AMD Radeon HD 8790M vs AMD Radeon R6 M255DX Comparison

FAQ

Q: Which GPU has the higher average benchmark score?

A: The AMD Radeon HD 8790M averages 5691 across its recorded tests, while the AMD Radeon R6 M255DX averages 4867.

Q: How do the two compare in the Geekbench Vulkan test?

A: The AMD Radeon HD 8790M scores 6365, which is 23.5% higher than the AMD Radeon R6 M255DX's 4867.

Q: What architecture do both GPUs use?

A: Both are built on GCN 1.0 architecture, produced at TSMC's 28 nm process node, with identical transistor density of 12.3M per mm².

Q: What is the memory configuration difference?

A: The R6 M255DX uses system shared memory with bandwidth described as system dependent. The HD 8790M has 2 GB of dedicated GDDR5 on a 128-bit bus, delivering 64.00 GB/s.

Q: Which GPU has a higher shading unit count?

A: The HD 8790M has 384 shading units, compared to 320 on the R6 M255DX.

Q: What is the production status of both cards?

A: Both are marked as end-of-life products. The R6 M255DX released in January 2014, while the HD 8790M released in March 2013.

Architecture Differences

Both GPUs share the same GCN 1.0 architecture and TSMC 28 nm process, but they are fundamentally different designs. The AMD Radeon R6 M255DX uses the Jet chip, a compact integrated-class processor with 690 million transistors on a 56 mm² die. The AMD Radeon HD 8790M employs the Mars chip, a larger discrete mobile solution with 950 million transistors on a 77 mm² die. This 260 million transistor difference and the larger die area give the HD 8790M more room for compute resources.

The R6 M255DX belongs to the Gem System Hybrid generation under the Rx M200 family, operating as an integrated graphics processor (IGP) with no dedicated memory. It relies entirely on system shared memory, meaning its performance scales with the host laptop's RAM configuration. The HD 8790M comes from the Solar System generation under the HD 8700M family and is a discrete MXM module with its own 2 GB GDDR5 memory. The memory clock is rated at 1000 MHz with 4 Gbps effective transfer, delivering 64.00 GB/s over a 128-bit bus.

Clock speeds differ notably. The R6 M255DX runs at a 780 MHz base with an 855 MHz boost. The HD 8790M runs at 850 MHz base and 900 MHz boost, giving it roughly a 9% clock advantage at base and 5% at boost. Combined with 64 more shading units, 4 more texture mapping units, and identical 8 ROPs, the HD 8790M achieves higher theoretical throughput: 691.2 GFLOPS FP32 versus 547.2 GFLOPS, 21.60 GTexel/s versus 17.10 GTexel/s, and 7.200 GPixel/s versus 6.840 GPixel/s.

Both support DirectX 12 (11_1), OpenGL 4.6, and Vulkan 1.2.170. Display outputs are portable-device dependent for both, which reflects their mobile nature. The R6 M255DX has no power connectors and occupies an IGP slot width; the HD 8790M is an MXM-A (3.0) module with no power connectors either. The HD 8790M names London as its predecessor and Gem System as its successor, while the R6 M255DX has no recorded predecessor or successor in the database.

The Verdict

The data points clearly toward the AMD Radeon HD 8790M as the stronger performer. Its average benchmark score of 5691 places it at the 33rd percentile among all GPUs, while the R6 M255DX sits at the 28th percentile. In the only head-to-head test recorded, the Geekbench Vulkan benchmark, the HD 8790M wins by 23.5%, a substantial margin for a single-generation comparison.

The HD 8790M's nearest rivals in the database include the Intel Iris Pro Graphics P6300 (avg 5712, 0.4% behind) and the NVIDIA GeForce GTX 670MX (avg 5721, 0.5% behind), placing it in the company of mid-range discrete mobile parts. The R6 M255DX competes with the NVIDIA GeForce GTX 560M (avg 4855, 0.2% ahead) and the GeForce 940MX (avg 4844, 0.5% ahead), a lower performance tier overall.

Users who need dedicated graphics memory, higher compute throughput, and better raw benchmark scores should select the HD 8790M. Its 2 GB GDDR5 avoids the system-dependent bandwidth bottleneck that plagues the R6 M255DX. The R6 M255DX is suitable only for systems where integrated graphics are the only option and where the host platform's shared memory can provide adequate bandwidth. For any workload where the discrete HD 8790M can be installed, the benchmark data recommends it without qualification.

Specification Differences

| Field | AMD Radeon R6 M255DX | AMD Radeon HD 8790M |

|---|---|---|

| Chip | Jet | Mars |

| Generation | Gem System Hybrid (Rx M200) | Solar System (HD 8700M) |

| Transistors | 690 million | 950 million |

| Die size | 56 mm² | 77 mm² |

| Base clock | 780 MHz | 850 MHz |

| Boost clock | 855 MHz | 900 MHz |

| Memory clock | System Shared | 1000 MHz, 4 Gbps effective |

| Memory size | System Shared | 2 GB |

| Memory type | System Shared | GDDR5 |

| Memory bus width | System Shared | 128 bit |

| Memory bandwidth | System Dependent | 64.00 GB/s |

| Shading units | 320 | 384 |

| TMUs | 20 | 24 |

| Pixel rate | 6.840 GPixel/s | 7.200 GPixel/s |

| Texture rate | 17.10 GTexel/s | 21.60 GTexel/s |

| FP32 | 547.2 GFLOPS | 691.2 GFLOPS |

| Slot width | IGP | MXM Module |

| Bus interface | IGP | MXM-A (3.0) |

| Release date | 2014-01-06 | 2013-03-31 |

| Predecessor | None recorded | London |

| Successor | None recorded | Gem System |

Identical fields include manufacturer (AMD), architecture (GCN 1.0), process node (28 nm), foundry (TSMC), transistor density (12.3M / mm²), ROPs (8), TDP (not recorded), display outputs (portable device dependent), DirectX version (12 11_1), OpenGL version (4.6), Vulkan version (1.2.170), and production status (end-of-life).

Head-to-Head Benchmarks

The database records one direct comparison: Geekbench Vulkan. The AMD Radeon HD 8790M scores 6365, while the AMD Radeon R6 M255DX scores 4867. This yields a delta of -23.5% for the R6 M255DX, meaning the HD 8790M outperforms it by nearly a quarter in this API workload. The margin is consistent with the theoretical compute gap: the HD 8790M delivers 691.2 GFLOPS against 547.2 GFLOPS, a 26% advantage in raw FP32 throughput.

Looking at the average scores across all recorded tests, the HD 8790M's 5691 average exceeds the R6 M255DX's 4867 by 824 points, or about 16.9%. The HD 8790M also has an additional OpenCL benchmark score of 5017, which the R6 M255DX lacks entirely; this suggests the discrete card was tested in more diverse workloads.

The percentile rankings reinforce the gap. The HD 8790M sits at the 33rd percentile of all GPUs, five points higher than the R6 M255DX's 28th percentile. In practical terms, the HD 8790M lands near the NVIDIA GeForce GTX 670MX (delta -0.5%) and GTX 550 Ti (delta -0.7%), while the R6 M255DX trades blows with the GeForce GTX 560M (delta 0.2%) and GeForce 940MX (delta 0.5%). The R6 M255DX even trails the GeForce GTS 450 by 0.5% and the GeForce RTX 5060 Ti 8 GB by 0.7%, though the latter is an odd pairing given its modern class.

The wins tally confirms the outcome: the HD 8790M wins 1 out of 1 recorded head-to-head tests; the R6 M255DX wins none.

Where Each One Wins

The AMD Radeon HD 8790M wins in every measurable category. It has higher clocks, more shading units, more texture units, and dedicated GDDR5 memory with 64.00 GB/s bandwidth. The Vulkan benchmark shows a 23.5% lead, and the average score advantage of 824 points places it in a higher performance tier among its nearest rivals. Users running Vulkan-based games, OpenCL compute workloads, or any graphics task that benefits from dedicated memory should choose the HD 8790M. Its 384 shading units and 24 TMUs give it clear headroom for texture-heavy scenes and parallel compute.

The AMD Radeon R6 M255DX has no benchmark wins in the recorded data. Its only potential advantage is form factor: as an IGP, it requires no separate module and uses system shared memory, which may simplify system integration in thin laptops. Its nearest rival comparisons show it within 0.5% of the GeForce 940MX and GTX 560M, meaning it performs at entry-level discrete graphics levels from several years prior. For users constrained to integrated graphics, the R6 M255DX provides a baseline experience, but the data shows it cannot match the HD 8790M in any recorded workload.

The use-case split is therefore simple. The HD 8790M is the choice for gaming, GPU-accelerated rendering, and any application where frame rates or compute throughput matter. The R6 M255DX is the fallback for portable devices that cannot accommodate a discrete MXM module. The specification sheet reinforces this: the HD 8790M has a larger die, more transistors, higher clocks, and its own memory pool. The R6 M255DX's system-dependent bandwidth is a fundamental limitation that no driver optimization can fully overcome. Benchmark results indicate that for equal workloads, the HD 8790M consistently delivers superior performance.