Intel HD Graphics 3000 vs Intel HD Graphics 4400 vs Intel HD Graphics 6000

Intel HD Graphics 3000

The Intel HD Graphics 3000 (or Intel Graphics Media Accelerator HD 3000, GMA HD 3000, Intel HD Graphics 200) is an integrated graphics card in the Sandy Bridge codenamed processors. The HD Graphics 3000 has no dedicated memory but shares the Level 3 / LLC Cache with the CPU cores and also part of the main memory. Due to TurboBoost, the GPU can be overclocked depending on the current CPU load and power consumption. The base speed and the turbo boost speed of the HD Graphics 3000 depend on the processor:

• ULV processors Core ix-2xx7 (base 350MHz, Turbo 900-1000MHz)
• LV processors Core ix-2xx5 (base 500MHz, Turbo 1000MHz)
• Mainstream and high-end Dual und Quad-Core Core ix-2xx0 (base 650MHz, Turbo 1100-1300MHz)
• Desktop K processors (base 850, Turbo 1100-1350MHz)

The HD 3000 offers 12 Execution Units (EUs) like the old Intel GMA HD but due to architectural changes each EU is now faster. The slower HD Graphics 2000 uses only 6 Execution Units. The EUs can be accessed using DirectX 10.1,  OpenGL 3.0 and DirectCompute 4.1. OpenCL is not supported (the media SDK only uses the CPU).

Our performance tests with the high clocked mainstream version of the HD Graphics 3000 in quad-core processors showed a performance level on par with older entry level generations from NVIDIA (e.g. GeForce 310M) and AMD (HD 5450). In comparison with AMD's APU models, the HD 3000 can only compete against the C-and E-series, but not the faster Llano and Trinity models. The quality of the Intel drivers has been improved over the years, so most games are flawlessly playable (if the minimum requirements are fullfilled).

We could measure smooth frame rates in games like Dead Space 3, World of Tanks, Fifa 13, Torchlight 2, Counter-Strike: GO reach, Diablo 3, and many older games, though usually only in minimal detail settings. For accurate performance analysis and gaming benchmarks, read our article "Intel HD Graphics 3000" or scroll down in this article.

Altought 2x Antialiasing (AA) is supported, the chip is as fast as when rendering 4x AA. In the Unigine Valley benchmark a i7-2637M for example reached the same score with 2x AA and 4x AA.

In addition to the GPU, the chip also houses some dedicated units for decoding and encoding HD videos. On the IDF Intel demonstrated the encoding of a 3 minute long 1080p video to an iPhone compatible format in 640x360 in only 14s. Another novelty of the Sandy Bridge GPU is the embedded DisplayPort eDP to connect internal Displays.

Due to the integration in the 32nm built CPU, the power consumption of the GPU should be quite low.

Intel HD Graphics 4400

The Intel HD Graphics 4400 (GT2) is a processor graphics card included in some of the ULV Haswell processors of 2013. The relatively low base clock can be automatically overclocked using Turbo Boost technology. Depending on the processor model, the turbo clock rates may differ, resulting in varying graphics performance between models.

In comparison to the HD 4000, the HD 4400 graphics core has been modified extensively. The GPU now supports DirectX 11.1, OpenCL 1.2 and OpenGL 4.0. It also features an improved decoder for 4K videos and the fast Quick Sync encoder. Compared to the faster 4600, the 4400 offers the same amount of shaders, but lower clock speeds (see table of clock speeds of the different CPU models below).

The performance of the HD Graphics 4400 is somewhat below the HD 4600, since the GPU is designed for ULV models. Therefore, the clock rates are relatively low. Furthermore, the reduced TDP limits the Turbo Boost. Compared to the ULV versions of the Ivy Bridge HD 4000, the HD 4400 is about 20 - 30 percent faster. This performance boost is achieved by architectural improvements and an increased number of execution units: The GT2 version integrates 20 EUs, compared to 16 EUs for the old HD 4000. Depending on the clock rate, the HD 4400 matches the performance of a dedicated Radeon HD 7550M.

Due to the 22nm 3D Tri-Gate production process, the power consumption is relatively low. The HD Graphics 4400 can be found on ULV dual-core Haswell models with a TDP of 15 watts.

Intel HD Graphics 6000

The Intel HD Graphics 6000 (GT3) is an integrated Broadwell graphics card revealed in Q1 2015. It can be found in several ULV SoCs (15 W TDP) such as the Core i5-5250U or i7-5650U. The so-called GT3 GPU offers 48 EUs (Execution Units) and therefore somewhat more shader power than the previous HD 5000 (40 EUs). Depending on the specific CPU, the maximum GPU frequency varies between 950 and 1000 MHz.

Architecture and Features

Broadwell features a GPU based on the Intel Gen8 architecture, which has been optimized in various aspects compared to the previous Gen7.5 (Haswell). Inter alia, the shader arrays called "subslice" have been reorganized and now offer 8 Execution Units (EUs) each. Three subslices form a "slice" for a total of 24 EUs. Combined with other improvements such as larger L1 caches and an optimized frontend, the integrated GPU has become faster and more efficient than its predecessor.

The HD Graphics 6000 represents the top-end version of the Broadwell GPU family and consists of two slices with 48 EUs. Beyond that, there is also a low-end (GT1, 12 EUs), a mid-range (GT2, 24 EUs) and a high-end variant (GT3e, 48 EUs + eDRAM).

All Broadwell GPUs support OpenCL 2.0 and DirectX 12 (FL 11_1). The video engine can now decode H.265 using both fixed function hardware as well as available GPU shaders. Up to three displays can be connected via DP 1.2/eDP 1.3 (max. 3840 x 2160 @ 60 Hz) or HDMI 1.4a (max. 3840 x 2160 @ 24 Hz). HDMI 2.0, however, is not supported.

Performance

Depending on the specific CPU, the maximum GPU frequency varies between 950 and 1000 MHz. Due to the low TDP, however, the average clock in 3D applications will be significantly lower. Overall, the HD Graphics 6000 outperforms the previous HD 5000 by 20 - 25 percent and offers a performance slightly below a dedicated GeForce 820M.

Games as of 2014/2015 will usually run fluently only at (very) low settings.

Power Consumption

Utilizing a new 14 nm process, the Broadwell ULV chips are specified at just 15 W TDP and therefore suited for thin ultrabooks. The TDP is flexible and can be further reduced (9.5 W), which has a significant impact on performance.