Kaby Lake information.

[mikk]

Of course, they have to change plans accordingly.
Cannonlake core in 2017 will not be the same Cannonlake as Intel planned years ago. Maybe I'm wrong but there are several reasons for this option.

Cannonlake is linked to Gen10 graphics since a very long time, I doubt there was a change. Of course they could possibly develope a better Gen10 with more time, but there is no sign that Cannonlake was planned for Gen9.5.

I also doubt that Kabylake comes with HW accelerated SHA unlike Cannonlake.

 

[Eddward]

no matter what, even 0.5 Gen requiers much work, a lot more than just plain refresh without any iGPU improvements... and this work can not be done in just 1-1,5 year or so, I think... to split Cannonlake improvements iGPU into 2 generations is maybe possible... or they just changed it completely.. Gen 9.5 or Gen 10 is just a word on the paper...

 

[mikk]

no matter what, even 0.5 Gen requiers much work

You can't know this.

 

[Phynaz]

Kabylake seems it will be release the same time as ZEN, Q4 2016. But, they dont directly compete.

Of course they do. If they are on the market at the same time they are competing products.

 

[frozentundra123456]

Of course, they have to change plans accordingly.
Cannonlake core in 2017 will not be the same Cannonlake as Intel planned years ago. Maybe I'm wrong but there are several reasons for this option.

@frozentundra123456: by new uarch I mean updates in core and iGPU... refresh like Haswell refresh has nothing new...Mhz bump, maybe little bugfixes and unlocked instruction set, it requires max new revision/stepping on the same silicon... in Kabylake there will be updates, it's basically another 14nm tock with tick improvements...

Well, I actually hope you are right. We sure could use a bump in cpu over and above skylake, but I am not expecting it. Igpu, yes, but not cpu (other than a small mhz bump).

 

[frozentundra123456]

Of course they do. If they are on the market at the same time they are competing products.

Technically, of course you are correct. However, Aten is also correct (never thought I would say that) in that at least initially, Zen will be competing (hopefully for AMD at least) with the HEDT platform and servers rather than mainstream cpus with IGP.

 

[ShintaiDK]

Technically, of course you are correct. However, Aten is also correct (never thought I would say that) in that at least initially, Zen will be competing (hopefully for AMD at least) with the HEDT platform and servers rather than mainstream cpus with IGP.

Like Bulldozer?

 

[AtenRa]

Like Bulldozer?

Well, if you believe the 8Core 16 Threads ZEN will only perform as the 4Core 8Threads Intel CPUs then yes they are competing.

But after the information we have on the ZEN architecture, if anyone still believes that, he is clueless.

 

[ShintaiDK]

But after the information we have on the ZEN architecture, if anyone still believes that, he is clueless.

We heard that story before.

Phenom 40-50% faster than Core 2, no limit to Bulldozer performance going above and beyond.

So now on a shoehorn budget they are to create this miracle? And deliver, something they haven't done in this segment the last 10 years? And even worse, depending on a 3rd party foundry that couldn't even develop their own node.

Realism called. But you obviously didn't return the call for ages.

 

[AtenRa]

We heard that story before.

Phenom 40-50% faster than Core 2, no limit to Bulldozer performance going above and beyond.

That was for a QUAD CORE Phenom vs Dual Core Core 2. And no they didnt benchmarked Cinebench

So now on a shoehorn budget they are to create this miracle? And deliver, something they haven't done in this segment the last 10 years? And even worse, depending on a 3rd party foundry that couldn't even develop their own node.

Realism called. But you obviously didn't return the call for ages.

You are free to believe what ever you want, but reality is way different looking the world without those blue PR glasses.

Even if AMD just made a excavator 8Module 16 Threads CPU at 14nm it would obliterate any Intel 4C 8Thread 2016-2017 CPU in MT loads. But yes you like to put ZEN in the same segment as those Intel Quads for obvious reasons. Feel free to do so, you are the one that will be quoted later on.

 

[ShintaiDK]

I am sure you will cherry pick some 7zip benchmark or so for us to see while you proclaim another AMD victory with the imminent change in market share. While the rest of us will enjoy better performance on intel parts.

And no, AMD compared 4 Conroe cores with 4 Phenom cores. Barcelona vs X5355 to be more precise. but we all know what happened. Including the spec.org scandal that followed.

 

[Phynaz]

That was for a QUAD CORE Phenom vs Dual Core Core 2. And no they didnt benchmarked Cinebench



You are free to believe what ever you want, but reality is way different looking the world without those blue PR glasses.

Even if AMD just made a excavator 8Module 16 Threads CPU at 14nm it would obliterate any Intel 4C 8Thread 2016-2017 CPU in MT loads. But yes you like to put ZEN in the same segment as those Intel Quads for obvious reasons. Feel free to do so, you are the one that will be quoted later on.

This is a Kabylake thread, please keep the mythical AMD CPUs where they belong.

 

[Systems analyst]

Your numbers are flawed. On Intel investor relation site you can find the quarterly comparisons of the PC groups with the mobile numbers included, they do not come even close to what you are posting here.

April 2014, Intel reported $2 .8 billion earnings for 2014 Q1 PC group.
http://www.intc.com/common/download...9&filename=Earnings Release Q1 2014_final.pdf
April 2015, Intel reported $1 .4 billion earnings in the combined PC + mobile group for 2015 Q1.
http://www.intc.com/common/download...&filename=Earnings_Release_Q1_2015_Final2.pdf
- This is a shortfall of $1 .4 billion earnings in a quarter.

July 2014, they reported $3.7 billion earnings for 2014 Q2 PC group.
http://www.intc.com/common/download...1D&filename=Earnings Release Q2 2014 r134.pdf
July 2015, they reported $1 .6 billion earnings in the combined PC + mobile group for 2015 Q2.
http://www.intc.com/common/download...E&filename=Earnings_Release_Q2_2015_Final.pdf
- This is a shortfall of $2.1 billion earnings in a quarter.

October 2014, they reported $4.1 billion earnings for 2014 Q3 PC group.
http://www.intc.com/common/download...3&filename=Earnings_Release_Q3_2014_final.pdf
October 2015, they reported $2.4 billion earnings in the combined PC + mobile group for 2015 Q3.
http://www.intc.com/common/download...BAFB&filename=Earnings_Release_Q3_2015_5_.pdf
- This is a shortfall of $1.7 billion earnings in a quarter.


Projected 2015 Mobile/PC group shortfall c.f 2014 PC profits:
Q1 = 1.4, Q2=2.1, Q3=1.7, assume Q4=1.4
Total shortfall = 6.6 billion USD.
If 4.2 billion is attributable to Mobile,as per 2014, then PC shortfall for 2015 is 2.4 billion.
If 3.4 billion is attributable to Mobile, then PC shortfall for 2015 is 3.2 billion USD.

Look on page 8 or 9 or 10 of the PDFs for the numbers, "Supplemental Operating Segment Results". The numbers are correct. Can Intel turn this around?

 

[mrmt]

April 2014, Intel reported $2 .8 billion earnings for 2014 Q1 PC group.
http://www.intc.com/common/download...9&filename=Earnings Release Q1 2014_final.pdf
April 2015, Intel reported $1 .4 billion earnings in the combined PC + mobile group for 2015 Q1.
http://www.intc.com/common/download...&filename=Earnings_Release_Q1_2015_Final2.pdf
- This is a shortfall of $1 .4 billion earnings in a quarter.

July 2014, they reported $3.7 billion earnings for 2014 Q2 PC group.
http://www.intc.com/common/download...1D&filename=Earnings Release Q2 2014 r134.pdf
July 2015, they reported $1 .6 billion earnings in the combined PC + mobile group for 2015 Q2.
http://www.intc.com/common/download...E&filename=Earnings_Release_Q2_2015_Final.pdf
- This is a shortfall of $2.1 billion earnings in a quarter.

October 2014, they reported $4.1 billion earnings for 2014 Q3 PC group.
http://www.intc.com/common/download...3&filename=Earnings_Release_Q3_2014_final.pdf
October 2015, they reported $2.4 billion earnings in the combined PC + mobile group for 2015 Q3.
http://www.intc.com/common/download...BAFB&filename=Earnings_Release_Q3_2015_5_.pdf
- This is a shortfall of $1.7 billion earnings in a quarter.


Projected 2015 Mobile/PC group shortfall c.f 2014 PC profits:
Q1 = 1.4, Q2=2.1, Q3=1.7, assume Q4=1.4
Total shortfall = 6.6 billion USD.
If 4.2 billion is attributable to Mobile,as per 2014, then PC shortfall for 2015 is 2.4 billion.
If 3.4 billion is attributable to Mobile, then PC shortfall for 2015 is 3.2 billion USD.

Look on page 8 or 9 or 10 of the PDFs for the numbers, "Supplemental Operating Segment Results". The numbers are correct. Can Intel turn this around?

You don't know how to read a financial statement.

 

[Systems analyst]

You don't know how to read a financial statement.

For example, the PC group made a profit of $4.1 billion in 2014Q3, as stated on page 8 of the PDF referenced.
The PC+mobile group made a profit of $2.4 billion in 2015Q3, as stated on page 10 of the PDF referenced.
Intel make this very clear. If you can not accept their numbers as fact, then I can not help you further.

 

[Eddward]

Well, I actually hope you are right. We sure could use a bump in cpu over and above skylake, but I am not expecting it. Igpu, yes, but not cpu (other than a small mhz bump).

Will see, but honestly, I'm aiming rather at BDW-E.. I had 3 or 4 generations of quadcore (currently 4670K@4.5Ghz) and that's just enough. No real progress for past years.

 

[IlllI]

ok I'm confused. I'm not sure what to make of this "Intel's upcoming Skylake desktop chip with Iris Pro (henceforth referred to as "Skylake 4+4e") will not work on the motherboards that support the Skylake processors currently in the market. Indeed, these chips will require next generation boards designed specifically for the company's upcoming Kaby Lake family of processors. Interestingly enough, those Kaby Lake chips will apparently work just fine on current generation Skylake boards."

how could kaby lake be backward compatible with sky lake motherboards, yet sky lake cpus are not with kaby motherboards?

 

[III-V]

ok I'm confused. I'm not sure what to make of this "Intel's upcoming Skylake desktop chip with Iris Pro (henceforth referred to as "Skylake 4+4e") will not work on the motherboards that support the Skylake processors currently in the market. Indeed, these chips will require next generation boards designed specifically for the company's upcoming Kaby Lake family of processors. Interestingly enough, those Kaby Lake chips will apparently work just fine on current generation Skylake boards."

how could kaby lake be backward compatible with sky lake motherboards, yet sky lake cpus are not with kaby motherboards?

I think you messed up the wording on your question. Skylake will work on KL motherboards -- 4+4e Skylake will not work on today's Skylake motherboards, though. I am not positive as to why, but yeah.

 

[IlllI]

you're right that I am confused hah. I've never heard of a processor not compatible with a future motherboard, but the future cpu will be compatible with a past motherboard. I assume they are pin compatible for that to be possible, but wouldn't that confuse a lot of people?

 

[Magic Carpet]

4+4e Skylake will not work on today's Skylake motherboards, though.

Same as with Z87/Z97 and BDW?

 

[chucky2]

So on wiki (hardly a source for advance accurate information that turns out to be true), it lists chipset for Cannonlake the same as Kaby Lake.

Does anyone know if this is accurate? And, even if so, will the socket change for Cannonlake (even if the chipset were to stay the same) to make it unsuitable for Kaby Lake CPUs?

It would be nice to build a Kaby Lake system and know there is a Cannonlake upgrade path for CPU vs building a Skylake now and maybe mobo used won't work with future Kaby Lake (mercy of Intel and mobo team).

Chuck

 

[Magic Carpet]

It would be nice to build a Kaby Lake system and know there is a Cannonlake upgrade path for CPU vs building a Skylake now and maybe mobo used won't work with future Kaby Lake (mercy of Intel and mobo team).

Yeah, I am the same. I would like a chipset for at least couple of CPU generations. I was owned big time for not able to run Broadwell on a non x97 board.

 

[AMDisTheBEST]

I'll bet its a 5% improvement over the Skylake just like every successive generation of cpu after sandy bridge. No reason to release something ground breaking when no one else is completing.

 

[ShintaiDK]

I'll bet its a 5% improvement over the Skylake just like every successive generation of cpu after sandy bridge. No reason to release something ground breaking when no one else is completing.

You do know the gains have been in that area ever since the PPro in 1995?

Everything else have come from frequency and cores.

 

[jpiniero]

It would be nice to build a Kaby Lake system and know there is a Cannonlake upgrade path for CPU vs building a Skylake now and maybe mobo used won't work with future Kaby Lake (mercy of Intel and mobo team).

It is very unlikely that Cannonlake would support the same socket.