• 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 Broadwell-K & Skylake (non-K) desktop CPUs to launch in Q2-2015

S

Sweepr

Diamond Member
There's a lot of confusion about Intel's recent plans for their desktop platforms and there's been two new interesting leaks today. I think they deserve their own thread.

First leak comes from CPU-World (they have been spot on about a lot of unreleased Intel chips in the past), according to them Core i5/Core i7 Broadwell-K will be launched around the same time as multiplier locked Core i5/Core i7 versions of Skylake, both in Q2/2015. Skylake-based Core i3, Pentium & Celeron should be launched later, sometime in H2/2015.

Intel Skylake desktop CPUs to launch in Q2 2015

...According to Intel's desktop roadmap, Skylake desktop Core i5 and i7 microprocessors, using Skylake-S core, along with 100 series chipset will be launched in the second quarter 2015. Broadwell desktop i5/i7 CPUs will be also released in Q2 2015. What is interesting is that all unlocked processors at launch will be Broadwell based, and all locked parts will be built on Skylake-S core. At this time, Intel does not plan to transition Celeron, Pentium and Core i3 processors to Skylake architecture before second half of 2015.

On the low end, Intel is going to replace Celeron and Pentium "Bay Trail-D" desktop processors with Braswell-based chips in Q1 2015.
Click to expand...

Then there's Intel's Q1'2014-Q1'2015 desktop roadmap, the most recent leak by Chinese VR-Zone. According to them Broadwell-K got pushed to Q1'2015 and both Haswell-E and Devil's Canyon will arrive in the second half of this year (an year after first Haswell desktop chips).

Intel Q1-2014/Q1-2015 desktop roadmap

Intel-Devils-Canyon-Roadmap-635x426.jpg


While this was published today it could be based on older info than the May-2014 rumour (also from VR-Zone).

http://chinese.vr-zone.com/114150/i...roadwell-desktop-schedule-is-2015-q1-05262014

www.cpu-world.com/news_2014/2014052601_Intel_Skylake_desktop_CPUs_to_launch_in_Q2_2015.html
 
Last edited:
The Roadmap from vr-zone is outdated. It was planned for Q1, it's now scheduled for Q2. You should change your thread title. CPU-world obviously has a current Roadmap.
 
To me it still doesn't make sense that broadwell and skylake overlap on desktop. What is the point of broadwell on desktop if skylake will be superior design? Unlocked multiplier is just a gimmick and intel can unlock skylake parts if they want. Also why would broadwell have better iGPU performance (since I saw it mentioned somewhere)? Both are 14nm parts, why strap on more cache or pair beefier GPU with inferior core (Vs skylake).
 
Man, how confusing is this? Desktop Broadwell and Skylake to be released at the same time?

Why would Intel ever want to do something like that? Are ticks and tocks released simultaneously now? 😵
 
The other thing is that Skylake requires DDR4, and prices for DDR4 are supposed to be bad until the end of 2015. Unless that's changed, I can't imagine that's going to go over well with OEMs.
 
inf64 said:
To me it still doesn't make sense that broadwell and skylake overlap on desktop. What is the point of broadwell on desktop if skylake will be superior design? Unlocked multiplier is just a gimmick and intel can unlock skylake parts if they want. Also why would broadwell have better iGPU performance (since I saw it mentioned somewhere)? Both are 14nm parts, why strap on more cache or pair beefier GPU with inferior core (Vs skylake).
Click to expand...

They would overlap anyway if you think about Broadwell-E and Skylake, that already happens now (Ivy Bridge-E and Haswell), just not in the socket/platform. They could very well launch Broadwell-K as 'Devil's Canyon' successor with very high clocks, thus negating any Skylake (non-K) IPC advantage. Broadwell-K comes with GT3e graphics, probably faster than Skylake GT2 IGPs (even though it's based on a newer architecture).
Also, if Skylake-K is a quad-core + GT4e part it might not be ready to launch till H2 (Q4? or even later), that could mean half a year or more after Broadwell-K's launch.
 
Last edited:
Sweepr said:
According to Intel's desktop roadmap, Skylake desktop Core i5 and i7 microprocessors, using Skylake-S core, along with 100 series chipset will be launched in the second quarter 2015.
Click to expand...

Could they have mistaken "launched" for "started production"? I.e. it won't actually be released until ~1 year later, in 2016Q2? I.e. roughly the same time frame from start of production to actual availability as for Broadwell? That should match earlier info that has been leaked. Otherwise I don't see how this makes sense.

Also, what's Skylake-S by the way? Is it the low power version of desktop Skylake, similar to e.g. Haswell 4670S?
 
inf64 said:
To me it still doesn't make sense that broadwell and skylake overlap on desktop. What is the point of broadwell on desktop if skylake will be superior design? Unlocked multiplier is just a gimmick and intel can unlock skylake parts if they want. Also why would broadwell have better iGPU performance (since I saw it mentioned somewhere)? Both are 14nm parts, why strap on more cache or pair beefier GPU with inferior core (Vs skylake).
Click to expand...

As Sweepr pointed out, conceptually it is no different than what Intel already does in its market segmentation for the extreme platform versus mainstream in which enthusiasts have to choose between 6-cores of last year's microarchitecture versus 4-cores of this year's microarchitecture.

Now the same choice will have to be made on the mainstream platform as well. 4 unlocked cores of last year's architecture, or 4 locked cores of this year's architecture.

Its just more of the same old progression of market segmentation that Intel has been working hard to create over the years. Should come as no surprise, not that this will lessen the disappointment in enthusiasts though.

However the segmentation itself is not entirely arbitrary or artificial. Just as the extreme lineup entails chips that are themselves destined for the server and workstation lineup and so do take additional development time for validation and certification procedures (thus explaining the timeline gap between extreme and K models), there is a technical argument to be made regarding Intel doing the same on the mainstream desktop which is "reliability and overclocking".

The last thing Intel needs is to release a couple unlocked SKUs only to have the internet abuzz over a bunch of enthusiasts having burned up their 14nm Skylake processors from overvolting and overclocking. It takes time to develop the dataset necessary to quantify the risk of failure with a new node (intrinsic reliability) especially when pushing chips outside their specifications.

If 14nm is "delicate" in a way that 22nm is not, then Intel may be safeguarding its brand by pushing out the timeline for bad press over this once OC'ers start trashing their unlocked 14nm chips.

With Broadwell being nearly a year old by the time enthusiasts will get their hands on the K SKUs, Intel will have the data it needs to understand where the early death envelope lies. Skylake will be too early for them to have that data.

Or it could all just be a bunch of artificially created market segmentation on the mainstream desktop, might not be any good engineering-based reason for it at all. Time will tell, unfortunately it will take years for time to tell though 🙁
 
I think it just supports what some of us have suspected:
-CPU gains are pretty minimal, so are process
-Intel realized that shoddy OC capabilities of Haswell does not sell in the enthusiast(aka K series) market. So they must have quickly mustered the better K. That naturally means the next K comes a year later.

Thus, justifying the existence of both previous archtecture K, and current architecture non-K.
 
Last edited:
IntelUser2000 said:
I think it just supports what some of us have suspected:
-CPU gains are pretty minimal, so are process
-Intel realized that shoddy OC capabilities of Haswell does not sell in the enthusiast(aka K series) market. So they must have quickly mustered the better K. That naturally means the next K comes a year later.

Thus, justifying the existence of both previous archtecture K, and current architecture non-K.
Click to expand...
The 14nm process is being talked about, to some extent, at the VLSI Symposium in the second week of June. Intel's detailing a clock generator they built on 14nm CMOS, so perhaps they'll go over some process details. Of course, they've let me down the past, I don't know, dozen or so times they had a chance to detail the process, but maybe this time will be the one...

If not, at least I'll have a fun paper to read on IEEE Xplore.
 
Idontcare said:
If 14nm is "delicate" in a way that 22nm is not, then Intel may be safeguarding its brand by pushing out the timeline for bad press over this once OC'ers start trashing their unlocked 14nm chips.

With Broadwell being nearly a year old by the time enthusiasts will get their hands on the K SKUs, Intel will have the data it needs to understand where the early death envelope lies. Skylake will be too early for them to have that data.
Click to expand...

That makes a lot of sense.

Also I have to wonder if Intel is delaying Broadwell-K because the Haswell Refresh-K (Devil's Canyon) will also use a specific higher performance/higher leakage 22nm xtor from the big socket Haswell-E?

Then Intel releases 14nm (with safeguards for fragility as you mentioned.) Once this testing period and fragility has been determined, maybe Intel develops a more performance oriented/higher leakage desktop/server 14nm xtor for Broadwell-E and then shares this with Broadwell-K? At the same the company develops a more mobile oriented 14nm for Skylake (in the same way we saw the initial Haswell benefit from a more mobile oriented 22nm xtor compared to Ivy Bridge).

(Essentially I potentially see mobile getting prioritized on the mainstream socket in the beginning, but possibly sharing xtor technology from the big socket in the later half of its cycle)
 
Last edited:
How fast will these new CPUs be? Will they be compatible with a Z97 chipset? Reason I ask is because I'm in the process of building a new computer and I'm coming from a Q6600. I will be most likely buying a Z97 MOBO.

Edit- Ah crap. Just read a thread about Broadwell and it appears that CPU is DDR4. And here I'm coming from DDR2 to DDR3.
 
Last edited:
John Connor said:
How fast will these new CPUs be? Will they be compatible with a Z97 chipset? Reason I ask is because I'm in the process of building a new computer and I'm coming from a Q6600. I will be most likely buying a Z97 MOBO.

Edit- Ah crap. Just read a thread about Broadwell and it appears that CPU is DDR4. And here I'm coming from DDR2 to DDR3.
Click to expand...
Z97 Broadwell's DDR3. As far as how fast it is, no one knows. Intel's been keeping details for almost everything tightly under wraps.
 
Homeles said:
The 14nm process is being talked about, to some extent, at the VLSI Symposium in the second week of June.
Click to expand...

When I mean lower gains in the process side, I mean on the high performance variant. The mobile gains are really good.

will also use a specific higher performance/higher leakage 22nm xtor from the big socket Haswell-E?
Click to expand...

I am not sure if they are. The E chips are using relatively low clock speeds because they have lots of cores. And they have LOT of I/O portions. Leakage must be a priority. Their presentation suggests that server chips are using high performance/high leakage process, but the products don't suggest that's the case.
 
CHADBOGA said:
Is Skylake going to be DDR4 only?

Maybe they are going to run Broadwell for DDR3 and Skylake for DDR4.
Click to expand...
Theoretically speaking, you could build a memory controller that supports both. Intel may choose to do that with Skylake. I believe AMD is doing something similar with one of their upcoming releases.
 
So are the K variants going the Xtreme series route then? Competition from AMD cant come sooner in this segment, Intel is really coasting on this one.
 
Interesting, especially the fact that dual core are missing at launch: on Wikipedia it says quad core only as mainstream since long ago, before this leak, maybe the other specs are also true? Can't wait for some more info...
 
SAAA said:
Interesting, especially the fact that dual core are missing at launch: on Wikipedia it says quad core only as mainstream since long ago, before this leak, maybe the other specs are also true? Can't wait for some more info...
Click to expand...
I'd be thrilled if dual cores were finally starting to be phased out. I just don't see that happening yet, though.
 
Homeles said:
I'd be thrilled if dual cores were finally starting to be phased out. I just don't see that happening yet, though.
Click to expand...

I'm dreaming of a 3-core i3 since ages, let's upgrade those ugly core count and match the nomenclature! 😉
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top