One of the most surprising product announced in 2017 was Intel’s 8th Generation processor with AMD Radeon Graphics, which is an unusual combination between two competitors. KabyLake G-series, which was announced back in 2017, is an excitement of 2018. Followers from blue team and red team are eagerly waiting for this processor.
Intel 8th Generation processor with Radeon RX Vega M Graphics having CPU part is Intel Kaby Lake-H-a higher version of Intel Kaby Lake R chips. The CPUs are paired with either a Radeon RX Vega M GH or a Radeon RX Vega M GL GPU, which in turn is connected to 4GB of second-generation High Bandwidth Memory (HBM2).
In total, Intel is launching five variants of Kaby Lake G-series with AMD Graphics. Four of them are i7 and other one is i5 branded.
Intel i7 8809G will be the high end from the variants. It clocks at 4.2 GHz boost frequency and 3.1 base frequency, also having 4-core/8-threads configuration. It consisting of 24 Compute units in all 1536 stream processors. Most of this processors are targeted at gaming and some enterprise solutions.
KabyLake G-series Micro Architectural changes
This changes are based on cost, performance and efficiency trade-offs, which will be driving from best of both the worlds. The GPU is joined to CPU with 8-PCI-E 3.0 Lanes, and another 8 lanes are joined with other devices. Now the inter-connectivity in the CPU replaces traditional copper traces and connects with Intel’s Embedded Muti-Die Interconnect Bridge. It is basically a high speed, short range connection for transferring data between CPU and GPU.
Why collaboration between Intel and AMD?
KabyLake G-series Intel processors doesn’t required increase in more cores as the value would be tremendously high, so faster GPU would have been a rise in performance for mobile users. Also, Intel is planning for Discrete Graphics. Intel also have now AMD’s Radeon Graphics Head ” Raja Koduri “, so in coming years, there might be increase in Graphics performance from Intel side.
So, Intel’s first step to Heterogeneous computing.
The CPU in the new chips is subject to an overall TDP of 65W or 100W