step it up or they will get steam rolled. AMD has woken a sleeping giant. I'm not a fan of Intel or AMD, I am a fan of performance. If K8L is the best thing since sliced bread when it comes out I will buy it. I just don't think they will be able to keep up like some of the fanboys believe.

[nn_step]

Funny considering a C2E scores 18.5 on the single-threaded SpecInt2006 benchmark while a 3GHz Opteron only scores 13.3. Cryptography is dependent on a few instructions and Stream is a memory test.
rather unique since techinically conroe should be able to do 4 integer operations per cycle while AMD can only do 3...
Funny, then again conroe has 1 complex integer unit and 3 simple and AMD just has 3 complex

[Kanavit]

too bad , cache dont' mean much for AMD architecture.

[Carfax]

You are a walking Intel advertisement. SSE4 does nothing and at 3.74 with a large heatdump I would like to see that on their stock cooler.

Obviously you've never seen nor heard anything about SSE4.. I think I'll just let you remain ignorant and laugh at you :banana:

Also, if this chip is real, it will be on 45nm, smaller process than what Core 2 is on now, so it will consume less energy and have less leakage.

[Carfax]

rather unique since techinically conroe should be able to do 4 integer operations per cycle while AMD can only do 3...
Funny, then again conroe has 1 complex integer unit and 3 simple and AMD just has 3 complex

NN_step, this is overtly biased even for you..

Seriously, the P-M family and all it's derivatives are well known to be extremely strong in integer.

Even Yonah rocks the K8 in integer, let alone Core 2..

[3NZ0]

funny considering that K8 beats conroe at the same clock speed in
SM Cryptography ( 8.2% faster) and SM STREAM (4.3% faster) which If I remember correctly are both mainly integer programs that use lots and lots of memory

But the question is, does that have a noticable impact on the real world use?

[perkam]

nn, don't try to threadjack,

Everyone else chill. There have been many sources of info on K8L but only 1 for Yorkfield.

Perkam

[fhpchris]

I like aggravate discussions wiiiiiiiii :banana:

[erwinz]

:hehe: i'll wait for the real thing an let it fight in a bout.. :D

[nn_step]

:hehe: i'll wait for the real thing an let it fight in a bout.. :D
a very valid statement. Honestly we have no idea how Intel's or AMD's monster will turn out but all we know is that the next few years are going to very interesting times for us

[Theli]

I'm wondering, what exactly are the advantages of reducing 90nm to 65nm and on to 45nm? Someone mentioned reduced power consumption.

[metro.cl]

funny considering that K8 beats conroe at the same clock speed in
SM Cryptography ( 8.2% faster) and SM STREAM (4.3% faster) which If I remember correctly are both mainly integer programs that use lots and lots of memory

that means nothing, maybe a crappy car can beat a ferrari at 20km/h but the ferrari at 300km/h beat it.

not saying C2D is a ferrari nor amd is a crappy car.

K8L will be good and also intel will have a good cpu this "new" cpu is just a conroe with some minor features added + the process shrink.

After this one Intel will have a awsome chip with many new things

[en4cer]

This is what the K8L will face, according to VR-Zone:



And Intel can pull it off too.. Their 45nm process will be top notch, and the Core architecture doesn't consume much power due to it's incredible efficiency.. It will also have SSE4, the most sigificant extension to Intel's SIMD since SSE2..

I hope the K8L is a badass mofo, because it will be facing an uphill battle next year..

Interesting times ahead folks....interesting times indeed!

Source (http://vr-zone.com/?i=4109)


:eek: :eek: OMG.... IT'S GOING TO TAKE OVER THE WORLD!!! :slap: :stick:


Let's reserve judgment until the product is on the market shall we?

[Carfax]

I'm wondering, what exactly are the advantages of reducing 90nm to 65nm and on to 45nm? Someone mentioned reduced power consumption.

Reduced power consumption, aswell as a higher transistor budget meaning they can add more features to the core, while not increasing the size.

Also, they can make more processors at lower prices.

All in all, process mastery is extremely important, and it's how Intel has managed to become such a giant in the semi-conductor industry.

[[XC] hipno650]

Are you kidding? Conroe slaughters the K8 in integer performance!

The K8L might be better than Yorkfield clock for clock in FP, but due to Yorkfield's clockspeed advantage, I think Yorkfield may have the upper hand overall.



The cache is still shared, just not between all four cores. The reason is probably because it's difficult to get four cores to share cache effectively at the same time.
What i mant was that i wanted it to share it! look at conroe and how it benifits surely it would multiply with 4 cores!

[.sentinel]

Obviously you've never seen nor heard anything about SSE4.. I think I'll just let you remain ignorant and laugh at you :banana:

Also, if this chip is real, it will be on 45nm, smaller process than what Core 2 is on now, so it will consume less energy and have less leakage.
Are you bull:banana::banana::banana::banana:ting me?

[fhpchris]

:eek: :eek: OMG.... IT'S GOING TO TAKE OVER THE WORLD!!! :slap: :stick:


Let's reserve judgment until the product is on the market shall we?

Or when people @ XS get working engineering samples.... ;)

Really, how can I get editied by saying something ontopic, yet NN_step gets to post like 6 messages flame baiting other people?

I do not understand perky?

[Fred_Pohl]

Funny fact that most people miss is that K8 still beats Conroe in interger performance...

They probably miss that "fact" because it's pure fiction. :slapass:

[The Ghost]

http://www.overclockers.ru/images/news/2006/10/05/amd_02.jpg

[arisythila]

http://www.overclockers.ru/images/news/2006/10/05/amd_02.jpg

Mmmmmmm!

~Mike

[fhpchris]

more pictures and power points :(

vaporware = :slap:

[xenolith]

Remember guys, the enthusiast/performance market is relatively small. As long as AMD meets price point projections and continue to grow in the larger market segments (i.e., servers, mobile, OEMs, etc.) they'll be fine as a business.

[zabomb4163]

I'm wondering, what exactly are the advantages of reducing 90nm to 65nm and on to 45nm? Someone mentioned reduced power consumption.

http://www.intel.com/technology/silicon/new_45nm_silicon.htm

"More than 20 percent improvement in transistor switching speed or more than a five-fold reduction in transistor current leakage."

[Entity_Razer]

I think I'll just let you remain ignorant and laugh at you :banana:


No your not.....

Stop the pissing contest and get back to your daily routines :p:

We'll see once ES samples hit Xs won't we? and NN almost always had a ES sample up till now so... :) or he could at least get one so lets just wait and find out instead of making this into a pissing contest

[zakelwe]

The original quote from VR-zone says 12MB L2 cache 2x6MB.

Should that not be 4x3MB for Yorkfield ?

I do like AMD's new naming convention for the series. It's easy to be underwhelmed by AMD bare figures as they tend to be smaller than Intels, but that does not mean it will not be competitive.

Regards

Andy

[metro.cl]

The original quote from VR-zone says 12MB L2 cache 2x6MB.

Should that not be 4x3MB for Yorkfield ?

I do like AMD's new naming convention for the series. It's easy to be underwhelmed by AMD bare figures as they tend to be smaller than Intels, but that does not mean it will not be competitive.

Regards

Andy


yes but because it is shared between two cores the said 2x6mb (or 3x2x2) the thing is that one core can get a max of 6mb if necesary from the cache

[kemo]

All the rumors about yorkfield doesn?t make any sense to me old road maps said it has two quad cores each sharing 6mb of cache now they are saying quad core with 6 X 2mb of cache I think it is better to wait and see official road map from Intel

Edit : penyrn is merom replacement (old road map)
________
ultimate fighters (http://ultimatefighters.org/)

[perkam]

more pictures and power points :(

vaporware = :slap:K8L is obviously superior to Conroe if Intel needs Yorkfield to compete with it...with that in mind, K8L has always been a future product.

Vaporware only occurs when you have a product launched or announced but nowhere to buy it...calling this vaporware would be like calling ddr3 vaporware because it isn't here yet.

Perkam

[Vassili]

K8L is obviously superior to Conroe if Intel needs Yorkfield to compete with it...with that in mind, K8L has always been a future product.
K8L = Future product
Yorkfield = Future product

Both companies are working on a better products, or should the company that has the best product at this time, stop working on future products until the other company has some better products?:slap:

[onewingedangel]

K8L is obviously superior to Conroe if Intel needs Yorkfield to compete with it


I expected better from you...

Intel are releasing yorksfield as it will be cheaper to produce and consume less power, as well as give them a good deal of time to improve yields on the 45nm process. A 3.73 clock on 45nm is also very conservative considering you can get this on 65nm so I suspect the idea is to further reduce power consumption rather than ramp up clocks to beat amd's new design.

Whatever the whole amd vs intel pissing match will result in, the most presient bit of information for me is that intel will be moving to 45nm 9 months after amd get to 65nm. Amd is not catching up in process technology like we all thought.

[Sanborn]

I'm not really sure how people expect Altair with 4mb total of L2 thats 2.9Ghz at 65nm to compete with a product that has 3x the amount of cache...a Ghz faster (not that it matters anymore)....at a smaller size.

and question: how much will lets say...normal temps of a Core 2 Duo be lowered if it were a 45nm ?

[The Ghost]

I'm not really sure how people expect Altair with 4mb total of L2 thats 2.9Ghz at 65nm to compete with a product that has 3x the amount of cache...a Ghz faster (not that it matters anymore)....at a smaller size.
hasn't this been done already ?

[milkcafe]

here are the High-end cpu 3dmark06 cpu score comparision

the Kentsfield , Clovertown and the Tulsa are very strong

I want to know the score of Yorkfield and Altair FX


AMD Athlon(tm) FX57 @ 4208 MHz
3DMark06 CPU Score 1696 Marks
http://img199.imageshack.us/img199/9243/cpu42089kx.jpg

Intel Core 2 EXtreme X6800 4MB L2cache 2.93GGHz
3DMark06 CPU score　2512
http://enthusiast.hardocp.com/images/articles/1159465423unhAYDa7RG_2_2_l.gif

AMD Athlon(tm) FX62 @ 3676 MHz
3DMark06 CPU Score 2787 Marks
http://service.futuremark.com/compare?3dm06=329639

Intel Core Duo T2600 @ 3516 MHz
3DMark06 CPU Score 2874 Marks
http://service.futuremark.com/compare?3dm06=303322

Intel Pentium 965XE @ 6172 MHz
3DMark06 CPU Score 3314 Marks
http://service.futuremark.com/compare?3dm06=86920

AMD 2x Opteron 275 @ 2797 MHz
3DMark06 CPU Score 3824 Marks
http://service.futuremark.com/compare?3dm06=91145

Intel(R) Xeon(R) Woodcrest CPU 5150M @ 2.66GHz
3DMark06 CPU score　4085
http://service.futuremark.com/compare?3dm06=491849

Intel Conroe X6800 @ 5058MHz
3DMark06 CPU Score 4387 Marks
http://service.futuremark.com/compare?3dm06=440284

Intel XEON 5160M @ 3000MHz (Woodcrest 3.0GHz 4MB L3cache )
3DMark06 CPU Score 4526 Marks
http://service.futuremark.com/compare?3dm06=519723

2x AMD Opteron(tm) 8220 SE processor @ 2800MHz
3DMark06 CPU Score　4782 Marks

Intel XEON 5150M @ 3333MHz (Woodcrest 3.0GHz 4MB L3cache )
3DMark06 CPU Score 4525 Marks
http://service.futuremark.com/compare?3dm06=531132

Intel Core 2 Quado QX6700 Kentsfield 8MB L2cache @ 3709MHz
3DMark06 CPU Score 5564 Marks
http://www.iamxtreme.net/coolaler/conroe/KENSFIELD/E6700/338/3700_3D06.gif

Intel(R) Xeon(R) Clovertown 2.66GHz (2x Intel Woodcrest Xeon 5150 CPU)
3DMark06 CPU score　5581
http://img180.imageshack.us/img180/5765/3dmarkkk6.jpg

Intel Core 2 Quado QX6700 Kentsfield 8MB L2cache @ 4104MHz
3DMark06 CPU Score 6057 Marks
http://www.tyrou.net/screens/12856-3d06.png

Intel XEON 7140M @ 3400MHz (Tulsa 3.40GHz 16MB L3cache )
3DMark06 CPU Score 6632 Marks

[Theli]

K8L is obviously superior to Conroe if Intel needs Yorkfield to compete with it...
A regression fallacy, I'm afraid. And a premature one at that. All it would take is for Intel to think that the K8L or any future AMD product-line will be superior to Conroe, or that consumers will at some point in the future wish for something faster than the Conroe, for them to want to keep developing new technologies. I'd say that the probability of that is around 100%. Again, I think we'll have to wait for benchmarks before we can be certain.

[Mr.X]

Will Yorkfield need a new socket and/or chipset? I assume it will but i just want to make sure.

[uOpt]

They are going back to individual L2 caches?

Interesting...

[nn_step]

A regression fallacy, I'm afraid. And a premature one at that. All it would take is for Intel to think that the K8L or any future AMD product-line will be superior to Conroe, or that consumers will at some point in the future wish for something faster than the Conroe, for them to want to keep developing new technologies. I'd say that the probability of that is around 100%. Again, I think we'll have to wait for benchmarks before we can be certain.
:stick: why would you say that conroe is better than K8L

[arisythila]

I expected better from you...

Intel are releasing yorksfield as it will be cheaper to produce and consume less power, as well as give them a good deal of time to improve yields on the 45nm process. A 3.73 clock on 45nm is also very conservative considering you can get this on 65nm so I suspect the idea is to further reduce power consumption rather than ramp up clocks to beat amd's new design.


Just because its cheaper to produce doesnt mean its going to cost less.... Im pretty sure Intel uses binning, and im PRETTY sure that Intel also charges ALOT more than it cost to make. I know this, because they are still in business. I also know the same thing about AMD. We were up into the 1000 dollars for a processor, when Intel had a far crappier product than AMD, Intel was STILL selling those processors for 900-1000 dollars. Even tho, the AMD part @ 900-1000 dollars was a faster product.

Now that Intel has a faster processor, It no longer cost 1000 dollars for AMD top of the line processor. Its more like 700 dollars. I really cannot say the same thing about Intel. Intel will always sell thier top of the line processor at the top of the line prices. AMD tends not to do that. If AMD is on top they will, Of they are on the bottom they wont.

Another reason I buy AMD. I dont have to pay top of the line prices for :banana::banana::banana::banana:.

~Mike

[vitaminc]

Just because its cheaper to produce doesnt mean its going to cost less.... Im pretty sure Intel uses binning, and im PRETTY sure that Intel also charges ALOT more than it cost to make. I know this, because they are still in business. I also know the same thing about AMD. We were up into the 1000 dollars for a processor, when Intel had a far crappier product than AMD, Intel was STILL selling those processors for 900-1000 dollars. Even tho, the AMD part @ 900-1000 dollars was a faster product.

Now that Intel has a faster processor, It no longer cost 1000 dollars for AMD top of the line processor. Its more like 700 dollars. I really cannot say the same thing about Intel. Intel will always sell thier top of the line processor at the top of the line prices. AMD tends not to do that. If AMD is on top they will, Of they are on the bottom they wont.

Another reason I buy AMD. I dont have to pay top of the line prices for :banana::banana::banana::banana:.

~Mike

it seems you don't grasp the idea of tiered pricing, and obviously you have not pay any attention to the numerous BIG price cuts Intel went through before C2D.

the price of the latest XE or FX edition is a extremely poor indicator of price/performance. its more a statuary tag.

EDIT: btw, i would cross fingers about a single die quad core yorksfield just to be safe.

[nn_step]

it seems you don't grasp the idea of tiered pricing, and obviously you have not pay any attention to the numerous BIG price cuts Intel went through before C2D.

the price of the latest XE or FX edition is a extremely poor indicator of price/performance. its more a statuary tag.

EDIT: btw, i would cross fingers about a single die quad core yorksfield just to be safe.
I believe the point both of you are trying to make is that AMD and Intel will only keep prices low IF they don't have the absolute best on earth. As LONG as they are both competing performance will always be good and the Prices will always be low

[Helmore]

That's because Intel's TDP right now are much lower than AMD's. Even with the doubling, Kentsfield's power consumption is roughly the same as a FX-62.
There is one thing that you are missing, that is the fact that AMD is much more strict in TDP ratings then Intel. So if AMD would rate a CPU at 125W, then Intel would rate it at 100W. Those figures ought to be taken with a grain of salt, as I do not know how much of a difference there is, but it is quite a well known fact that AMD is stricter in thier ratings.

[vitaminc]

There is one thing that you are missing, that is the fact that AMD is much more strict in TDP ratings then Intel. So if AMD would rate a CPU at 125W, then Intel would rate it at 100W. Those figures ought to be taken with a grain of salt, as I do not know how much of a difference there is, but it is quite a well known fact that AMD is stricter in thier ratings.

Not more strict but different methodology. Intel is using RMS power, AMD is using maximum power.

[nn_step]

Not more strict but different methodology. Intel is using RMS power, AMD is using maximum power.
Tue, as well as AMD includes the Memory controller, Intel on the hand doesn't

[Fred_Pohl]

There is one thing that you are missing, that is the fact that AMD is much more strict in TDP ratings then Intel. So if AMD would rate a CPU at 125W, then Intel would rate it at 100W. Those figures ought to be taken with a grain of salt, as I do not know how much of a difference there is, but it is quite a well known fact that AMD is stricter in thier ratings.

Since when is unsubstantiated conjecture a well known fact? Based on component wattage estimates a Woodcrest system consumes more power on paper than a comparable Opteron system, yet in real world testing the opposite is true.

I think you're confusing manufacturer specific TDP verbage with cpu family TDP ratings and reality. Obviously a 125W TDP FX62 consumes less than 125W just as a 65W TDP E6300 consumes less than 65W since neither cpu is at the top of their respective cpu family TDP ratings. At this time we can be fairly certain that a E6700 does consume 65W TDP but we still don't know if FX74 will be the top model in AMD's 125W family TDP, although it seems very likely.

[arisythila]

I believe the point both of you are trying to make is that AMD and Intel will only keep prices low IF they don't have the absolute best on earth. As LONG as they are both competing performance will always be good and the Prices will always be low

I was trying to make the point of when the FX-57 was out, and around 1000 dollars, They had the 3.7 ghz or 3.8ghz EE chip was also 1000 dollars. Even tho the FX-57 was a bit faster than it. It seems like to AMD if they are not king of the hill they cut prices, to were Intel doesnt really, unless there is a new technology. IE: C2D.

~Mike

[jimmyz]

:stick: why would you say that conroe is better than K8L

because you can buy a conroe now!

[nn_step]

because you can buy a conroe now!
the same could have been said about K8 and Conroe about 6 months ago :stick:
To claim one company's product is better than another's future product is a blatant flame bait and complete fanboyism.

[arisythila]

LOL, Werd™

~Mike

[jimmyz]

the same could have been said about K8 and Conroe about 6 months ago :stick:
To claim one company's product is better than another's future product is a blatant flame bait and complete fanboyism.

not meant as flame bait or fanboyism at all. i am just pointing out that the conroe is available now and therefore it wins. and for the record i hope amds next chips run circles around intel. i own amd and intel pcs

[Absolute_0]

Why is Intel unfair? Like AMD would have done their 50% price cuts without Intel putting the pressure on? :rolleyes:. They're both the same, big companies, don't be hatin.

Perkam, Intel doesn't NEED to release a new one to compete with K8L; Conroe with some clock increases would have no problem competing. But since the success of C2D, they've decided on an extremely fast an aggressive course of rapid die shrinks and Mhz increases. Which is exactly what you want to do once you have a base architecture that's extremely sound and efficient.

Right now they've got momentum and they're making the "wheel" on the steamroller even bigger.

As for 45nm quad cores at 3.46-3.73 Ghz, i called this months ago :fact:

[accord99]

There is one thing that you are missing, that is the fact that AMD is much more strict in TDP ratings then Intel. So if AMD would rate a CPU at 125W, then Intel would rate it at 100W. Those figures ought to be taken with a grain of salt, as I do not know how much of a difference there is, but it is quite a well known fact that AMD is stricter in thier ratings.
That's no longer the case:

http://www.xbitlabs.com/images/cpu/core2duo-shootout/power-2.png


A 31W TDP Yonah uses less power than a 25W Turion MT.

http://www.silentpcreview.com/article313-page5.html

[oijkue]

Is Yorkfield a native quad core chip?
If it's yes,why vr-zone says "2x6M L2" ?

[Absolute_0]

No, it's probably going to be like Kentsfield, 2x4mb, each 4mb shared by 2 cores, except this is 50% increased to 2x6mb. Since Connies love cache this should give a nice performance boost.

[uOpt]

There is one thing that you are missing, that is the fact that AMD is much more strict in TDP ratings then Intel. So if AMD would rate a CPU at 125W, then Intel would rate it at 100W. Those figures ought to be taken with a grain of salt, as I do not know how much of a difference there is, but it is quite a well known fact that AMD is stricter in thier ratings.

The TDP is not the power consumption of the processor.

The TDP is what the mechanical design of the casing can dissipate. It does not take into account what the die inside will actually take. That is why many AMD64 that have drastically different power consumption (same core same volts but at different clockspeeds) have the same TDP. Because it's the casing design and not the core.

Intel socket 775 and 771 CPUs have a lower paper TDP because they don't have as much surface (as much contact to the heatsink). But that doesn't mean the CPUs take more or less in practice.

To find out the actual consumption you have to measure. Neither Intel nor AMD specify that info. And it wouldn't make sense since as every overclocker know every die is different. We have all seen CPUs that take much more or less power at the same volts and frequency. In particular those CPUs that clock very well at low voltage already take a great deal of power more at that voltage that others.

This is why you cannot use TDP as the power consumption. It isn't. This is, BTW, one of the major reasons why the popular extreme PSU calculator is nonsense.

[adamant415]

This thread should have just been info about Yorkfield. Unfortunately since k8l was mentioned in the original title and the original post it seems doomed to turn into a flame bait thread. The die hard amd people need to refrain from posting comments in this thread as far as I'm concerned, you know who you are. That is just my rant, I know it happens both ways but maybe if one side stopped it would eventually quit.

Just my .02

[arisythila]

I know seriously.. I hate it when some people bring up Conroe in K8L threads, or even none K8L threads... Worthless in my opinion. I guess you guys have to live with my frusteration.

~Mike

[thunderstruck!]

or even none K8L threads...
This is a non K8L thread. Should Conroe not be mentioned?:p:

[arisythila]

Was talking about AMD based none K8L conversations.

~Mike

[PetNorth]

That's no longer the case:

http://www.xbitlabs.com/images/cpu/core2duo-shootout/power-2.png


It depends on the link http://www.lostcircuits.com/cpu/low_e/5.shtml


A 31W TDP Yonah uses less power than a 25W Turion MT.

http://www.silentpcreview.com/article313-page5.html

I don't see MT, but ML (TDP 35W).

[vitaminc]

Tue, as well as AMD includes the Memory controller, Intel on the hand doesn't

Memory controller doesn't consume much power or real estate.

If you really want to compare the actual power, you will have to take all CPU, MC, FSB/HT controller, PCIe switchs into account (aka include both CPU andd NB). Or even easier, measure power draw from the wall using the same power supply.

TDP is TDP, its for thermal mechanical design and does not matter where the MC is located. If you are packing 200W of heat on the CPU socket, its a bad thermal design compare to spreading the heat around CPU socket, NB socket, etc.

[vitaminc]

I was trying to make the point of when the FX-57 was out, and around 1000 dollars, They had the 3.7 ghz or 3.8ghz EE chip was also 1000 dollars. Even tho the FX-57 was a bit faster than it. It seems like to AMD if they are not king of the hill they cut prices, to were Intel doesnt really, unless there is a new technology. IE: C2D.

~Mike

I will say it again. The flagship chips are not a good indicator of pricing strategy.

Intel has drastically cut their CPU prices 3 times this year, before C2D is out.

You are just finding a null reason to hate. :p

[nn_step]

Memory controller doesn't consume much power or real estate.

If you really want to compare the actual power, you will have to take all CPU, MC, FSB/HT controller, PCIe switchs into account (aka include both CPU andd NB). Or even easier, measure power draw from the wall using the same power supply.

TDP is TDP, its for thermal mechanical design and does not matter where the MC is located. If you are packing 200W of heat on the CPU socket, its a bad thermal design compare to spreading the heat around CPU socket, NB socket, etc.
:rofl:

and lets us be completely HONEST
intel Does NOT include the chips or the Memory or ANYTHING but the processor its self. Period.
End of discussion

[arisythila]

I will say it again. The flagship chips are not a good indicator of pricing strategy.

Intel has drastically cut their CPU prices 3 times this year, before C2D is out.

You are just finding a null reason to hate. :p

This is defently not the reason i 'hate' intel. I wont bring up how they crushed my way of life back in 1999-2000. I dont want to have to bring up any of that stuff. I could have VERY valid reasons to 'hate' intel. but i choose not too. C2D is currently the fastest chip on the market right now. 100%.

I personally think K8L will beat Yorkfield clock for clock hands down. HANDS down...

~Mike

[onewingedangel]

Power usage should be measured at the mains for the whole system to make fair comparisons.

At the minute intel offers better processor and system level power usage.

When AMD finally get to 65nm maybe this will change, maybe it won't.

Remember nvidia's amd northbridges use more power than intels northbridges even though they dont have a memory controller.

[vitaminc]

:rofl:

and lets us be completely HONEST
intel Does NOT include the chips or the Memory or ANYTHING but the processor its self. Period.
End of discussion

and Intel C2D systems draw less power at the wall compare to AMD A64X2/Opty systems. Period.
End of discussion.

[thunderstruck!]

This is defently not the reason i 'hate' intel. I wont bring up how they crushed my way of life back in 1999-2000.

Intel is responsible for your hardship 6-7 years ago? Did you work for them? They hurt a family member? Come on, they make CPUs...


I personally think K8L will beat Yorkfield clock for clock hands down. HANDS down...
Your high hopes may be crushed (no pun intended) or they may not. There havn't been any results from either chip, how can you make that conclusion?

[accord99]

It depends on the link http://www.lostcircuits.com/cpu/low_e/5.shtml
Only the undervolted and slower EEs have lower power consumption. On the other hand, the E6700 barely uses half the power of the slower FX-62.



I don't see MT, but ML (TDP 35W).
The ML is a misprint, it's voltage is 1.22v which is indicative of a MT and it matches the power consumption of a MT in an earlier review:
http://www.silentpcreview.com/article300-page6.html

[Absolute_0]

We could argue all day :p:

But let's consider instead the overclocking of these chips. 3.73 Ghz stock clocks and 45nm manufacturing process? What's that mean, 4.5 Ghz on air!?! :D

[theteamaqua]

i aint rooting for no one. who ever makes faster chip is what matters. that includes after OCed ...

intel does have a bad habbit of releasing ES chips OC higher than what retail can do ... E6600 ES 4GHz , E6700 4.1GHz on air anyone? i havent seen E6700 4.1GHz on air, also my crappy E6400 L624A264 wont do anything higher than 425fsb. 421fsb good for 24/7 .
PcCI2iminal agrees with me right here: (on E6400 L624A264 425fsb wall)
http://www.xtremesystems.org/forums/showpost.php?p=1702168&postcount=121

so if yorksfield ES can do 4.5GHz, im subtreacting 300MHz from that. Thats 4.2GHz with retail.

Now for AMD, AMD's 90nm doesnt OC very well, X3800 > 2.8GHz on air , thats it. also it is a memory latency whore. i hope they fix that in K8L , but they prolly wont, though. If K8L cant beat yorksfield clock for clock, i prolly wont buy their product for at least 2 years .... since intel's true quadcore is com'n 2008....

DDR3 will prolyl be out by then, i hope they dont jack up the price much, both mobo/cpu maker.

i am very excited about intel's 45nm, the jump from 90 to 65nm was pretty nice. pentium d 930 4GHz stcok vcore . i do think that yorksfield has 4.5GHz potential.

[kemo]

Come on guys why all that flame there is ain?t enough information about k8l and yorkfield so why keep arguing
________
List of bellhousing patterns (http://www.chevy-wiki.com/wiki/List_of_GM_bellhousing_patterns)

[PetNorth]

Only the undervolted and slower EEs have lower power consumption. On the other hand, the E6700 barely uses half the power of the slower FX-62.


The ML is a misprint, it's voltage is 1.22v which is indicative of a MT and it matches the power consumption of a MT in an earlier review:
http://www.silentpcreview.com/article300-page6.html

Remember these measurements are with Conroe B1 stepping, and you know what hapened with B1 and load... I'd like to see power measurements with the retail parts (B2)...

[coldpower27]

This is going to be interesting if correct, this is likely 2 Ridgefield Cores in MCM on the 45nm process which is going to be pretty efficient for production.

Though 45nm process by Q3 2007 would be damn amazing.

[Turtle 1]

I am a little confused. Me and wife just got back from 2 weeks in Hawaii so I am trying to catch up. I read an article before I left that said the Yorkfield would be 2 seperate dies as is kentsfield . But all the cache would be shared on both dies.

Since this info comes from VR zone . Which is less reliable than the inquirer I think we need more info.

I am still looking for the link but here's some interesting reading.

http://www.xbitlabs.com/news/cpu/display/20061002041328.html

From above link . Native quad core in 2007

Intel was originally expected to launch dual-core Wolfdale (for desktops) and dual-core Penryn (for laptops) processors made using 45nm process technology in 2008, according to earlier information. However, considering the recent progress with 65nm product shipments and ongoing progress with 45nm development, the company seemingly speeds up its roadmap. In addition to 45nm dual-core chips in 2007, there are rumours about single-die quad-core chip made using 45nm process technology in the Q3 2007.


http://www.hkepc.com/bbs/itnews.php?tid=675864&starttime=0&endtime=0&www.nordichardware.se

http://www.firingsquad.com/news/newsarticle.asp?searchid=12475

Now if you look at VRzones report it says 50 new sse4 instructions . If you read more reliable sources you will see that the (penryn wolfdale cpu's will only recieve 30 new instructions and the latter nehalem will use 50 instuctions from sse4 .

So what most are saying is 12mb. of shared cache= 3mb per core or 1 core can use all 12 mb of cache. Also most reports are saying that yorksfield will be a native 4core cpu. Wolfdale will be the dual core model clocked at 4ghz with 6mb of shared cache.
You add in the 30 sse4 instructions aong with the high k gates (metal ) and the higher clocked speeds the wolfdale should be 75% faster than Conroe.

Isn't it amazing . The big question is who is out of the gate first. Wolfdale/ yorksfield or K8L(K9). My guess is the the K9 doggy will be first but will be no match for the Wolf/ yorksfield.

Intel is really turning on the heat. In 2008 we get nehalem . Man I don't want to ever piss Intel off.

[kemo]

I am a little confused. Me and wife just got back from 2 weeks in Hawaii so I am trying to catch up. I read an article before I left that said the Penryn would be 2 seperate dies as is kentsfield . But all the cache would be shared on both dies.

Since this info comes from VR zone . Which is less reliable than the inquirer I think we need more info.

I am still looking for the link but here's some interesting reading.

http://www.xbitlabs.com/news/cpu/display/20061002041328.html

From above link . Native quad core in 2007

Intel was originally expected to launch dual-core Wolfdale (for desktops) and dual-core Penryn (for laptops) processors made using 45nm process technology in 2008, according to earlier information. However, considering the recent progress with 65nm product shipments and ongoing progress with 45nm development, the company seemingly speeds up its roadmap. In addition to 45nm dual-core chips in 2007, there are rumours about single-die quad-core chip made using 45nm process technology in the Q3 2007.


http://www.hkepc.com/bbs/itnews.php?tid=675864&starttime=0&endtime=0&www.nordichardware.se

http://www.firingsquad.com/news/newsarticle.asp?searchid=12475

Now if you look at VRzones report it says 50 new sse4 instructions . If you read more reliable sources you will see that the (penryn wolfdale cpu's will only recieve 30 new instructions and the latter nehalem will use 50 instuctions from sse4 .

So what most are saying is 12mb. of shared cache= 3mb per core or 1 core can use all 12 mb of cache. Also most reports are saying that yorksfield will be a native 4core cpu. Wolfdale will be the dual core model clocked at 4ghz with 6mb of shared cache.
You add in the 30 sse4 instructions aong with the high k gates (metal ) and the higher clocked speeds the wolfdale should be 75% faster than Conroe.

Isn't it amazing . The big question is who is out of the gate first. Wolfdale/ yorksfield or K8L(K9). My guess is the the K9 doggy will be first but will be no match for the Wolf/ yorksfield.

Intel is really turning on the heat. In 2008 we get nehalem . Man I don't want to ever piss Intel off.
The bad thing is that there is nothing clear these days everything is just reports with no confirmation or reliable source
for example wolfdale in some reports they say it should replace Conroe with 6mb of cache others say to replace allendle with 3 Mb also others said penyrn is replacement of Conroe
but old reports said something called ?Ridgefield? should replace Conroe with the 6 Mb
any way yorkfield they said it is hybrid quad core which means two quad cores in one package and each quad shares 6 Mb of cache now the are saying quad core with 2 X 6mb of cache so nothing for sure till intel says so

Any way cpu manufacturing is getting very exciting these years and every thing is great for consumers
________
Mazda 767 (http://www.ford-wiki.com/wiki/Mazda_767)

[camouflage]

:slobber: 2x6MB L2 cache.........:slobber: :slobber: :slobber: :slobber: :slobber:

This chip will do 10 sec SP1M even You removed the RAM before.....:lol:

[oldblue]

The bad thing is that there is nothing clear these days everything is just reports with no confirmation or reliable source

I agree. Has anyone sorted out what information we have for Yorkfield officially from Intel (e.g. from IDF), and what is just floating about on the rumor sites? I find the former information is generally much more reliable than the latter.

[nn_step]

You add in the 30 sse4 instructions aong with the high k gates (metal ) and the higher clocked speeds the wolfdale should be 75% faster than Conroe.

Isn't it amazing . The big question is who is out of the gate first. Wolfdale/ yorksfield or K8L(K9). My guess is the the K9 doggy will be first but will be no match for the Wolf/ yorksfield.

Intel is really turning on the heat. In 2008 we get nehalem . Man I don't want to ever piss Intel off.
well it better be 75% faster, it is twice the cores :rolleyes:
But since no one has specifically said what it is 75% faster in, one can only assume multi-threaded applications.
But if they mean single threaded applications they would have to switch from a 4 issue to an 8 issue processor and that would be one hell of a feat

[Turtle 1]

well it better be 75% faster, it is twice the cores :rolleyes:
But since no one has specifically said what it is 75% faster in, one can only assume multi-threaded applications.
But if they mean single threaded applications they would have to switch from a 4 issue to an 8 issue processor and that would be one hell of a feat
No ! Actually I was referring to the 2core wolfdale.

It been stated it will be a 4 ghz. part. and it will also be a 45nm high K (metal gates process) With the higher Clock and the high k gates running at 20%+ switching speed . Its rather easy to to get a 75% increase in performance. What isn't known yet is if the new 45nm parts will also take the Tigerton point to point from each core to the NB. If it does look for a 100% increase in performance. All this while having 5x less leakage.:toast:

For Intel to go from 4 issue to 8 issue processor wouldn't be as hard as you think. THink Itanium. But thats not whats going to happen . Maybe nehalem will go to a 6 issue core .

http://www.reghardware.co.uk/2006/09/28/intel_readies_yorkfield/

http://www.theregister.co.uk/2005/12/05/intel_45nm_roadmap/

[Turtle 1]

i aint rooting for no one. who ever makes faster chip is what matters. that includes after OCed ...

intel does have a bad habbit of releasing ES chips OC higher than what retail can do ... E6600 ES 4GHz , E6700 4.1GHz on air anyone? i havent seen E6700 4.1GHz on air, also my crappy E6400 L624A264 wont do anything higher than 425fsb. 421fsb good for 24/7 .
PcCI2iminal agrees with me right here: (on E6400 L624A264 425fsb wall)
http://www.xtremesystems.org/forums/showpost.php?p=1702168&postcount=121

so if yorksfield ES can do 4.5GHz, im subtreacting 300MHz from that. Thats 4.2GHz with retail.

Now for AMD, AMD's 90nm doesnt OC very well, X3800 > 2.8GHz on air , thats it. also it is a memory latency whore. i hope they fix that in K8L , but they prolly wont, though. If K8L cant beat yorksfield clock for clock, i prolly wont buy their product for at least 2 years .... since intel's true quadcore is com'n 2008....

DDR3 will prolyl be out by then, i hope they dont jack up the price much, both mobo/cpu maker.

i am very excited about intel's 45nm, the jump from 90 to 65nm was pretty nice. pentium d 930 4GHz stcok vcore . i do think that yorksfield has 4.5GHz potential.

Have you ever considered the guys that are getting those ES cpu's before anyone else can dream to have one . Are guys that are really really sharp when it comes to hardware. The guys that were showing off those es c2d chips here are really quit good , I go so far as to say they are :banana: :banana: :banana: :banana: :banana: :banana: :banana: out standing.

[Turtle 1]

the same could have been said about K8 and Conroe about 6 months ago :stick:
To claim one company's product is better than another's future product is a blatant flame bait and complete fanboyism.


It was said 100's of times go back and check the Threads . :stick:

[theteamaqua]

Have you ever considered the guys that are getting those ES cpu's before anyone else can dream to have one . Are guys that are really really sharp when it comes to hardware. The guys that were showing off those es c2d chips here are really quit good , I go so far as to say they are :banana: :banana: :banana: :banana: :banana: :banana: :banana: out standing.

http://www.xtremesystems.org/forums/showpost.php?p=1702168&postcount=121

.... read it

and i do believe that PcCI2iminal is a good OCer. and yet he got the same fsb wall as me, so it must be the CPU's problem

[Turtle 1]

Ya I looked at it. You know I kinda gotta leave this alone. 3.4 ghz. from a 2.13 ghz $300 chip. That is what 45% faster than an FX 62. That cost $700 . Do all X2 3800 hit 2.8ghz. All chips are not equal. I thinK 3.4 ghz is great as long as its stable.

I remember befor the es chips showed up over at [H] we were discussing this and everyone agreeded that if these chips did 3.2 how great they would be. I of course being a Intel fanboy was saying I exspected 3.6 Ghz from a 6800. . I went so far as to say maybe 4 Ghz on water.

Well you gotta to know the the AMD fan-boys had a field day with that. Even some Intel fanboys said I had gone to far. Be happy with your 3.4 ghz performance monster.

My wife has a c2d 6800 @ 3.8.5 ghz useing corsair 5400 ul running 1:1 @ 350 fsb with 3-2-2-8 timings. She hasn't tried to go any further with it. I got to play online gaming with it earlier tonight. God its sweet no doubts about it. When I got done playing . I shout it done to enter the bios to see what it would do. . Dam she has it password locked. After 33 years of marriage she knows me to well.

I don't know if PcC12iminal is a good O/Cer or not. I can say and I have no desire to offend him or you but his desk top is sloppy.

[ochw]

yorkfeid is looking well nice

[gn0s1s]

I don't know if PcC12iminal is a good O/Cer or not. I can say and I have no desire to offend him or you but his desk top is sloppy.


LOL Saving Screens to desktop is sloppy?

[Turtle 1]

Ya . I guess if you want to do a simple one click its ok but For me a couple more clicks and a few seconds is nice.

[osiris999]

everything is saved to my desktop

[doompc]

K8L might be at least as fast as Conroe, clock by clock, maybe even faster than Wolfdale, thanks to it's massive FPUs, single cicle SSE and the integrated memory controller.
But we don't know how power hangry is it, nor it's overclocking potencial.

AMD will demo K8L Opteron in december, also will release 65nm K8 then.
So will have a preview of AMD's 65nm process power/clocking capabilities (but it will be like Winchester over Newcastle).

Question is, AMD will release native quad-core in soon 2007 (server, desktop in Q2/Q3) and Intel won't do it til 2008, gotta sell these dual-die-fsb-shared-crap til then...
Multiple K8L Opteron with HTT3.0 will rule server market til 2008/9 when Intel goes CSI + integrated memory controller.

But I think Wolfdale will remain desktop #1 even when Antares (K8L dual core with L3 cache) apear.
Intel's 45nm process might clock really well (4GHz+ on air), 65nm K8L won't be able to keep up.

My 2 cents. :rolleyes:
Sorry my bad english...