Intel UHD Graphics 615 vs Intel UHD Graphics (Jasper Lake 32 EU) vs Intel Iris Plus Graphics G4 (Ice Lake 48 EU)

The Intel UHD Graphics 32 EUs (Jasper Lake, Gen. 11) is an integrated graphics card in the Jasper Lake Pentium SoCs for laptops and small desktops. It is the fastest GPU version in the Jasper Lake series (Pentium Silver) and offers all 32 EUs (Execution Units / Shader Blocks). It offers no dedicated graphics memory and the clock rate depends on the processor model. At launch there are two Pentium Silver models (N6000 with 6W and N6005 with 10W) clocked at 350 - 900 MHz.

Thanks to the improved architecture and more EUs, the UHD Graphics is clearly faster than the old Gemini Lake SoCs (e.g. N5000). Intel states up to 78% better 3DMark Fire Strike scores and up to 22% more FPS in Rocket League. Still only very low demanding games should be playable.

A special new feature of the Gen11 graphics card is the new Variable Rate Shading (VRS) support. With it game designers can decide where to spend shading time and e.g. shade object in the background or behind fog with less resolution (up to using only one source for a 4x4 block). With this technique early results show up to 1.3x performance in Unreal Engine POC and 1.2x speedup in Civ 6. Up to now VRS is only supported by the new Nvidia Turing architecture (GTX 1650 and up).

Another improved hardware piece is the integrated video de- and encoder that was improved significantly according to Intel. They did not specify any more details, but the previous generation was able to decode VP9 and H.265/HEVC in Main10 profile with 10 bit color depth using the dedicated hardware.

The Jasper Lake SoCs and therefore the integrated GPU are manufactured in the 10nm process (like Ice Lake probably) at Intel that should be comparable to the 7nm process of TSMC.

The Intel Iris Plus Graphics G4(Ice Lake with 48 EUs) is an integrated graphics card in the Ice Lake SoCs (10th generation core 10xxG4) for laptops. It offers no dedicated graphics memory (no eDRAM cache like the Intel Iris Graphics 655 predecessor of the Coffee Lake SoCs). It offers 48 of the 64 EUs (Execution Units / Shader Blocks) of the Ice Lake chip and the clock rate depends on the processor model. At launch it looks like all models have a 300 MHz base clock and only differ in the boost clock (1050 - 1100 MHz). Another difference is the TDP as it can be configured from 12 - 25 Watt in the 15 Watt U-models and maybe even higher in the 25 Watt models that will come later.

The performance was not yet communicated, but due to the 25% less cores and similar clock speeds the performance should be clearly lower than the full Iris Plus G7 models. Although, the Ice Lake G4 graphics does not feature any dedicated graphics memory it should perform better than the old Iris Plus 655 with 128 MB eDRAM. Compared to AMDs offering, we suspect the performance should be similar to an AMD Radeon RX Vega 8 GPU. Therefore, the gaming performance should be only sufficient for low demanding games like Rocket League or Overwatch.

A special new feature of the Gen11 graphics card is the new Variable Rate Shading (VRS) support. With it game designers can decide where to spend shading time and e.g. shade object in the background or behind fog with less resolution (up to using only one source for a 4x4 block). With this technique early results show up to 1.3x performance in Unreal Engine POC and 1.2x speedup in Civ 6. Up to now VRS is only supported by the new Nvidia Turing architecture (GTX 1650 and up).

Another improved hardware piece is the integrated video de- and encoder that was improved significantly according to Intel. They did not specify any more details, but the previous generation was able to decode VP9 and H.265/HEVC in Main10 profile with 10 bit color depth using the dedicated hardware.

The Iris Plus G4 supports three display pipes that can each output 5K60 signals (via DisplayPort 1.4 HBR3 or HDMI 2.0b). Combining two pipes, the chip is capable to output 8k content.

The Ice Lake SoCs and therefore the integrated GPU are manufactured in the modern 10nm+ process at Intel that should be comparable to the 7nm process of TSMC.