AMD Radeon R7 M460 vs Intel UHD Graphics P750 Comparison

The AMD Radeon R7 M460 and Intel UHD Graphics P750 are two end-of-life mobile graphics solutions that land in the same performance tier, separated by less than one percent in the only available benchmark. The R7 M460, a discrete part from AMD’s GCN 3.0 family, edges out Intel’s integrated Generation 12.1 solution by a razor-thin margin, with both cards sitting at the 38th percentile of all GPUs. This parity makes the choice between them less about raw speed and more about the architectural trade-offs each brings to a system.

Head-to-Head Benchmarks

The sole head-to-head data point comes from the Geekbench OpenCL test, where the AMD Radeon R7 M460 scores 6,612 points against the Intel UHD Graphics P750’s 6,554. That works out to a 0.9% delta in favor of AMD, a lead so narrow it falls within typical run-to-run variance. The R7 M460 also sits 0.5% above the AMD Radeon HD 7730M (6,581) and 1.5% above the NVIDIA GeForce GTX 670M (6,513), while trailing the NVIDIA GeForce GT 1010 (6,698) by 1.3%. For Intel, the P750 is 0.6% ahead of the GTX 670M and 0.9% ahead of the NVIDIA GeForce GT 555M (6,493), but it trails the HD 7730M by 0.4% and the R7 M460 by the same 0.9% margin that defines this matchup.

Look closer at the scoring structure, and the two GPUs achieve their near-identical results through very different means. The R7 M460 has 384 shading units operating at a boost clock of 1024 MHz, while the P750 deploys just 256 shading units but runs them at a higher 1300 MHz boost. AMD’s card also has 24 texture mapping units and 8 ROPs, versus Intel’s 64 TMUs and 32 ROPs. The result is that Intel’s pixel fill rate (41.60 GPixel/s) is more than five times AMD’s (8.192 GPixel/s), and its texture rate (83.20 GTexel/s) is over three times higher (24.58 GTexel/s). Yet in the OpenCL compute workload, AMD’s higher FP32 throughput (786.4 GFLOPS vs 665.6 GFLOPS) compensates for its fill-rate deficit, producing the near-deadlock.

The delta between the two is also reflected in their rival sets. Both cards share two common rivals: the HD 7730M and the GTX 670M. AMD’s scores against those are +0.5% and +1.5%, while Intel’s are -0.4% and +0.6%. The P750’s closest competitor is actually the HD 7730M, just 0.4% behind, whereas the R7 M460’s closest is the HD 7730M at 0.5% ahead. Neither card can claim a decisive victory in any direction; the entire competitive envelope spans less than three percentage points across four rival GPUs.

Where Each One Wins

The AMD Radeon R7 M460 wins the only benchmark category available, taking the Geekbench OpenCL crown by 58 points. That edge suggests a slight advantage in compute-oriented tasks that leverage FP32 arithmetic, given its 786.4 GFLOPS of single-precision throughput versus Intel’s 665.6 GFLOPS. The R7 M460 also benefits from dedicated GDDR5 memory with 36.00 GB/s of bandwidth and a 64-bit bus, which isolates its performance from system memory contention. For workloads that are shader-bound or rely on raw floating-point math, the data points to AMD holding a marginal but real lead.

The Intel UHD Graphics P750 counters with structural advantages that don’t show up in the OpenCL score but matter in other contexts. Its 41.60 GPixel/s pixel rate and 83.20 GTexel/s texture rate are dramatically higher than AMD’s, which indicates far stronger fill-rate-bound performance in traditional rasterization scenarios. The P750 also has double the ROP count (32 vs 8) and nearly triple the TMU count (64 vs 24), plus a 1300 MHz boost clock that tops AMD’s 1024 MHz. In games or applications that stress pixel shading and texture filtering rather than compute, the Intel part should pull ahead despite losing the synthetic compute test. Its FP16 throughput of 1,331.2 GFLOPS (2:1 ratio) also gives it a clear edge for half-precision workloads, where the R7 M460 manages only 786.4 GFLOPS at a 1:1 ratio.

The P750’s 15 W TDP and IGP form factor also make it the low-power choice, though the R7 M460 has no TDP listed to compare directly. For systems where memory bandwidth is shared with the CPU, the P750’s "System Shared" memory setup is a liability; the R7 M460’s dedicated 2 GB GDDR5 pool avoids that contention. Conversely, the P750’s "Motherboard Dependent" display outputs and Ring Bus interface tie it to the host platform in ways a discrete card is not.

Architecture Differences

The two GPUs come from fundamentally different design philosophies. AMD’s R7 M460 is built on the Meso chip using GCN 3.0 architecture, fabricated on a 28 nm process at TSMC. The die measures 125 mm² and packs 1,550 million transistors, yielding a density of 12.4 million transistors per mm². It belongs to the Gem System generation (R7 M400 family) and was released in May 2016. Intel’s P750 uses the Rocket Lake chip with Generation 12.1 architecture, built on Intel’s 14 nm+++ process. Its transistor count and die size are not listed, and it carries no release date in the data.

The R7 M460’s memory subsystem is straightforward: 2 GB of GDDR5 on a 64-bit bus, delivering 36.00 GB/s of bandwidth. The P750 instead uses system shared memory for both capacity and type, with bandwidth described as "System Dependent." That means the Intel part’s memory performance varies with the host platform’s RAM configuration, while AMD’s is fixed by its VRAM. The R7 M460 connects via PCIe 3.0 x8, whereas the P750 uses a Ring Bus architecture.

Compute resources favor AMD on paper: 384 shading units against 256, though Intel counters with 64 TMUs and 32 ROPs versus 24 and 8. Clock speeds tell a similar story, with AMD’s base at 730 MHz and boost at 1024 MHz, while Intel’s base is a low 350 MHz but boost reaches 1300 MHz. FP32 performance is higher on AMD (786.4 GFLOPS vs 665.6 GFLOPS), but FP16 reverses decisively (786.4 GFLOPS vs 1,331.2 GFLOPS). API support is comparable, with both offering DirectX 12 (AMD at 12_0, Intel at 12_1), OpenGL 4.6, and Vulkan (AMD at 1.2.170, Intel at 1.4).

Intel’s TDP is listed at 15 W, confirming its integrated nature, while AMD’s is absent from the data. The P750’s slot width is "IGP," meaning it occupies no expansion slot, and its display outputs are motherboard-dependent. The R7 M460’s display outputs are not listed. AMD’s transistor density advantage (12.4M / mm²) reflects the older but denser 28 nm node, while Intel’s 14 nm+++ process is not quantified in the pack.

FAQ

Q: Which GPU has the higher benchmark score?

A: The AMD Radeon R7 M460 scores 6,612 in Geekbench OpenCL, which is 0.9% higher than the Intel UHD Graphics P750’s 6,554.

Q: How do they compare to their nearest rivals?

A: The R7 M460 is 0.5% above the AMD Radeon HD 7730M and 1.5% above the NVIDIA GeForce GTX 670M, but 1.3% below the NVIDIA GeForce GT 1010. The P750 is 0.6% above the GTX 670M and 0.9% above the NVIDIA GeForce GT 555M, while sitting 0.4% below the HD 7730M.

Q: Which GPU has better raw compute performance?

A: AMD’s R7 M460 delivers 786.4 GFLOPS of FP32 throughput, higher than Intel’s 665.6 GFLOPS. However, Intel’s P750 offers 1,331.2 GFLOPS of FP16, far above AMD’s 786.4 GFLOPS.

Q: What are the memory configurations for each?

A: The R7 M460 uses 2 GB of dedicated GDDR5 with a 64-bit bus and 36.00 GB/s bandwidth. The P750 relies on system shared memory with system-dependent bandwidth.

Q: Which GPU has higher pixel and texture rates?

A: The Intel UHD Graphics P750 has a 41.60 GPixel/s pixel rate and 83.20 GTexel/s texture rate. The AMD R7 M460 has 8.192 GPixel/s and 24.58 GTexel/s, respectively.

Q: What process nodes are used?

A: AMD uses a 28 nm process at TSMC for the Meso chip, while Intel uses its own 14 nm+++ process for Rocket Lake.

The Verdict

Based strictly on the benchmark data, the AMD Radeon R7 M460 is the faster GPU by a 0.9% margin in Geekbench OpenCL, making it the pick for anyone prioritizing compute performance. Its dedicated 2 GB GDDR5 memory with 36.00 GB/s bandwidth is a concrete advantage over Intel’s system-dependent shared memory, which can bottleneck in memory-heavy workloads. The R7 M460 also leads in FP32 throughput (786.4 GFLOPS) and comes from a mature GCN 3.0 architecture with a 28 nm process. For users running OpenCL-based applications or shader-heavy tasks that favor single-precision math, the data supports choosing AMD.

The Intel UHD Graphics P750, however, is not a clear loser. Its 41.60 GPixel/s and 83.20 GTexel/s rates are far superior to AMD’s 8.192 GPixel/s and 24.58 GTexel/s, which points to much stronger fill-rate performance in rasterization-bound scenarios. Its 15 W TDP and IGP form factor make it the right choice for low-power or space-constrained systems, and its FP16 capability (1,331.2 GFLOPS) doubles AMD’s for half-precision workloads. The P750 also supports DirectX 12_1 and Vulkan 1.4, slightly ahead of AMD’s 12_0 and 1.2.170.

In practical terms, the choice hinges on workload type. If the primary use is compute or gaming that stresses shader units, the R7 M460’s 0.9% OpenCL lead and dedicated VRAM give it the edge. If the workload involves heavy pixel processing, high-resolution texturing, or FP16 compute, the P750’s architectural strengths in fill rate and half-precision throughput make it the more capable part despite its lower raw FP32 score. Both GPUs sit at the 38th percentile and are end-of-life, so neither offers a future-proofing advantage. The data shows a statistical tie in the only measured benchmark, but the underlying architectures favor different use cases—AMD for compute density, Intel for fill-rate-bound and low-power scenarios.