• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

Intel’s 10nm Cannonlake Delayed to 2H 2017

DmitryD

DmitryD

DmitryD

Associate
Joined
5 Sep 2014
Posts
289
Intel’s 10nm Cannonlake Delayed to 2H 2017 – Kaby Lake Y/U/H Series in Q3/Q4 2016, Kaby Lake-S Desktop Chips in Q1 2017

For people who were planning to upgrade to Intel’s latest 10nm architecture codenamed Cannonlake, there’s bad news from Benchlife as the processor family has been delayed to second half of 2017. Cannonlake which was previously pushed to early 2017 will now be moving several months back from its launch date while Kaby Lake will act as the gap filler family to offer desktop and mobility users an upgrade path on existing platforms.

More from the source:

http://wccftech.com/intels-10nm-can...ies-q3-2016-kaby-lakes-desktop-chips-1h-2017/

----

Looks like I'll stick with my 2500K until 2018...
 
DmitryD said:
Intel’s 10nm Cannonlake Delayed to 2H 2017 – Kaby Lake Y/U/H Series in Q3/Q4 2016, Kaby Lake-S Desktop Chips in Q1 2017

For people who were planning to upgrade to Intel’s latest 10nm architecture codenamed Cannonlake, there’s bad news from Benchlife as the processor family has been delayed to second half of 2017. Cannonlake which was previously pushed to early 2017 will now be moving several months back from its launch date while Kaby Lake will act as the gap filler family to offer desktop and mobility users an upgrade path on existing platforms.

More from the source:

http://wccftech.com/intels-10nm-can...ies-q3-2016-kaby-lakes-desktop-chips-1h-2017/

----

Looks like I'll stick with my 2500K until 2018...
Click to expand...

Plenty of time to save up then hopefully we will actually see a bump in performance then. :D
 
ukxenon said:
Agree time to save and now we can see what zen will offer
Click to expand...

Problem i can see with Zen is it's trying to catch up all the lost years, Intel has widened the gap every gen even if it hasn't made a big difference with each generation. I hope AMD put some real pressure on Intel but i doubt it unfortunately.
 
Dinkleberg said:
Problem i can see with Zen is it's trying to catch up all the lost years, Intel has widened the gap every gen even if it hasn't made a big difference with each generation. I hope AMD put some real pressure on Intel but i doubt it unfortunately.
Click to expand...

I am more worried that AMD has to deal with their subpar foundry partner Globalfoundries,who seem to delay and have issues with anything they do.

Samsung already has 14NM chips out and Globalfoundries can't even get any out yet in quantity after LICENSING the SAME technology and getting a whole load of IBM people too.

If they can actually get working chips out,at least AMD will fighting Intel with 14NM chips when they have 14NM chips otherwise it will be another case of Intel being a process node ahead which means if Zen is good it will have one hand tied behind its back.
 
Nutella33 said:
Latest rumours are AMD are both working with GF and TSMC. Not sure they are just tried to GF
Click to expand...

I really hope for their sake,that they have a backup with TSMC.

The subpar foundry will kill AMD otherwise. Look at the last few process nodes:
1.)32NM delayed and had problems with clockspeed misses with proven tech like Llano
2.)28NM MASSIVELY delayed and had problems with clockspeed misses with Kaveri
3.)20NM was a failure
4.)14NM already delayed

For instance,Kaveri not only launched with significantly lower clockspeeds but was delayed massively too,from a 2013 release date to 2014.

While the subpar foundry keeps delaying it means AMD has to keep selling existing products which are not competitive in CPU performance when compared to Intel and their IGP advantage gets eaten away,which means they need to keep righting down inventory, due to WSA commitments,which means they are wedded to usng GF who keep ****** up.

Part of the fab selloffs,was the WSA,which means they HAVE to use GF for the next few years too. Some of the former management of AMD really balsed up on it.

But it doesn't stop them from using TMSC,so I really hope they have some contingency in place.

It worries me if they can hit their clockspeed targets using GF and more important everytime the crap company delays it means instead of 14NM products from Intel,AMD will be fighting 10NM ones.
 
Last edited:
If they are using GF 14nm it will be a custom process like 28SHP was. The low-power ones which Samsung designed do not allow the characteristics needed for high-perf chips.

Been trying to beat this into peoples' heads but wccf keep running articles about LPP this LPP that... that's the only publicly available info so they just assume and run with it like the hacks they are.

They said they have taped out, about 6 months after "volume ramp" early in the year. Guess we will see who they end up using next year. Maybe the Fujitsu deal was to pay for the extra tapeout to hedge against GF fails.
 
Last edited:
Intel probably won't put out anything faster until 7nm new arch in 2021.

2016 - nowt
2017 - worthless refresh ("Kaby Lake"), then 10nm shrink
2018 - another mediocre uninspired arch ("Icelake")
2019 - refresh of mediocre arch
2020 - 7nm shrink of mediocre arch

2021 - Intel finally have to do something. If AMD still around and fighting, that is.
 
Kaby Lake is basically gonna be +100 MHz and slightly better IGPs compared to Skylake, right?

AMD have a big opportunity here. If they blow it, it's gonna suck for all of us.
 
7nm is possibly about the limit. They might be able to crack 5nm. But anything smaller and thats about it.

I think by then it will be more refining the transistors and such. Maybe it will be time to use the AMD approach and start using "moar coars" :D. Or more and more transistors.

AMD im not sure. I think they are just about stuck already.
 
Last edited:
That's a good point, I've been wondering about this on-and-off. At some point, they're going to reach a limit with how small they can go, and once they've reached it, then what do they do?

Eventually, they're going to reach the point where they can't physically make the things any smaller.
 
AnthonyA85 said:
That's a good point, I've been wondering about this on-and-off. At some point, they're going to reach a limit with how small they can go, and once they've reached it, then what do they do?

Eventually, they're going to reach the point where they can't physically make the things any smaller.
Click to expand...

They can go much smaller than this again.

Silicon atoms are 0.2nm. Though it may take another 50 years to get there, and by then a new material will have been used most likely.
 
You know somethings gone very wrong when socket 1366 is still a viable platform. I don't ever recall there being such small increments in performance til these last few years. Remember the jump from p4 to core2 or athlon xp to 64, that seemed like night and day!
 
davida76 said:
You know somethings gone very wrong when socket 1366 is still a viable platform. I don't ever recall there being such small increments in performance til these last few years. Remember the jump from p4 to core2 or athlon xp to 64, that seemed like night and day!
Click to expand...

I built my dad an Athlon XP system - I think it was the best one of that line (a Barton 3200+), and that ended up being obsolete rather quickly. Bought it just before the advent of things like dual-core, 64-bit, SATA, etc. In comparison, X58's lack of PCIe 3, USB 3, SATA 6 Gb/s, M.2, NVMe, etc. is nowhere near as big of an issue as the above was, especially since consumer SSDs using AHCI on SATA 3 Gb/s are still great.

I was debating for a while whether to go with a Core i7-920 or a Core 2 Q9450 (IIRC) when building my current system and in the end decided to spend more on newer tech. Boy am I glad I did! :D
 
davida76 said:
You know somethings gone very wrong when socket 1366 is still a viable platform. I don't ever recall there being such small increments in performance til these last few years. Remember the jump from p4 to core2 or athlon xp to 64, that seemed like night and day!
Click to expand...

Viable if you want to buy a second hand Xeon only, and miss out on all the new fun stuff, such as USB3, USB3.1, Sata3, PCI-E v3, M.2, UEFI fan control, Sata Express, DDR4 speed/capacity, and a cooler, quiet, power efficient system (1366 xeons power consumption when overclocked is crazy)
 
You must log in or register to reply here.
Back
Top Bottom