Intel Graphics Technology

Intel Graphics Technology (GT) is a series of integrated graphics processors (IGPs) designed by Intel. The graphics processor is integrated into the processor package, either on the same die as the central processing unit (CPU) or on a separate die or tile. The series was introduced in 2010 as Intel HD Graphics, succeeding Graphics Media Accelerator (GMA). Intel introduced the Intel UHD Graphics name in 2017.

Intel has also used the Iris, Iris Pro, Iris Plus and Iris Xe brands for integrated graphics. Some Iris Pro and Iris Plus models include eDRAM. With Meteor Lake and subsequent Intel Core Ultra processors, integrated GPUs are sold under the Intel Graphics and Intel Arc brands.

Intel Xe architectures are used in both integrated and discrete GPUs. The Intel Arc name therefore does not by itself distinguish a processor's integrated graphics from a separate graphics card. Integrated Arc products include the 130V and 140V in Lunar Lake, the 130T and 140T in Arrow Lake-H, and the B370 and B390 in Panther Lake.

History

Before the introduction of Intel HD Graphics, Intel integrated graphics were built into the motherboard's northbridge, as part of the Intel's Hub Architecture. They were known as Intel Extreme Graphics and Intel GMA. As part of the Platform Controller Hub (PCH) design, the northbridge was eliminated and graphics processing was moved into the processor package.

The previous Intel integrated graphics solution, Intel GMA, had a reputation of lacking performance and features, and therefore was not considered to be a good choice for more demanding graphics applications, such as 3D gaming. The performance increases brought by Intel's HD Graphics made the products competitive with integrated graphics adapters made by its rivals, Nvidia and ATI/AMD.

Generations

Earlier Intel HD and Iris Graphics are divided into generations and, within a generation, into tiers denoted by the 'GTx' label. Each generation corresponds to the implementation of a Gen graphics microarchitecture with a corresponding GEN instruction set architecture since Gen4.

Later Xe-based products use architecture names such as Xe-LP, Xe-LPG, Xe2 and Xe3. Xe-cores contain vector engines, but the number and capabilities of these engines vary between architectures.

Gen5 architecture

Westmere In January 2010, Clarkdale and Arrandale processors with Ironlake graphics were released, and branded as Celeron, Pentium, or Core with HD Graphics. There was only one specification: 12 execution units, up to 43.2 GFLOPS at 900 MHz. It can decode a H.264 1080p video at up to 40 fps.

Its direct predecessor, the GMA X4500, featured 10 EUs at 800 MHz, but it lacked some capabilities.

Model number Execution units Shading units Base clock (MHz) Boost clock (MHz) GFLOPS (FP32) HD Graphics 12 24500 900 24.0–43.2

Gen6 architecture

Sandy Bridge In January 2011, the Sandy Bridge processors were released, introducing the "second generation" HD Graphics: Model number Tier Execution units Boost clock (MHz) Max GFLOPS FP16 FP32 FP64 HD Graphics GT1 6 1000 192 96 24 HD Graphics 2000 1350 259 129.6 32 HD Graphics 3000 GT2 12 1350 518 259.2 65 HD Graphics P3000 GT2 12 1350 518 259.2 65

Sandy Bridge Celeron and Pentium have Intel HD, while Core i3 and above have either HD 2000 or HD 3000. HD Graphics 2000 and 3000 include hardware video encoding and HD postprocessing effects.

Gen7 architecture

Ivy Bridge On 24 April 2012, Ivy Bridge was released, introducing the "third generation" of Intel's HD graphics: Model number Tier Execution unitsShading units Boost clock (MHz) Max GFLOPS (FP32) HD Graphics [Mobile] GT1 648 1050 100.8 HD Graphics 2500 1150 110.4 HD Graphics 4000 GT2 16128 1300 332.8 HD Graphics P4000 GT2 16128 1300 332.8

Ivy Bridge Celeron and Pentium have Intel HD, while Core i3 and above have either HD 2500 or HD 4000. HD Graphics 2500 and 4000 include hardware video encoding and HD postprocessing effects.

For some low-power mobile CPUs there is limited video decoding support, while none of the desktop CPUs have this limitation. HD P4000 is featured on the Ivy Bridge E3 Xeon processors with the 12X5 v2 descriptor, and supports unbuffered ECC RAM.

Gen7.5 architecture

Haswell

In June 2013, Haswell CPUs were announced, with four tiers of integrated GPUs:

Model number Tier Execution units Shading units eDRAM(MB) Boost clock(MHz) Max GFLOPS FP16 FP32 FP64 Consumer HD Graphics GT1 10 80 N/A 1150 384 192 48 HD Graphics 4200 GT2 20 160 850 544 272 68 HD Graphics 4400 950–1150 608-736 304–368 76-92 HD Graphics 4600 900–1350 576-864 288–432 72-108 HD Graphics 5000 GT3 40 320 1000–1100 1280–1408 640–704 160-176 Iris Graphics 5100 1100–1200 1408–1536 704–768 176-192 Iris Pro Graphics 5200 GT3e 128 1300 1280–1728 640-864 160-216 Professional HD Graphics P4600 GT2 20 160 N/A 1200–1250 768-800 384–400 96-100 HD Graphics P4700 1250–1300 800-832 400–416 100-104

The 128 MB of eDRAM in the Iris Pro GT3e is in the same package as the CPU, but on a separate die manufactured in a different process. Intel refers to this as a Level 4 cache, available to both CPU and GPU, naming it Crystalwell. The Linux drm/i915 driver is aware and capable of using this eDRAM since kernel version 3.12.

Gen8 architecture

Broadwell In November 2013, it was announced that Broadwell-K desktop processors (aimed at enthusiasts) would also carry Iris Pro Graphics.

The following models of integrated GPU are announced for Broadwell processors:

Model number Tier Execution units Shading units eDRAM (MB) Boost clock (MHz) Max GFLOPS (FP32) Consumer HD Graphics GT1 12 96 rowspan="6" 850 163.2 HD Graphics 5300 GT2 24 192 900 345.6 HD Graphics 5500 950 364.8 HD Graphics 5600 1050 403.2 HD Graphics 6000 GT3 48 384 1000 768 Iris Graphics 6100 1100 844.8 Iris Pro Graphics 6200 GT3e 128 1150 883.2 Professional HD Graphics P5700 GT2 24 192 – 1000 384 Iris Pro Graphics P6300 GT3e 48 384 128 1150 883.2

Braswell Model number CPUmodel Tier Executionunits Clock speed(MHz) HD Graphics 400 E8000 GT1 12 320 N30xx 320–600 N31xx 320–640 J3xxx 320–700 HD Graphics 405 N37xx 16 400–700 J37xx 18 400–740

Gen9 architecture

Skylake The Skylake line of processors, launched in August 2015, retires VGA support, while supporting multi-monitor setups of up to three monitors connected via HDMI 1.4, DisplayPort 1.2 or Embedded DisplayPort (eDP) 1.3 interfaces.

The following models of integrated GPU are available or announced for the Skylake processors:

Model number Tier Execution units Shading units eDRAM (MB) Boost clock (MHz) Max GFLOPS (FP32) Consumer HD Graphics 510 GT1 12 96 rowspan="4" 1050 201.6 HD Graphics 515 GT2 24 192 1000 384 HD Graphics 520 1050 403.2 HD Graphics 530 1150 441.6 Iris Graphics 540 GT3e 48 384 64 1050 806.4 Iris Graphics 550 1100 844.8 Iris Pro Graphics 580 GT4e 72 576 128 1000 1152 Professional HD Graphics P530 GT2 24 192 – 1150 441.6 Iris Pro Graphics P555 GT3e 48 384 128 1000 768 Iris Pro Graphics P580 GT4e 72 576 1000 1152

Apollo Lake The Apollo Lake line of processors was launched in August 2016. Model number CPUmodel Tier ExecutionunitsShadingunits Clock speed(MHz) HD Graphics 500 E3930 GT1 1296 400 – 550 E3940 400–600 N3350 200–650 N3450 200–700 J3355 250–700 J3455 250–750 HD Graphics 505 E3950 18144 500–650 N4200 200–750 J4205 250–800

Gen9.5 architecture

Kaby Lake The Kaby Lake line of processors was introduced in August 2016. New features: speed increases, support for 4K UHD "premium" (DRM encoded) streaming services, media engine with full hardware acceleration of 8- and 10-bit HEVC and VP9 decode. Model number Tier Execution units Shading units eDRAM(MB) Base clock (MHz) Boost clock (MHz) Max GFLOPS (FP32) Used in Consumer HD Graphics 610 GT1 12 96 rowspan="4" 300−350 900−1100 172.8–211.2 Desktop Celeron, Desktop Pentium G4560, i3-7101 HD Graphics 615 GT2 24 192 300 900 – 1050 345.6 – 403.2 m3-7Y30/32, i5-7Y54/57, i7-7Y75, Pentium 4415Y HD Graphics 620 1000–1050 384–403.2 i3-7100U, i5-7200U, i5-7300U, i7-7500U, i7-7600U HD Graphics 630 350 1000–1150 384−441.6 Desktop Pentium G46**, i3, i5 and i7, and Laptop H-series i3, i5 and i7 Iris Plus Graphics 640 GT3e 48 384 64 300 950–1050 729.6−806.4 i5-7260U, i5-7360U, i7-7560U, i7-7660U Iris Plus Graphics 650 1050–1150 806.4−883.2 i3-7167U, i5-7267U, i5-7287U, i7-7567U Professional HD Graphics P630 GT2 24 192– 350 1000–1150 384−441.6 Xeon E3-**** v6

Kaby Lake Refresh / Amber Lake / Coffee Lake / Coffee Lake Refresh / Whiskey Lake / Comet Lake The Kaby Lake Refresh line of processors was introduced in October 2017. New features: HDCP 2.2 support Model number Tier Executionunits Shadingunits eDRAM(MB)Base clock(MHz) Boost clock(MHz) MaxGFLOPS (FP32)Used in Consumer UHD Graphics 610 GT1 12 96 – 350 1050 201.6 Pentium Gold G54**, Celeron G49** i5-10200H UHD Graphics 615 GT2 24 192 300 900–1050 345.6–403.2 i7-8500Y, i5-8200Y, m3-8100Y UHD Graphics 617 1050 403.2 i7-8510Y, i5-8310Y, i5-8210Y UHD Graphics 620 1000–1150 422.4–441.6 i3-8130U, i5-8250U, i5-8350U, i7-8550U, i7-8650U, i3-8145U, i5-8265U, i5-8365U, i7-8565U, i7-8665U i3-10110U, i5-10210U, i5-10310U, i7-10510U i7-10610U i7-10810U UHD Graphics 630 23 184 350 1100–1150 404.8–423.2 i3-8350K, i3-8100 with stepping B0 24 192 1050–1250 403.2–480 i9, i7, i5, i3, Pentium Gold G56**, G55** i5-10300H, i5-10400H, i5-10500H, i7-10750H, i7-10850H, i7-10870H, i7-10875H, i9-10885H, i9-10980HK Iris Plus Graphics 645 GT3e 48 384 128 300 1050–1150 806.4-883.2 i7-8557U, i5-8257U Iris Plus Graphics 655 1050–1200 806.4–921.6 i7-8559U, i5-8279U, i5-8269U, i5-8259U, i3-8109U Professional UHD Graphics P630 GT2 24 192– 350 1100–1200 422.4–460.8 Xeon E 21**G, 21**M, 22**G, 22**M, Xeon W-108**M

Gemini Lake/Gemini Lake Refresh New features: HDMI 2.0 support, VP9 10-bit Profile2 hardware decoder Model number Tier ExecutionunitsShadingunits CPUmodel Clock speed(MHz) GFLOPS (FP32) UHD Graphics 600 GT1 1296N4000200–65038.4–124.8 N4100 200–70038.4–134.4 J4005 250–70048.0–134.4 J4105 250–75048.0–144.0J4125250–75048.0–144.0 UHD Graphics 605 GT1.5 18N5000200–75057.6–216 J5005 250–80072.0–230.4

Gen11 architecture

Ice Lake New features: 10 nm Gen 11 GPU microarchitecture, two HEVC 10-bit encode pipelines, three 4K display pipelines (or 2× 5K60, 1× 4K120), variable rate shading (VRS), and integer scaling.

While the microarchitecture continues to support double-precision floating-point as previous versions did, the mobile configurations of it do not include the feature and therefore on these it is supported only through emulation.

Name Tier Executionunits Shadingunits Base clock(MHz) Boost clock(MHz) GFLOPS Used in FP16 FP32 FP64 Consumer UHD Graphics G1 32 256 300 900–1050 921.6–1075.2 460.8–537.6 115.2 Core i3-10**G1, i5-10**G1 Iris Plus Graphics G4 48 384 300 900–1050 1382.4–1612.8 691.2–806.4 96-202 Core i3-10**G4, i5-10**G4 G7 64 512 300 1050–1100 2150.4–2252.8 1075.2–1126.4 128-282 Core i5-10**G7, i7-10**G7

Xe-LP architecture (Gen12)

Model Process Execution units Shading units Max boost clock (MHz) Processing power (GFLOPS) Notes FP16 FP32 FP64 INT8 Intel UHD Graphics 730 Intel 14++ nm 24 192 1200–1300 922–998 461–499 rowspan="7" 1843–1997 Used in Rocket Lake-S Intel UHD Graphics 750 32 256 1200–1300 1228–1332 614–666 2457–2662 Intel UHD Graphics P750 32 256 1300 1332 666 2662 Used in Xeon W-1300 series Intel UHD Graphics for 11th Gen Intel ProcessorsIntel 10SF 32 256 1400–1450 1434–1484 717–742 2867–2970 Used in Tiger Lake-H Intel UHD Graphics for 11th Gen Intel Processors G4 48 384 1100–1250 1690–1920 845–960 3379–3840 Used in Tiger Lake-U/H35 Iris Xe Graphics G7 80 640 1100–1300 2816–3328 1408–1664 5632–6656 Iris Xe Graphics G7 96 768 1050–1450 3379–4454 1690–2227 6758–8909