Intel Iris Pro Graphics 5200 vs NVIDIA GeForce GTX 970M Comparison

The NVIDIA GeForce GTX 970M and Intel Iris Pro Graphics 5200 represent two fundamentally different approaches to mobile graphics: a dedicated, high-performance discrete GPU versus an integrated processor graphics solution. Benchmark data shows the GTX 970M is decisively faster, but the Iris Pro 5200 holds its own as a low-power IGP. The GTX 970M’s average benchmark score of 4628 places it in the 27th percentile of all GPUs, while the Iris Pro 5200’s average score of 4360 lands just one percentile lower at 26th. Despite the similar overall percentiles, the head-to-head benchmark results reveal a massive performance gulf, with the GTX 970M winning both comparison tests by margins exceeding 275%.

Where Each One Wins

The NVIDIA GeForce GTX 970M wins every single head-to-head benchmark in the data, making it the clear choice for any graphics-intensive workload. In the two tests where both GPUs have results — Geekbench OpenCL and Geekbench Vulkan — the GTX 970M dominates. The Geekbench OpenCL score of 18946 is roughly 3.75 times higher than the Iris Pro 5200’s 5042, and the Vulkan score of 18292 is nearly five times the Iris Pro’s 3677. The GTX 970M’s PassMark G3D score of 5704 and PassMark GPU Compute score of 2289 further underscore its compute and rendering strength, though the Iris Pro 5200 lacks direct comparisons in those tests.

The Intel Iris Pro Graphics 5200 wins in one specific area: power efficiency and integration. With a 45 W TDP, it is designed for systems where a dedicated GPU is impractical. As an IGP, it uses system shared memory, which means no dedicated VRAM allocation is required. Its 22 nm process node from Intel contrasts with the GTX 970M’s 28 nm node from TSMC, giving Intel a manufacturing advantage. However, this does not translate into performance wins; the Iris Pro 5200 simply offers a baseline graphics capability for everyday tasks, while the GTX 970M is engineered for demanding 3D workloads. The data shows zero benchmark wins for Intel, so any use-case favoring the Iris Pro would be based on system design constraints rather than performance.

FAQ

Q: How much faster is the NVIDIA GeForce GTX 970M in OpenCL compute workloads?

A: The GTX 970M scores 18946 in Geekbench OpenCL, which is 275.8% higher than the Iris Pro 5200’s score of 5042. This indicates the NVIDIA GPU delivers roughly 3.75 times the compute performance in this test.

Q: Does the Intel Iris Pro Graphics 5200 beat the GTX 970M in any benchmark?

A: No. In the head-to-head benchmark data, the GTX 970M wins both the Geekbench OpenCL and Geekbench Vulkan tests. The Iris Pro 5200 records zero wins in the comparison.

Q: What is the difference in Vulkan performance between the two GPUs?

A: The GTX 970M scores 18292 in Geekbench Vulkan, while the Iris Pro 5200 scores 3677. This represents a 397.5% delta, meaning the NVIDIA GPU is nearly five times faster in this API workload.

Q: How do their overall average benchmark scores compare?

A: The GTX 970M has an average benchmark score of 4628, while the Iris Pro 5200 averages 4360. The difference is 268 points, which is approximately a 6% gap in favor of NVIDIA.

Q: What are the percentile rankings for each GPU?

A: The GTX 970M sits in the 27th percentile of all GPUs, while the Iris Pro 5200 sits in the 26th percentile. Despite the huge performance gap in head-to-head tests, their overall percentile rankings are nearly identical.

Q: Which GPU has a higher memory bandwidth?

A: The GTX 970M features 120.3 GB/s of bandwidth with a 192-bit bus and 6 GB of GDDR5 memory. The Iris Pro 5200 uses system shared memory, with bandwidth listed as system dependent and no dedicated memory bus.

Head-to-Head Benchmarks

The only two benchmarks where both GPUs have recorded scores are Geekbench OpenCL and Geekbench Vulkan, and the NVIDIA GeForce GTX 970M wins both by staggering margins. In Geekbench OpenCL, the GTX 970M posts 18946 points versus the Iris Pro 5200’s 5042 points. The deltaPct of 275.8% means the NVIDIA GPU is more than three and a half times faster. This test measures general-purpose compute performance across a variety of workloads, and the GTX 970M’s 1280 shading units, 80 TMUs, and 48 ROPs provide the raw hardware resources for such a dominant showing.

The Geekbench Vulkan test shows an even larger gap. The GTX 970M scores 18292, while the Iris Pro 5200 manages only 3677. The deltaPct here is 397.5%, meaning the NVIDIA GPU is nearly five times faster. Vulkan is a low-overhead graphics API, and the GTX 970M’s support for DirectX 12 (12_1) and Vulkan 1.4, combined with its 2.657 TFLOPS of FP32 compute, allows it to leverage the API’s efficiency far better than Intel’s IGP. The Iris Pro 5200 only supports Vulkan 1.0 and offers 736.0 GFLOPS of FP32 performance, which is less than a third of NVIDIA’s throughput.

Beyond the head-to-head tests, the GTX 970M’s PassMark results reinforce its strength. It scores 5704 in PassMark G3D and 2289 in GPU Compute, along with 99 in PassMark DirectX 9 and 42 in DirectX 11. These scores are not directly comparable to the Iris Pro 5200, which lacks PassMark entries, but they show a GPU capable of handling legacy and modern DirectX titles. The GTX 970M’s pixel rate of 49.82 GPixel/s and texture rate of 83.04 GTexel/s dwarf the Iris Pro’s 4.600 GPixel/s and 46.00 GTexel/s, respectively, further illustrating the performance chasm.

Specification Differences

The two GPUs differ dramatically across nearly every specification. The GTX 970M is built on NVIDIA’s GM204 chip using a 28 nm process from TSMC, while the Iris Pro 5200 uses Intel’s Haswell GT3e chip on a 22 nm process. The GTX 970M has 5,200 million transistors on a 398 mm² die, whereas the Iris Pro 5200’s transistor count and die size are not listed. The NVIDIA GPU features 1280 shading units, 80 TMUs, and 48 ROPs, compared to Intel’s 320 shading units, 40 TMUs, and just 4 ROPs.

Memory configurations could not be more different. The GTX 970M has 6 GB of GDDR5 memory on a 192-bit bus with 120.3 GB/s bandwidth. The Iris Pro 5200 uses system shared memory, with its size, type, bus width, and bandwidth all listed as system dependent. Clock speeds also vary widely: the GTX 970M runs at a 924 MHz base and 1038 MHz boost, while the Iris Pro 5200 starts at 200 MHz and boosts to 1150 MHz. The NVIDIA GPU’s memory clock is 1253 MHz (5 Gbps effective), while Intel’s memory clock is tied to the system.

The GTX 970M is a mobile module with an MXM-B (3.0) bus interface and no power connectors, while the Iris Pro 5200 is an IGP with a Ring Bus interface and a 45 W TDP. The GTX 970M has no TDP listed. Display outputs are portable device dependent for NVIDIA and motherboard dependent for Intel. The GTX 970M supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4, while the Iris Pro 5200 supports DirectX 12 (11_1), OpenGL 4.3, and Vulkan 1.0. The GTX 970M was released on 2014-10-06, while the Iris Pro 5200 came earlier on 2013-06-02.

Architecture Differences

The architectural divide between these two GPUs starts with their fundamental design philosophy. The GTX 970M uses NVIDIA’s Maxwell 2.0 architecture, which is a discrete GPU design with dedicated resources for shading, texturing, and rasterization. Maxwell 2.0 is known for its efficiency improvements over previous NVIDIA architectures, and the GTX 970M’s 28 nm process from TSMC houses 5,200 million transistors. The Iris Pro 5200, in contrast, uses Intel’s Generation 7.5 architecture, which is the integrated graphics solution built into Haswell processors. It is fabricated on Intel’s own 22 nm process and shares system memory with the CPU.

Shader and compute capabilities highlight the architectural gap. The GTX 970M offers 2.657 TFLOPS of FP32 compute, while the Iris Pro 5200 offers only 736.0 GFLOPS — a difference of nearly 3.6 times. The NVIDIA GPU’s pixel rate of 49.82 GPixel/s is more than ten times the Intel’s 4.600 GPixel/s. Texture rate tells a similar story: 83.04 GTexel/s versus 46.00 GTexel/s, a 1.8 times advantage for NVIDIA despite the GTX 970M having twice the TMUs.

Feature support also diverges. The GTX 970M supports DirectX 12 (12_1), which includes features like conservative rasterization and rasterizer-ordered views, while the Iris Pro 5200 only supports DirectX 12 (11_1), which is a more limited feature set. Vulkan support differs as well: NVIDIA offers version 1.4, while Intel offers version 1.0. OpenGL support is 4.6 on NVIDIA versus 4.3 on Intel. The GTX 970M’s 48 ROPs enable high-resolution rendering with more efficient pixel throughput, while the Iris Pro 5200’s 4 ROPs are severely limited, making it suitable only for basic display output and light 2D workloads. The GTX 970M’s 80 TMUs provide substantial texture filtering capabilities, whereas Intel’s 40 TMUs are halved. These architectural differences explain the benchmark results: the GTX 970M is a purpose-built graphics processor with massive parallel resources, while the Iris Pro 5200 is an integrated solution designed for minimal power draw and basic visual tasks.