Intel Planning To Bring 14nm Coffee Lake 6 Core Chips in 2018 on Mainstream PCs – To Coexist With 10nm Cannonlake Processors
Intel is planning to launch their first 6 core mainstream processors in 2018 with a new family of CPUs known as Coffee Lake. Based on the 14nm node, Coffee Lake will be the first mainstream processor family which will see a bump in the total number of cores since their first Core 2 Extreme era processors. While Intel’s HEDT lineup has seen a significant increase in cores, it’s finally time that Intel has decided to bump up the number on their main stream family.
Intel’s 14nm Coffee Lake and 10nm Cannonlake CPUs would coexist in 2018. (Image Credits: PCWatch)
Intel’s 14nm Coffee Lake Family To Bring 6 Core Processors on Mainstream PCs – Initially Aiming H-Series SKUs in 2018
Just in case you are worried about a potential delay of 10nm chips, then there’s no need to worry. The roadmap posted by PCWatch is very clear and shows 14nm Kaby Lake (Tock+) being replaced by 10nm Cannonlake (Tick) in second half of 2017. Cannonlake would first launch in two variants initially, the Y-Series chips featuring 2 core modules and GT2 graphics featured in Core M processors and the U-Series chips that also feature 2 cores along with GT2 graphics.
The difference between these chips will be clock speeds and TDPs. While both are based on 10nm Cannonlake architecture, the Y-Series SKUs have TDPs under 5W and branded as Core M processors. The U-Series processor has TDP ranging for 15-28W and is branded under the regular Core i7 / Core i5 processors. The U-Series chips are found in several laptops and notebook designs while the Core M series is aimed at Ultrabooks and ultra low power devices.
Intel killed off its Tick-Tock strategy in favor of three year process technology cadence!
Now here’s the interesting part, we know that Cannonlake is based on a entirely new process node and is going to launch in late mid-2017 but Intel plans to keep on updating their 14nm family of processors. We considered Kaby Lake to be the last 14nm family of CPUs but that doesn’t seem to be the case. In fact, Intel will extend the 14nm process to at least one more family in 2018, a year after their Kaby Lake launch. This family will be known as Coffee Lake.
Intel’s Coffee Lake 14nm CPUs Will Ship in Y-Series and U-Series SKUs
The 14nm Coffee Lake family has been referenced in a memo from Intel. While they haven’t been discussed publicly, it seems like a roadmap detailing the new family has appeared on the web. The memo we are talking about is from Intel’s new President, Murthy Renduchintala, whose memo listed a total of six products, all still in development (via Oregonlive):
- Kaby Lake, the code name for a new microprocessor due to begin production late this year using Intel’s current 14-nanometer technology.
- Cannon Lake, Intel’s first 10nm microprocessor, due late in 2017.
- Ice Lake, the second generation of 10nm technology
- Intel’s forthcoming 7560 baseband modem, a future generation of mobile wireless technology.
- The memo referenced two other products – Coffee Lake and Glenview – that Intel hasn’t discussed publicly.
We know much about Kaby Lake and Cannonlake chips which are the upcoming family of processors from the silicon giant. Ice lake and Tiger Lake which are part of Intel’s next-gen 10nm processors have also shown up in several leaks but this is the first time we are getting detailed information on Coffee Lake.
Intel’s 14nm and 10nm products have faced numerous delays due to difficulties involved in fabricating smaller process nodes.
According to the roadmap, Coffee Lake will make use of the most advanced 14nm process node which will be optimized beyond Kaby Lake series. These chips will co-exist with 10nm Cannonlake processors and offered in both U-Series and U-Series SKUs. We are looking at 6 core chips with GT3e graphics in the H-Series line up and up to 2 cores in the U-Series lineup with GT3e graphics.
Giving a more closer look, we can see what’s happening here, Coffee Lake appears to feature GT3e graphics while Cannonlake series offers GT2 graphics. The GT3e graphics chips means that these will feature eDRAM cache for higher on-chip bandwidth which will allow faster graphics and compute processing capabilities on laptops and low-power devices. The H-series lineup is the more performance oriented lineup aimed at high performance notebooks and with the Multi-Core Era upon us with the emergence of low-level APIs such as DirectX 12 and Vulkan, we can finally see good scaling on mobility platforms.
Coffee lake sounds great but why Intel is launching two different families at the same time still needs to be answered. It is mentioned that Intel has had much experience with 14nm process node and considering that the 10nm process won’t have great yields when it’s available in 2018, it’s better to built large chip dies on an existing yet matured and optimized process node which is 14nm.
Intel’s Cannonlake would be the first CPU family to feature 10nm process node when it arrives in 2017.
The only chips that Intel will be offering in 2018 in Cannonlake generation will start off with U and Y-Series SKUs which are small dies consisting of GT2 graphics and dual cores. The GT3e graphics with embedded DRAM that serves as L4 cache and use of up to 6 core variants requires good yields which won’t be the case with 10nm initially. Even as of right now, Intel has yet to offer Skylake quad core chips with GT4e graphics although its been almost a year since they were announced. 10nm and 14nm would co-exist for the mainstream family in 2018 but merged together with 10nm Ice lake family which arrives in late 2018 or early 2019.
It isn’t confirmed if Intel plans to offer Coffee Lake 6 core chips on mainstream desktop platforms but there’s a high possibility that would launch such chips on LGA sockets. For more information on Kaby Lake-S and Kaby Lake-X processors, visit this link. We will have more information on Cannonlake and Coffee Lake processors as their production commences.
Intel CPU Generation Comparison:
|Intel Sandy Bridge Platform||Intel Ivy Bridge Platform||Intel Haswell Platform||Intel Broadwell Platform||Intel Skylake Platform||Intel Kaby Lake Platform||Intel Coffee Lake Platform||Intel Cannonlake Platform|
|Processor Architecture||Sandy Bridge||Ivy Bridge||Haswell||Broadwell||Skylake||Kaby Lake||Coffee Lake||Cannonlake|
|Processors Cores (Max)||4||4||4||4||4||4||6||TBA|
|Platform Chipset||6-Series “Cougar Point”||7-Series “Panther Point”||8-Series “Lynx Point”||9-Series “Wild Cat Point”||100-Series “Sunrise Point”||200-Series “Union Point”||300-Series "Cannon-Lake PCH"||TBA|
|Platform Socket||LGA 1155||LGA 1155||LGA 1150||LGA 1150||LGA 1151||LGA 1151||TBA||TBA|
|Platform||Desktop LGA||Desktop LGA||Desktop LGA||Desktop LGA||Desktop LGA||Desktop LGA||Desktop LGA||Desktop LGA|