Intel Iris Xe Graphics G7 80EUs vs NVIDIA GeForce 910M vs Intel UHD Graphics 32EUs (Alder Lake)

The Intel Xe Graphics G7 80EUs (Tiger-Lake U GPU with 80 EUs) is a integrated graphics card in the mid range Tiger-Lake U CPUs (15 - 28 Watt, e.g. Core i5 11th Gen CPUs). It is using the new Xe architecture (Gen12) and should be announced later in 2020.

The performance depends on the TDP settings of the laptop and the used cooling. First informations show that the chip can be configured at 15 and 28 Watt TDP default (as the Ice Lake-U chips) and the performance should be around a dedicated GeForce MX330 in 3DMark benchmarks. For gaming we are expecting a bit worse performance due to the missing dedicated graphics memory and driver support. Compared to the older Iris Plus G4 in Ice Lake chips, the Tiger Lake GPU should be nearly twice as fast. Benchmarks with various laptops show that the 80EU version lacks about 18% behind the faster 96EU version and on par with the MX330. Compared to AMD iGPUs, the G7 80EU positions itself around the Vega 7 and Vega 8 chips (see chart below for details).

Therefore, the iGPU is still only for lowest graphical settings and low resolutions in demanding games. Some modern games like Days Gone or Cyberpunk wont run over 30 fps in minimum settings (see below).

The Tiger Lake SoCs and therefore the integrated GPU are manufactured in the modern 10nm+ (10nm Superfin) process (improved 10nm process) at Intel and therefore should offer a very good efficiency.

The NVIDIA GeForce 910M is an entry-level DirectX 12 compatible graphics card announced in late 2015. Its core is based on the 28nm GF117 chip (Fermi architecture) and is equipped with 64-bit DDR3 memory. The 910M is a renamed GeForce 820M, which already was based on the the old GT 620M, GT 720M and GeForce 710M.

Architecture

The GF117 is based on the optimized GF108 Fermi chip (GeForce GT 540M) and offers 96 shaders, 16 TMUs and 4 ROPs. Each shader core is clocked twice as fast as the rest of the graphics chip, a technique known as hot clocking. More detailed information on Fermi can be found on the GT 435M GPU page.

It should be noted that the GF117 does not offer dedicated graphic ports and can therefore only be used in conjunction with Optimus.

The 820M supports GPU Boost 2.0, which can automatically overclock the card if the laptop cooling system allows it.

Performance

The gaming performance should be identical to the old GeForce 820M (see this page for benchmarks and game tests) and therefore also GT 720M and the Intel Iris Graphics 5100. This places the card in the entry level segment. Only old or low demanding games run fluent on this graphics card.

Features

The shader cores (also called CUDA cores) can be used for general calculations with APIs such as CUDA, DirectCompute 2.1 and OpenCL. PhysX is theoretically possible, but the 910M is too slow to handle both PhysX and 3D rendering in modern games. 3D Vision is not supported according to Nvidia.

Power consumption

The power consumption of the Geforce 910M should be similar to the old 820M and GT 720M. As a result, the GPU is best suited for notebooks 13 inches in size or greater. The efficiency however cant be compared to a modern Maxwell based GPU like the GeForce 920MX.