AMD Radeon HD 8730M vs NVIDIA GeForce GT 555M Comparison

FAQ

Q: Which GPU is faster in the recorded OpenCL benchmark?

A: The NVIDIA GeForce GT 555M scores 6493, which is 9% higher than the AMD Radeon HD 8730M's 5955 in the geekbench_opencl test.

Q: How do these two mobile GPUs compare to their nearest rivals?

A: The GT 555M sits 0.2% above the NVIDIA Quadro M5000M (6481) and 0.3% above the AMD Radeon Vega 10 Mobile (6476), while trailing the GTX 670M by 0.3% and the Intel UHD Graphics P750 by 0.9%. The HD 8730M matches the Quadro K620M (5957) within 0.0%, sits 0.3% below the HD 8750M (5970), and leads the Intel UHD Graphics 730 (5929) by 0.4% and the Quadro K4000 (5982) by 0.5%.

Q: What are the memory specifications for both cards?

A: Both use DDR3 memory on a 128-bit bus with 28.80 GB/s bandwidth. The NVIDIA card has 1024 MB, while the AMD card has 2 GB. Both run memory at 900 MHz with 1800 Mbps effective speed.

Q: What is the difference in shading unit counts?

A: The AMD Radeon HD 8730M has 384 shading units, while the NVIDIA GeForce GT 555M has 144. Despite this, the NVIDIA part still wins in the recorded benchmark score.

Q: Which GPU supports Vulkan?

A: The AMD Radeon HD 8730M supports Vulkan version 1.2.170, while the NVIDIA GeForce GT 555M has no Vulkan support listed in the database.

Q: What are the pixel and texture rates for each card?

A: The AMD card achieves 5.600 GPixel/s and 16.80 GTexel/s, while the NVIDIA card achieves 3.540 GPixel/s and 14.16 GTexel/s.

The Verdict

The benchmark data points to a clear but narrow winner: the NVIDIA GeForce GT 555M takes the single recorded OpenCL test with a 9% margin. Its score of 6493 places it in the 37th percentile of all GPUs, while the AMD Radeon HD 8730M lands at 5955 and the 34th percentile. For tasks that rely on OpenCL compute performance, the GT 555M is the stronger pick.

However, the choice is not purely about the headline score. The AMD part carries 2 GB of memory versus 1 GB, double the shading units (384 vs 144), a smaller and denser chip, and a newer architecture generation (GCN 1.0 vs Fermi). The HD 8730M also supports Vulkan, which the GT 555M does not. Users who prioritize compute throughput in the recorded test should choose the NVIDIA card. Users who value memory capacity, shader count, and modern API support for compatibility reasons may prefer the AMD option, accepting the lower benchmark score.

The delta is modest: 9% separates them. Neither card dominates decisively. The database shows the GT 555M outperforming its rival in raw OpenCL, but the HD 8730M offers structural advantages that may matter in specific workloads not captured by this single test.

Head-to-Head Benchmarks

The only recorded head-to-head benchmark is geekbench_opencl, and the result favors NVIDIA. The GeForce GT 555M scores 6493, while the Radeon HD 8730M scores 5955. That is a 9% advantage for the NVIDIA card, representing a lead of 538 points.

This margin is meaningful but not overwhelming. In the context of each card's nearest rivals, the GT 555M's score places it in a cluster where competitors range from 6476 (Radeon Vega 10 Mobile) to 6554 (Intel UHD Graphics P750). The HD 8730M's score sits in a lower cluster, with rivals ranging from 5929 (Intel UHD Graphics 730) to 5982 (Quadro K4000). The two clusters do not overlap, which means the 9% gap between the two cards is consistent across their respective competitive sets.

The AMD card's closest rival, the Quadro K620M, scores 5957, just 2 points above the HD 8730M. The NVIDIA card's closest rival, the Quadro M5000M, scores 6481, just 12 points below the GT 555M. These near-identical margins highlight how tightly packed these mobile GPUs are within their performance tiers.

For the AMD card, the 0.3% deficit to the HD 8750M (5970) shows that it is at the lower end of its own family. For the NVIDIA card, the 0.3% gap to the GTX 670M (6513) shows it is also near the bottom of its performance class. The head-to-head result, therefore, reflects two entry-level mobile parts where NVIDIA holds a measurable but not transformative edge in OpenCL compute.

Specification Differences

The two cards differ across nearly every major specification field. The NVIDIA GeForce GT 555M uses a 40 nm process, while the AMD Radeon HD 8730M uses a 28 nm process, both fabricated by TSMC. The NVIDIA chip, GF106, contains 1,170 million transistors on a 238 mm² die, yielding a density of 4.9M transistors per mm². The AMD chip, Mars, packs 950 million transistors on a 77 mm² die, yielding 12.3M per mm².

Memory capacity differs: the GT 555M has 1024 MB, the HD 8730M has 2 GB. Both use DDR3 on a 128-bit bus with identical 28.80 GB/s bandwidth and 900 MHz memory clock. The shading unit count diverges sharply: 144 for NVIDIA versus 384 for AMD. Texture mapping units match at 24 each, but ROPs differ: 16 for NVIDIA versus 8 for AMD.

Clock behavior also differs. The AMD card has a base clock of 650 MHz and a boost clock of 700 MHz. The NVIDIA card lists no base or boost clock in the database, only its 900 MHz memory clock. Pixel rate favors AMD at 5.600 GPixel/s versus 3.540 GPixel/s. Texture rate favors AMD at 16.80 GTexel/s versus 14.16 GTexel/s. FP32 compute favors AMD at 537.6 GFLOPS versus 339.8 GFLOPS.

The bus interface differs: PCIe 2.0 x16 for NVIDIA versus PCIe 3.0 x8 for AMD. Power consumption is listed at 35 W for the NVIDIA card, while the AMD card has no TDP recorded. Display outputs are listed as portable-device-dependent for NVIDIA, with no listing for AMD.

Architecture Differences

The NVIDIA GeForce GT 555M is built on the Fermi architecture, released in the GeForce 500M generation. The AMD Radeon HD 8730M uses GCN 1.0, part of the Solar System (HD 8700M) generation. These architectures represent different design philosophies. Fermi, from NVIDIA, uses a 40 nm process and a 238 mm² die. GCN 1.0, from AMD, uses a 28 nm process and a much smaller 77 mm² die.

The transistor counts reflect this: NVIDIA packs 1,170 million transistors into its larger die, while AMD fits 950 million into less than a third of the area. The density numbers tell the story: 4.9M per mm² for NVIDIA versus 12.3M per mm² for AMD. The newer 28 nm node allows AMD to achieve more than twice the transistor density.

Shader architecture differs fundamentally. NVIDIA's Fermi design places 144 shading units, while AMD's GCN 1.0 design arrays 384 shading units. The ROP count goes the other way: 16 for NVIDIA versus 8 for AMD. This creates an interesting split where AMD has more compute-oriented hardware (shaders, FP32 throughput) but fewer pixel output units.

API support shows the generational gap. Both support DirectX 12, but the NVIDIA card supports version 12 (11_0) while the AMD card supports 12 (11_1). OpenGL is 4.6 for both. Vulkan is supported only by the AMD card, at version 1.2.170, while the NVIDIA card has no Vulkan entry in the database.

The AMD card also has a boost clock mechanism (650 MHz base, 700 MHz boost), which the NVIDIA card lacks in the recorded specifications. The bus interface reflects their respective generations: PCIe 2.0 x16 for the 2011 NVIDIA part, PCIe 3.0 x8 for the 2013 AMD part.

Where Each One Wins

The NVIDIA GeForce GT 555M wins in the only recorded benchmark: OpenCL compute. Its 6493 score beats the AMD card's 5955 by 9%. This is the decisive metric in the database, and for any workload that behaves like the geekbench_opencl test, the GT 555M is the faster part.

The NVIDIA card also holds a specification advantage in ROP count (16 vs 8) and has a recorded TDP of 35 W, which may indicate a lower power draw for the GPU itself, though the database does not provide a comparable figure for the AMD card.

The AMD Radeon HD 8730M wins in several structural categories. It has 2 GB of memory versus 1 GB, which matters for workloads that exceed the smaller frame buffer. It has 384 shading units versus 144, a 2.7x advantage that explains its higher FP32 throughput (537.6 GFLOPS vs 339.8 GFLOPS). Pixel rate is 58% higher (5.600 vs 3.540 GPixel/s), and texture rate is 19% higher (16.80 vs 14.16 GTexel/s).

The AMD card also wins on process technology and density: 28 nm versus 40 nm, with 12.3M transistors per mm² versus 4.9M. It supports Vulkan 1.2.170, while the NVIDIA card does not list Vulkan support. Its boost clock of 700 MHz provides dynamic performance scaling that the NVIDIA card lacks in the recorded data.

In practical terms, the database suggests this split: the GT 555M is the choice for OpenCL compute performance, where its 9% lead is the only measured outcome. The HD 8730M is the choice for memory-intensive tasks, higher shader throughput, newer API compatibility, and efficiency of design. The benchmark result is the only direct comparison available, and it favors NVIDIA, but the AMD card's architecture and memory capacity offer capabilities that the single test does not capture.