Intel Iris Pro Graphics 6200 vs NVIDIA GeForce GTX 750 Comparison

The NVIDIA GeForce GTX 750 and Intel Iris Pro Graphics 6200 represent two very different answers to the same question: how do you get acceptable GPU performance without a huge power budget. The recorded data makes the split clear. The GTX 750 is a discrete card built on NVIDIA's Maxwell architecture, drawing 55 W and delivering its own GDDR5 memory. The Iris Pro 6200 is an integrated GPU soldered into Intel's Broadwell processors, sipping 15 W and sharing system memory. Across the three recorded Geekbench tests, the GTX 750 wins two and the Iris Pro 6200 wins one, and the average benchmark scores land at 7222 for the GTX 750 versus 6117 for the Iris Pro. That gap of roughly 1100 points is real but not enormous: the GTX 750 sits at the 40th percentile against all GPUs in the database, while the Iris Pro 6200 sits at the 35th. Both are end-of-life parts, so the practical question for anyone choosing between them today is not which is newer, but which profile fits the build.

The Verdict

Pick the GTX 750 if compute throughput and driver breadth matter. It wins Geekbench OpenCL by a wide margin and Geekbench Vulkan by a solid one, and it carries newer API support in OpenGL 4.6 and Vulkan 1.4 compared to the Iris Pro's OpenGL 4.4 and Vulkan 1.0. Pick the Iris Pro 6200 if power envelope is the constraint and Metal-style throughput is the workload. It wins the Geekbench Metal test outright with 7764 versus 4274, an 82 percent lead in raw score terms, and it does so inside a 15 W package with no slot, no power connector, and no dedicated memory to buy. What the data does not support is calling either a gaming solution in absolute terms: both cluster in the middle-to-lower band of the database, with the GTX 750 nearly identical to parts like the AMD Radeon Vega 8 Mobile (0.3 percent delta) and the Iris Pro 6200 trading blows with the AMD Radeon HD 8690M (0.3 percent delta). These are entry-level performers by current standards.

Architecture Differences

The GTX 750 is the GM107 chip on NVIDIA's Maxwell architecture, part of the GeForce 700 generation, fabricated by TSMC on a 28 nm process. It packs 1870 million transistors into a 148 mm² die, which works out to a density of 12.6M per mm². The Iris Pro 6200 is Intel's Broadwell GT3e, built on Intel's own 14 nm process, roughly half the node size of the Maxwell part. The database records no transistor count or die size for the Intel chip.

The execution resources differ in shape as much as in size. NVIDIA gives the GTX 750 512 shading units, 32 TMUs, and 16 ROPs. Intel gives the Iris Pro 6200 fewer shaders at 384 and far fewer ROPs at 6, but more texture units at 48. That imbalance shows up directly in the throughput figures: the GTX 750 pushes 17.36 GPixel/s against 6.600 GPixel/s for the Iris Pro, a pixel fill rate advantage of nearly three to one, while the Iris Pro actually leads on texture rate at 52.80 GTexel/s versus 34.72 GTexel/s. Raw FP32 compute lands at 1111.0 GFLOPS for the GTX 750 against 844.8 GFLOPS for the Iris Pro.

Memory is the deepest divide. The GTX 750 carries 1024 MB of dedicated GDDR5 on a 128-bit bus, clocked at 1253 MHz for 5 Gbps effective, yielding 80.19 GB/s of guaranteed bandwidth. The Iris Pro 6200 has no memory of its own: size, type, bus width, and bandwidth are all system shared and system dependent. That means its real-world bandwidth rides on whatever system memory the host platform provides, an inherent variance a discrete card simply does not have. Clock behavior also differs: the GTX 750 runs a tight 1020 MHz base to 1085 MHz boost range, while the Iris Pro swings from a 300 MHz base up to a 1100 MHz boost.

Where Each One Wins

The GTX 750 owns general-purpose compute and modern API paths. Its OpenCL score of 9315 is more than double the Iris Pro's 4556, which matters for any workload that dispatches through OpenCL. Its Vulkan score of 8078 versus 6032 gives it a clear lead in the most current graphics API recorded here, and its Vulkan 1.4 and OpenGL 4.6 support means it can run software the Iris Pro's Vulkan 1.0 and OpenGL 4.4 cannot. Its much higher pixel fill rate also favors any workload bound on rasterization output.

The Iris Pro 6200 wins where the benchmark matches its strengths: the Metal test, where its 7764 towers over the GTX 750's 4274. Its higher texture rate and higher boost clock explain part of that, and its 14 nm process explains how it does this at 15 W. For a system where a discrete slot is unavailable or the power budget is sealed shut, it is the only option of the two, and the data shows it is not a weak one, sitting just 0.3 percent from the Radeon HD 8690M and within about one percent of the NVIDIA GeForce MX230 and the NVIDIA Quadro P2000 in average score.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The GTX 750, at 7222 versus 6117 for the Iris Pro 6200, a gap of about 1100 points. It also sits five percentile points higher against all GPUs in the database, 40th versus 35th.

Q: Does the Iris Pro 6200 win anything?

A: Yes. It takes the Geekbench Metal test with 7764 against 4274, a 45 percent deficit for the GTX 750. That is the single largest win either way.

Q: How do their power profiles compare?

A: The GTX 750 has a 55 W TDP, needs no power connectors, is single-slot, and carries a suggested PSU of 250 W. The Iris Pro 6200 is a 15 W integrated GPU with no slot, no connectors, and no suggested PSU figure recorded.

Q: Which has dedicated memory?

A: The GTX 750, with 1024 MB of GDDR5 on a 128-bit bus delivering 80.19 GB/s. The Iris Pro 6200 shares system memory for everything.

Q: Are either of these still in production?

A: No. Both are marked end-of-life. The GTX 750 launched first, in February 2014, followed by the Iris Pro 6200 in September 2014. The GTX 750 had a launch MSRP of 119 USD; no MSRP is recorded for the Iris Pro.

Q: Which supports newer graphics APIs?

A: The GTX 750, on every count: DirectX 12 (11_0) versus 12 (11_1) is the one category where the Intel part lists the higher feature level, but OpenGL 4.6 beats 4.4 and Vulkan 1.4 beats 1.0.

Head-to-Head Benchmarks

The OpenCL result is the blowout. The GTX 750 scores 9315 and the Iris Pro 6200 scores 4556, a delta of 104.5 percent in NVIDIA's favor. For any compute pipeline that runs through OpenCL, the discrete card delivers roughly twice the throughput. Given that the raw FP32 difference is only about 24 percent (1111.0 versus 844.8 GFLOPS), the OpenCL gap suggests the dedicated GDDR5 bandwidth and mature compute drivers are doing heavy lifting here, not shader count alone.

Vulkan goes to the GTX 750 as well, 8078 to 6032, a 33.9 percent advantage. This is the most balanced result of the three and the most representative of modern graphics work, since both parts support Vulkan and the API layer is current on both, at least minimally. A one-third lead is meaningful but not decisive; it positions the GTX 750 clearly ahead without the humiliation of the OpenCL spread.

Metal flips the script completely. The Iris Pro 6200 posts 7764 against 4274 for the GTX 750, a 45 percent win for Intel. Its higher boost clock of 1100 MHz and its 48 TMUs likely contribute, but the magnitude of the reversal is striking: the same chip that manages barely half the GTX 750's OpenCL output more than doubles it here. Any workload aligned with this API's characteristics strongly favors the integrated part.

The overall tally, two wins to one for the GTX 750, matches the average score gap. Neither part dominates.

Specification Differences

• Manufacturer and chip: NVIDIA GM107 (Maxwell, GeForce 700) versus Intel Broadwell GT3e (Generation 8.0, HD Graphics).
• Process: 28 nm at TSMC versus 14 nm at Intel.
• Shading units: 512 versus 384.
• TMUs: 32 versus 48.
• ROPs: 16 versus 6.
• Pixel rate: 17.36 GPixel/s versus 6.600 GPixel/s.
• Texture rate: 34.72 GTexel/s versus 52.80 GTexel/s.
• FP32 compute: 1111.0 GFLOPS versus 844.8 GFLOPS.
• Clocks: 1020/1085 MHz base/boost versus 300/1100 MHz.
• Memory: 1024 MB GDDR5, 128-bit, 80.19 GB/s, 1253 MHz (5 Gbps effective) versus fully system shared.
• TDP: 55 W versus 15 W.
• Form factor: single-slot PCIe 3.0 x16 card, 145 mm long, no power connectors, 250 W suggested PSU versus an IGP on a ring bus with motherboard-dependent display outputs.
• Display outputs: 2x DVI and 1x mini-HDMI 1.4a versus motherboard dependent.
• API support: DirectX 12 (11_0), OpenGL 4.6, Vulkan 1.4 versus DirectX 12 (11_1), OpenGL 4.4, Vulkan 1.0.
• Release dates: February 2014 versus September 2014.
• Launch MSRP: 119 USD for the GTX 750; none recorded for the Iris Pro 6200.
• Lineage: the GTX 750 sits between the GeForce 600 and GeForce 900 generations; no predecessor or successor is recorded for the Iris Pro.

The bottom line from the data: the GTX 750 is the stronger all-around performer with dedicated memory and broader API support, while the Iris Pro 6200 is the efficiency pick that wins its preferred benchmark decisively at roughly a quarter of the power draw.