Intel UHD Graphics P750 vs NVIDIA GeForce GTX 460 Comparison

The Verdict

The NVIDIA GeForce GTX 460 is the clear winner in the recorded benchmark data, taking the only head-to-head test with a 20.9% advantage. Its Geekbench OpenCL score of 7925 places it in the 41st percentile of all GPUs in the database, while the Intel UHD Graphics P750 scores 6554 and sits in the 38th percentile. The GTX 460 is a discrete graphics card that remains competitive with later mobile and workstation parts, landing within 1.6% of the GeForce GTX 880M and within 0.2% of the Quadro K4100M. The Intel UHD Graphics P750 is an integrated processor graphics solution that trades blows with older mobile discrete chips, sitting 0.4% behind the Radeon HD 7730M and 0.9% ahead of the GeForce GT 555M.

For gaming or any workload where raw compute throughput matters, the GTX 460 is the pick from the data. It has dedicated GDDR5 memory and over 900 GFLOPS of FP32 performance. For an ultra-low-power system where the CPU's integrated graphics is the only option and the 15 W TDP is a hard constraint, the P750 is the only feasible choice. The GTX 460 demands 160 W, a dual-slot cooler, and two 6-pin power connectors, while the P750 uses no separate power connectors and is motherboard dependent for its display outputs. The verdict is simple: the GTX 460 for performance, the P750 for integration and power efficiency.

Architecture Differences

The two parts come from fundamentally different design philosophies. The GTX 460 uses the GF104 chip built on Fermi architecture, fabricated by TSMC on a 40 nm process. It packs 1,950 million transistors into a 332 mm² die, giving a transistor density of 5.9 million per square millimeter. The P750 is part of Intel's Generation 12.1 architecture, built on the Rocket Lake chip using Intel's 14 nm+++ process. The database records no transistor count or die size for the Intel part, which reflects its nature as an integrated GPU sharing the CPU die.

The compute resources differ substantially. The GTX 460 has 336 shading units, 56 texture mapping units, and 24 ROPs. The P750 has 256 shading units, 64 TMUs, and 32 ROPs. Despite fewer shading units, the P750 posts higher pixel and texture rates: 41.60 GPixel/s and 83.20 GTexel/s against the GTX 460's 9.45 GPixel/s and 37.80 GTexel/s. The P750's FP32 throughput is 665.6 GFLOPS, while the GTX 460 reaches 907.2 GFLOPS. The P750 also supports FP16 at 1,331.2 GFLOPS (2:1 ratio), a capability the GTX 460 does not record.

Memory architecture is where the gap widens. The GTX 460 uses 768 MB of GDDR5 on a 192-bit bus, delivering 86.40 GB/s of bandwidth with a memory clock of 900 MHz (3.6 Gbps effective). The P750 uses system shared memory with a system dependent bandwidth figure and no dedicated VRAM. This explains the benchmark gap despite the P750's higher pixel and texture rates; the GTX 460 feeds its shaders from dedicated high-speed memory, while the P750 competes with the CPU for system memory bandwidth.

API support also differs. The GTX 460 supports DirectX 12 (11_0) and OpenGL 4.6, but records no Vulkan support. The P750 supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The P750 has a newer DirectX feature level and adds Vulkan, while the GTX 460 has none recorded in the database.

Head-to-Head Benchmarks

Only one benchmark appears in the head-to-head comparison: Geekbench OpenCL. The GTX 460 scores 7925 points, and the P750 scores 6554 points. The delta is 20.9% in favor of the GTX 460. That is a substantial margin, larger than the percentile gap might suggest. The GTX 460 sits at the 41st percentile while the P750 sits at the 38th, but the raw score difference of 1371 points is decisive.

Context from the nearest rivals strengthens the picture. The GTX 460's 7925 sits just 0.2% above the Quadro K4100M's 7906, and only 1.4% below the Quadro P5000's 8039 and the GTX 880M's 8040. The P750's 6554 is 0.6% above the GTX 670M's 6513 and 0.9% above the GT 555M's 6493, but 0.4% below the Radeon HD 7730M's 6581 and 0.9% below the Radeon R7 M460's 6612. The P750 competes in an older mobile discrete class, while the GTX 460 competes in a higher tier that includes workstation parts like the Quadro P5000.

The win count reflects this: the GTX 460 records 1 win in the head-to-head comparison, the P750 records 0. The GTX 460 wins the only recorded test, and no test in the database favors the Intel part.

FAQ

Q: Which GPU is faster in OpenCL compute?

A: The NVIDIA GeForce GTX 460 scores 7925 in Geekbench OpenCL, which is 20.9% higher than the Intel UHD Graphics P750's 6554.

Q: Does the Intel UHD Graphics P750 support DirectX 12 at a higher feature level?

A: Yes. The P750 supports DirectX 12 (12_1), while the GTX 460 supports DirectX 12 (11_0). The P750 also supports Vulkan 1.4, while the GTX 460 has no Vulkan support recorded.

Q: How much memory bandwidth does each GPU have?

A: The GTX 460 has 86.40 GB/s of bandwidth from 768 MB of GDDR5 on a 192-bit bus. The P750 uses system shared memory with a system dependent bandwidth figure.

Q: What are the power requirements for each GPU?

A: The GTX 460 has a 160 W TDP, requires a dual-slot cooler, two 6-pin power connectors, and a suggested PSU of 450 W. The P750 has a 15 W TDP, is an IGP, and records no power connectors or suggested PSU.

Q: How does the GTX 460 compare to its nearest rivals?

A: The GTX 460 scores 0.2% above the Quadro K4100M and 1.4% below both the Quadro P5000 and the GeForce GTX 880M. It also sits 1.6% below the GeForce GTX 650 Ti.

Q: How does the P750 compare to its nearest rivals?

A: The P750 scores 0.6% above the GeForce GTX 670M and 0.9% above the GeForce GT 555M. It sits 0.4% below the Radeon HD 7730M and 0.9% below the Radeon R7 M460.

Where Each One Wins

The GTX 460 wins in raw compute performance. Its 907.2 GFLOPS of FP32 throughput, 86.40 GB/s of dedicated memory bandwidth, and 20.9% OpenCL advantage over the P750 make it the choice for any task that stresses shader throughput or memory-bound workloads. It also competes in a higher performance tier, as shown by its proximity to workstation GPUs like the Quadro P5000 (1.4% behind) and the Quadro K4100M (0.2% ahead). The GTX 460 is an end-of-life discrete card with a 199 USD launch MSRP, but its benchmark standing remains respectable even against much newer parts.

The P750 wins in integration and efficiency. Its 15 W TDP is dramatically lower than the GTX 460's 160 W, it needs no power connectors, and it occupies no expansion slot as an IGP. Its display outputs are motherboard dependent, which means it is built into the platform rather than added to it. The P750 also has higher pixel and texture rates on paper: 41.60 GPixel/s and 83.20 GTexel/s versus 9.45 GPixel/s and 37.80 GTexel/s. It supports FP16 at 1,331.2 GFLOPS, a feature the GTX 460 lacks entirely. For a system where power draw is critical and the CPU's integrated graphics is sufficient, the P750 is the only option between these two.

The API advantage also belongs to the P750. It supports DirectX 12 (12_1) and Vulkan 1.4, while the GTX 460 is limited to DirectX 12 (11_0) with no Vulkan support. That matters for modern titles that lean on newer API features. The GTX 460 is a legacy Fermi part from the GeForce 400 generation, while the P750 is Generation 12.1 from Intel. The GTX 460's process node is 40 nm versus the P750's 14 nm+++, and the GTX 460's bus interface is PCIe 2.0 x16 while the P750 uses a Ring Bus. The GTX 460's display outputs are 2x DVI and 1x mini-HDMI 1.3a, whereas the P750's are motherboard dependent.

Specification Differences

The two GPUs differ across nearly every recorded specification. The GTX 460 is built on the GF104 chip with Fermi architecture from the GeForce 400 generation, using a 40 nm TSMC process with 1,950 million transistors on a 332 mm² die. The P750 uses the Rocket Lake chip with Generation 12.1 architecture from the HD Graphics-W (Rocket Lake) generation, on Intel's 14 nm+++ process, with no transistor or die size recorded.

Clock behavior differs in kind. The GTX 460 has no base or boost clock recorded, only a memory clock of 900 MHz (3.6 Gbps effective). The P750 has a 350 MHz base clock and a 1300 MHz boost clock, with system shared memory. The GTX 460's memory is 768 MB of GDDR5 on a 192-bit bus with 86.40 GB/s bandwidth. The P750's memory size, type, and bus width are all system shared, with system dependent bandwidth.

Compute resources differ: 336 shading units, 56 TMUs, and 24 ROPs for the GTX 460; 256 shading units, 64 TMUs, and 32 ROPs for the P750. Pixel rates are 9.45 GPixel/s versus 41.60 GPixel/s. Texture rates are 37.80 GTexel/s versus 83.20 GTexel/s. FP32 is 907.2 GFLOPS versus 665.6 GFLOPS. FP16 is absent for the GTX 460 and 1,331.2 GFLOPS (2:1) for the P750.

Power and physical specs separate them further. The GTX 460 has a 160 W TDP, a dual-slot design, two 6-pin power connectors, a 450 W suggested PSU, a PCIe 2.0 x16 interface, and a 210 mm (8.3 inches) length. The P750 has a 15 W TDP, an IGP slot width, no power connectors, no suggested PSU, a Ring Bus interface, and no dimensions recorded. Display outputs are 2x DVI and 1x mini-HDMI 1.3a for the GTX 460, versus motherboard dependent for the P750. API support shows DirectX 12 (11_0) and OpenGL 4.6 for the GTX 460, with no Vulkan; DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4 for the P750. The GTX 460 has a release date of 2010-07-11 and a launch MSRP of 199 USD, while the P750 has no release date or MSRP recorded. Both are end-of-life products.