Intel Celeron 3867U vs AMD A12-9700P vs Intel Core i7-7920HQ

Intel Celeron 3867U

The Intel Celeron 3865U is an ULV (ultra low voltage) dual-core SoC based on the Kaby-Lake architecture and has been launched in the first quarter of 2017. The CPU can be found in ultrabooks as well as normal notebooks. In addition to two CPU cores clocked at 1.8 GHz (no Turbo Boost, no HyperThreading), the chip also integrates an HD Graphics 610 GPU and a dual-channel DDR4-2133/DDR3L-1600 memory controller. The SoC is manufactured using a 14 nm process with FinFET transistors.

Compared to the similar Celeron 3865, the 3867 offers a different cTDP-down option (12.5 versus 10 Watt).

Architecture

Intel basically uses the same micro architecture compared to Skylake, so the per-MHz performance does not differ. The manufacturer only reworked the Speed Shift technology for faster dynamic adjustments of voltages and clocks, and the improved 14nm process allows much higher frequencies combined with better efficiency than before.

Performance

Due to the missing Turbo Boost and the low clock speeds, especially the single thread performance is very limited which results in a lower performance even for lower demanding tasks. The performance should be noticeably slower than the Celeron 3965 which offers 400 MHz higher clocked CPU cores. Therefore, the CPU is only suited for entry level tasks like office, web surfing and multimedia.

Graphics

The integrated graphics unit called HD Graphics 610 (similar to the HD Graphics 510) represents the "GT1" version of the Kaby Lake GPU (Intel Gen. 9). Its 12 Execution Units, also called EUs, are clocked at 300 - 900 MHz and offer a performance somewhat below the older HD Graphics 4400. Only a few games of 2015 can be played smoothly in lowest settings.

Power Consumption

Specified at a TDP of 15 W (including CPU, GPU and memory controller), the CPU is best suited for small notebooks and ultrabooks (11-inches and above). Optionally, the TDP can be lowered to 10 watts (cTDP down), reducing both heat dissipation and performance and allowing even more compact designs.

AMD A12-9700P

The AMD A12-9700P is a mainstream SoC from the Bristol-Ridge APU series for notebooks (7th APU generation), which was announced mid 2016. The ULV chip with a TDP of 15 Watts (can be configured to 12 Watts) has four CPU cores (two Excavator modules), a Radeon R7 GPU as well as a dual-channel DDR4-1866 memory controller. Carrizo is a full-fledged SoC and is also equipped with an integrated chipset, which provides all I/0 ports.

Architecture

Bristol Ridge is the successor of the Carrizo architecture and the design is almost identical. Thanks to optimized manufacturing processes and more aggressive Boost behavior, however, the clocks are a bit higher at the same power consumption. The memory controller now also supports DDR4-RAM, in this case up to 1866 MHz. More technical details are available in the following articles:

• AMD's Bristol Ridge and Stoney Ridge Architecture
• AMD: Bristol Ridge APUs announced

Performance

The performance of the A12-9700P is between the two 15-Watt processors A10-9600P and FX-9800P, and therefore roughly on par with a 15-Watt Core i3 from the Skylake or Kaby Lake series, respectively. Compared to the Intel model, the AMD chip has a small advantage in multi-thread scenarios, but is beaten when you only stress one or two cores. Due to the TDP limitation, the performance of the A12-9700P will drop significantly under sustained workloads.

This means there is sufficient performance for typical office and web applications as well as light multitasking.

Graphics Card

The integrated Radeon R7 (Bristol Ridge) GPU has 512 active shader units (8 compute cores) clocked at up to 758 MHz. Thanks to the better utilization of the clock range as well as faster DDR4-RAm, the GPU can slightly beat its predecessors Radeon R7 (Carrizo) and competes with a dedicated GeForce 920MX in the best-case scenario (dual-channel memory, low CPU requirements). Many games from 2015/2016 can be played smoothly at low settings. 

Power Consumption

AMD specifies the TDP of the A12-9700P with 15 Watts, which is comparable to Intel's ULV models. This means the CPU is a good choice for thin notebooks starting with a 12-inch screen.

Intel Core i7-7920HQ

The Intel Core i7-7920HQ is a fast quad-core processor for notebooks based on the Kaby Lake architecture and was announced in January 2017. It is the top model of the consumer series at the time of the announcement. Only the Xeon E3-1535M v6 is faster (100 - 200 MHz higher Turbo clocks). Besides four cores including Hyper-Threading support running at 3.1-4.1 GHz (4 cores up to 3.7 GHz, 2 cores up to 3.9 GHz), the processor is also equipped with the HD Graphics 630 GPU as well as a dual-channel memory controller (DDR3L-1600/DDR4-2400). It is manufactured in a 14nm process with FinFET transistors.

Architecture

Intel basically uses the same micro architecture compared to Skylake, so the per-MHz performance does not differ. The manufacturer only reworked the Speed Shift technology for faster dynamic adjustments of voltages and clocks, and the improved 14nm process allows much higher frequencies combined with better efficiency than before.

Performance

The performance is a bit higher compared to the Skylake predecessor Core i7-6920HQ (2.9-3.8 GHz) thanks to slightly higher clocks (up to 7.9%), so it is the fastest mobile consumer processor in the beginning of 2017 (mobile Xeon E5-1535M v6 has even higher Turbo clocks). The performance is sufficient even for very demanding tasks and on par with powerful desktop quad-core processors.

Graphics

The integrated Intel HD Graphics 630 has 24 Execution Units (similar to previous HD Graphics 530) running at 350 - 1100 MHz. The performance depends a lot on the memory configuration; it should be comparable to a dedicated Nvidia GeForce 920M in combination with fast DDR4-2133 dual-channel memory.

Contrary to Skylake, Kaby Lake now supports hardware decoding for H.265/HEVC Main 10 with a 10-bit color depth as well as Google's VP9 codec. The dual-core Kaby Lake processors, which were announced in January, should also support HDCP 2.2.

Power Consumption

The chip is manufactured in an improved 14nm process with FinFET transistors, which improves the efficiency even further. Intel still specifies the TDP with 45 Watts, but it can also be reduced to 35 Watts by the notebook manufacturers (cTDP down). This will obviously affect the performance, because the Turbo Boost cannot be maintained for longer periods.