Intel UHD Graphics 605 vs NVIDIA GeForce GTX 860M vs Intel Iris Graphics 550

Intel UHD Graphics 605

The Intel UHD Graphics 605 is an integrated processor graphics unit from the Gemini Lake generation (e. g. Pentium Silver N5000), which was announced late 2017. The GPU can convince with low consumption figures, but the performance is only in the low-end segment and is rarely sufficient for modern games. Being the maximum configuration of the Gemini Lake GPU, the UHD Graphics 605 is equipped with 18 Execution Units (EUs) running at up to 750 MHz. The technical specifications of the GPU are based on Intel's Gen9 architecture, which is also used for the more expensive Skylake series and also in the predecessor HD Graphics 505.

Due to its lack of dedicated graphics memory or eDRAM cache, the UHD 605 has to access the main memory via processor (2x 64bit DDR4/LPDDR4-2400).

Performance

The UHD Graphics 605 offers a similar performance as the Intel HD Graphics 505 in the Apollo Lake SoCs (same Gen 9 architecture). Therefore, the GPU is only suited for some very low demanding games like Dirt Rally or Sims 4.

Features

The video engine decodes H.265/HEVC (8b and 10b) and VP9 completely in hardware. Up to three displays can be connected via DP 1.2a/eDP 1.3 or HDMI 2.0a (including HDCP 2.2).

Power Consumption

The TDP of the whole Pentium Silver N5000 chip is 6 Watts by default, so the GPU is primarily used for (often passively cooled) netbooks or thin notebooks. The TDP can also be reduced to 4.9 Watts, but this will obviously affect the performance.

NVIDIA GeForce GTX 860M

The NVIDIA GeForce GTX 860M is an upper mid-range, DirectX 12-compatible graphics card for laptops unveiled in March 2014. It is one of the first cards based on Nvidia's new Maxwell architecture (GM107 chip), but is still manufactured in 28 nm. The GTX 860M offers 640 shader units clocked at 1029 MHz (+ Boost) as well as 4 GB of GDDR5 memory (128 bit, 5000 MHz effective).

Alternatively, there is also a version based on the Kepler chip GK104, which has 1152 shader units clocked at 797 MHz.

Architecture

Compared to Kepler, Maxwell has been optimized in several details to increase power efficiency. Smaller Streaming Multiprocessors (SMM) with only 128 ALUs (Kepler: 192) and an optimized scheduler should lead to better utilization of the shaders. Nvidia promises that a Maxwell SMM with 128 ALUs can offer 90 percent of the performance of a Kepler SMX with 192 ALUs. GM107 features 5 SMMs and thus 640 shader cores, 40 TMUs and 16 ROPs (128-bit interface).

Another optimization is the massively enlarged L2 cache (GM107: 2 MB). The increased size can handle some of the memory traffic to allow for a relatively narrow memory interface without significantly reducing the performance.

Similar to Fermi and Kepler, Maxwell / GM107 supports DirectX 11.2 with feature level 11.0 only.

Performance

Although the GTX 860M has much less memory bandwidth and computing power, it can compete with (or sometimes even beat) the old Kepler-based GeForce GTX 770M while drawing significantly less power. Many games of 2013/2014 (Tomb Raider or BioShock Infinite, etc.) can be played fluently in FullHD resolution and very high detail settings. However, very demanding games such as Crysis 3 or Battlefield 4 will require lower resolutions and/or details.

Despite its considerably higher shader count, the Kepler version is about 10 percent slower, mainly due to the worse bandwidth efficiency.

Features

GM107 integrates the sixth generation of the PureVideo HD video engine (VP6), offering a better decoding performance for H.264 and MPEG-2 videos. Of course, VP6 supports all features of previous generations (4K support, PIP, video encoding via NVENC API). Unfortunately, HDMI 2.0 is still not supported.

Power Consumption

The power consumption of the GeForce GTX 860M should be similar to the old GTX 765M (about 60 watts). The low-clocked Kepler version draws just slightly more power. Therefore, the 860M is best suited for 15-inch laptops or larger. The 800M series also supports Optimus to automatically switch between an integrated graphics card and the Nvidia GPU.

Intel Iris Graphics 550

The Intel Iris Graphics 550 (GT3e) is an integrated graphics unit of the Skylake generation. As successors of the Iris Graphics 6100 (Broadwell), the Iris Graphics 550 can be found in CPUs of 28-watt series and offers 64 MB of dedicated eDRAM memory. Furthermore, the so-called GT3e-version features 48 Execution Units (EUs) clocked at up to 1100 MHz (depending on the CPU model). Besides the eDRAM cache, the Iris 550 is able to access the main memory (2x 64bit DDR3L-1600 / DDR4-2133).

Compared to the Iris Graphics 540 of the 15-watt series, the Iris Graphics 550 differs only by its slightly higher maximum clock and a nearly doubled TDP, which allows a better utilization of the Turbo Boost.

Performance

The exact performance of the Iris Graphics 550 depends on various factors like L3 Cache size, memory (DDR3/DDR4) and maximum clock rate of the specific model. The fastest versions (Core i7-6567U) may compete with a dedicated GeForce 930M and will handle modern games (as of 2015) in low or medium settings.

Features

The revised video engine now decodes H.265/HEVC completely in hardware and thereby much more efficiently than before. Displays can be connected via DP 1.2 / eDP 1.3 (max. 3840 x 2160 @ 60 Hz), whereas HDMI is limited to the older version 1.4a (max. 3840 x 2160 @ 30 Hz). However, HDMI 2.0 can be added using a DisplayPort converter. Up to three displays can be controlled simultaneously.

Power Consumption

The Iris Graphics 550 can be found in mobile processors specified at 28 W TDP and is therefore suited for medium-sized laptops and ultrabooks.