Intel's 8th Generation Core Family - Coffee Lake (LGA 1151, 6C/12T)

Where do you expect Core i7-8700K's Turbo to land?

  • 3.8/3.9 GHz

    Votes: 0 0.0%

  • 4.0/4.1 GHz

    Votes: 3 23.1%

  • 4.2/4.3 GHz

    Votes: 6 46.2%

  • 4.4/4.5 GHz

    Votes: 3 23.1%

  • 4.6/4.7 GHz

    Votes: 1 7.7%
  • Total voters
    13
  • Poll closed .
OrangeKhrush said:
I can see why you follow that, he is like a poster child of all things Intel, It is like a mecca of Intel.
Click to expand...

Don´t tell me you got upset because you bought the hype of the PR slide nodes that couldn´t deliver to their names in the hope that the evil Intel wouldn´t be in the lead? How do you think GloFos "7nm" node compares? ;)
 
Shintai said:
Don´t tell me you got upset because you bought the hype of the PR slide nodes that couldn´t deliver to their names in the hope that the evil Intel wouldn´t be in the lead? How do you think GloFos "7nm" node compares? ;)
Click to expand...

Probably less dense as it is a not as wealthy process
 
The only thing Intel has released on 10nm is paper and a coming soon logo....
 
With the ICL PCH being released to mobo manufacturers I imagine it's in track, but I can only hope.
 
Gideon said:
The only thing Intel has released on 10nm is paper and a coming soon logo....
Click to expand...

Intel is very conservative with "coming soon" sorts of messaging to OEMs. There's a very high degree of confidence it is actually coming soon.

And as Shintai and others have indicated, it really is what most would consider a 10nm process not "10nm class" or whatever justification others use. Dayman and juanrga's posts about density contains the important thing to understand with this stuff - the density is the telling factor.
 
Dayman said:
Huh? Where did he say the 8 core was now 14nm?He said Intel had problems with 14nm which made Intel take 100% focus off 10nm, then they had bigger problems with 10nm than 14nm it seems, thus they introduced 14nm+ and 14nm++.
profile_large.png


Here's the reference for 8 core cannonlake.

Originally we were meant to have (desktop)

Skylake -> Cannonlake -> Icelake -> Tigerlake

Instead, due to 10nm problems we got

Skylake -> Kaby lake -> Coffee Lake -> Icelake -> Tigerlake
Click to expand...

too bad i dont count quoting 2yr + news release to be any more accurate. any businesses changes their plan, especially how intel m!lked hard with its haswell refresh, then KBL now CFL. intel might be planning to design 8 core, bet your life that 2 cores would have been disabled if not due to AMD.


juanrga said:
I just said you that Intel originally planned an 8-core 10nm chip for mainstream, and then I said you that now that 10nm is finally ready, Intel is going to release a 10nm 8-core chips for mainstream. Where you get that it is 14nm?

The 10nm 8-core was planned years before Zen was even tapped out.
Click to expand...

thats what i thought, but we wont know until it comes. to be frank with u intel changed so much due to AMD its not even funny and they aren't even trying to hide it anymore rofl.

its good that we can get 8 cores mainstream due to AMD, thank you AMD.
 
PhaseNoise said:
Intel is very conservative with "coming soon" sorts of messaging to OEMs. There's a very high degree of confidence it is actually coming soon.

And as Shintai and others have indicated, it really is what most would consider a 10nm process not "10nm class" or whatever justification others use. Dayman and juanrga's posts about density contains the important thing to understand with this stuff - the density is the telling factor.
Click to expand...

Yes Intel has touted it's process quite a bit and it's supposed advantages. However one needs to see this evidence at some point and so far Intel has been all talk rather then show. They know others are about to surpass them in node size which will be a advantage they will no longer have. I mean Intel touts how much better their 14nm is over GF yet AMD is right there with Intel these days. 10nm was supposed to show up in 2015, then 2016 and now 2017 and still nothing other then papers so it's no surprise some of us are a bit skeptical.
 
Gideon said:
Yes Intel has touted it's process quite a bit and it's supposed advantages. However one needs to see this evidence at some point and so far Intel has been all talk rather then show. They know others are about to surpass them in node size which will be a advantage they will no longer have. I mean Intel touts how much better their 14nm is over GF yet AMD is right there with Intel these days. 10nm was supposed to show up in 2015, then 2016 and now 2017 and still nothing other then papers so it's no surprise some of us are a bit skeptical.
Click to expand...

GF's process is not at parity with Intel's. Ryzen does incredibly well due to a very solid design, but is kneecapped hard on the upper end by the process not allowing higher clocks.

Based upon my experience, I strongly believe AMD was expecting the high end (1800s) to reach much higher clocks than it can. There's a hard wall that voltage and cooling aren't able to rectify.
 
PhaseNoise said:
GF's process is not at parity with Intel's. Ryzen does incredibly well due to a very solid design, but is kneecapped hard on the upper end by the process not allowing higher clocks.

Based upon my experience, I strongly believe AMD was expecting the high end (1800s) to reach much higher clocks than it can. There's a hard wall that voltage and cooling aren't able to rectify.
Click to expand...

That has to do with the choice of LLP which is great for mobile and server but not as good for high speeds on the desktop. But like Shintai likes to say the money is in mobile and server so I guess it was a brilliant choice for AMD then. The reality is as the core count goes up the speed advantage Intel has on MHz disappears as we have seen with Threadripper. Parity or superiority at a supposed process node means nothing unless the chip performs, Intel needs to stop pointing to paper and show us something otherwise it's all smoke and mirrors.
 
Gideon said:
That has to do with the choice of LLP which is great for mobile and server but not as good for high speeds on the desktop. But like Shintai likes to say the money is in mobile and server so I guess it was a brilliant choice for AMD then. The reality is as the core count goes up the speed advantage Intel has on MHz disappears as we have seen with Threadripper. Parity or superiority at a supposed process node means nothing unless the chip performs, Intel needs to stop pointing to paper and show us something otherwise it's all smoke and mirrors.
Click to expand...

The 14nm LPP process is GF's highest performing process which was available for Ryzen. What process would you suggest they have used? I'm clearly missing something here.

I am not sure what to make of the last sentence. Intel does have very high performance processors, this is the first time I've heard their products referred to as smoke and mirrors.
 
oleNBR said:
thats what i thought, but we wont know until it comes. to be frank with u intel changed so much due to AMD its not even funny and they aren't even trying to hide it anymore rofl.

its good that we can get 8 cores mainstream due to AMD, thank you AMD.
Click to expand...

Sorry I don't buy your theory that Intel fabricated a time machine to travel in the future, see the performance of RyZen on reviews and then did travel back in time to change the plans and roadmaps even before RyZen had been tapeout.
 
"But Intel does not stop there because these drivers also already support the next two generations of processors namely the Cannonlake platform (Core 9th Gen?) Which will use this same socket LGA 1151 V2 and the same chipsets 300 Series but will go to a finesse of etching of 10 nm. Cannonlake is scheduled for late 2017 or early 2018 following the models. The next platform Icelake (Core 10th Gen?) Scheduled for 2019 is also supported by these drivers but we still have only a few details about it except the fact that it will be a 10 nm etching that new 400 Series chipsets should be born (and again a new socket)."

This is French to English on google translate, so expect it to be iffy.


https://www.touslesdrivers.com/index.php?v_page=3&v_code=6741
 
"They were claiming over 3.3GHz for A75 at 250uW(microwatts) per MHz" No idea if this is true, I found it in the comments of the Anandtech article I posted below the screen shots, but thats below a watt if true. 0.825w to be exact.
 
Last edited:
PhaseNoise said:
The 14nm LPP process is GF's highest performing process which was available for Ryzen. What process would you suggest they have used? I'm clearly missing something here.

I am not sure what to make of the last sentence. Intel does have very high performance processors, this is the first time I've heard their products referred to as smoke and mirrors.
Click to expand...

There is also a 14nm HP process as well.

I never said they didnt perhaps talking about 14nm and 10nm was confusing. I am referring to the missing Intel 10nm processors that were promised some time ago. They love to show pictures of the 10nm process and slides but they have yet to deliver this superior process in the form of a processor anyone can buy.
 
Gideon said:
There is also a 14nm HP process as well.

I never said they didnt perhaps talking about 14nm and 10nm was confusing. I am referring to the missing Intel 10nm processors that were promised some time ago. They love to show pictures of the 10nm process and slides but they have yet to deliver this superior process in the form of a processor anyone can buy.
Click to expand...

Its funny you mention it, yet we have to hear endless about a "7nm" process that may be due in 2019 if it doesn't slip to 2020 or gets cancelled like its previous 2 nodes.
 
Shintai said:
Its funny you mention it, yet we have to hear endless about a "7nm" process that may be due in 2019 if it doesn't slip to 2020 or gets cancelled like its previous 2 nodes.
Click to expand...

That GF 7nm node might just be here before that eternally delayed Intel 10nm node. Of course you know if you were always right we would be able to buy Coffeelake processors by now as well. There is a reason the media is talking about Intel 10nm and not in a flattering light, it's long overdue.
 
Gideon said:
That GF 7nm node might just be here before that eternally delayed Intel 10nm node. Of course you know if you were always right we would be able to buy Coffeelake processors by now as well. There is a reason the media is talking about Intel 10nm and not in a flattering light, it's long overdue.
Click to expand...

I dont think the fake 7nm will be here anytime soon, despite all the hype and hopes. Specially not from a company that failed not once, but twice in a row trying to develop their own node. But we can see in a couple of years after its being hyped to the sky and beyond reaching 5ghz and what not... :D
 
AbRASiON said:
Doesn't that just mean the documentation is released. Surely the physical chips won't exist for a few more months?
Click to expand...

I believe that means the details are released to the mobo manufactures, you likely wont see these chips for another year or so.
 
juanrga said:
Sorry I don't buy your theory that Intel fabricated a time machine to travel in the future, see the performance of RyZen on reviews and then did travel back in time to change the plans and roadmaps even before RyZen had been tapeout.
Click to expand...

idont buy your theory either, sounds beautified intel biased opinion. we have had 6c HEDT for a long time since first gen i7, intel could have easily given it to us 3-4 yrs ago on consumer mainstream level but they wouldnt, they had no reason to theres simply nothing from AMD out perform or match it. now if ryzen2 do catch up a bit more on frequency and IPC, ontop of that being as efficient or more efficient than intel, having 2 less cores intel simply cant charge $100-150 more to justified a tiny tiny bit performance over AMD and a iGPU.

face it, intel got pressured hardcore into giving us 8 core mainstream, thats a GOOD thing.
 
oleNBR said:
idont buy your theory either, sounds beautified intel biased opinion. we have had 6c HEDT for a long time since first gen i7, intel could have easily given it to us 3-4 yrs ago on consumer mainstream level but they wouldnt, they had no reason to theres simply nothing from AMD out perform or match it. now if ryzen2 do catch up a bit more on frequency and IPC, ontop of that being as efficient or more efficient than intel, having 2 less cores intel simply cant charge $100-150 more to justified a tiny tiny bit performance over AMD and a iGPU.

face it, intel got pressured hardcore into giving us 8 core mainstream, thats a GOOD thing.
Click to expand...

HEDT != mainstream
 
You must log in or register to reply here.
Back
Top