AMD Radeon 780M vs AMD Radeon R9 M390X Comparison

Where Each One Wins

The benchmark split between these two is unusually one-sided. The AMD Radeon 780M wins the only shared test in the database, and it wins decisively. In Geekbench Vulkan, the 780M scores 33683 against 16951 for the Radeon R9 M390X. That is a 49.7% gap, meaning the 780M delivers nearly double the raw Vulkan performance of the older discrete card. For any modern workload that leans on Vulkan, the 780M is the clear pick.

The R9 M390X does hold one intangible advantage: it is a discrete GPU with dedicated GDDR5 memory. That matters in scenarios where system RAM bandwidth is a bottleneck, though the recorded data does not include a direct test of that scenario. Its Geekbench Metal score of 24373 is strong for its era, and its average benchmark score of 20662 places it in the 66th percentile of all GPUs. The 780M, by comparison, sits at the 61st percentile with an average score of 17588.

What this means in practice: if you are running Vulkan-based titles or compute workloads, the 780M is dramatically faster. If you are working with older APIs like Metal or OpenGL, the R9 M390X remains respectable, but the data shows its best days are behind it. The 780M's win is not narrow, it is a generational blowout.

Architecture Differences

These two GPUs come from completely different eras of AMD design. The R9 M390X uses GCN 3.0 architecture on a 28 nm process, built by TSMC. Its chip, codenamed Amethyst, packs 5,000 million transistors into a 366 mm² die. That works out to 13.7 million transistors per square millimeter. The 780M uses RDNA 3.0 on a 4 nm process, also TSMC, with 25,390 million transistors in a much smaller 178 mm² die. Its density of 142.6 million transistors per square millimeter is over ten times higher.

The R9 M390X has 2048 shading units, 128 texture mapping units, and 32 ROPs. The 780M has only 768 shading units and 48 TMUs, but it also has 32 ROPs. Despite fewer shading units, the 780M achieves higher throughput because of its much higher clock speeds. The 780M boosts to 2900 MHz, while the R9 M390X has no listed base or boost clock, only a memory clock of 1250 MHz. The 780M also includes 12 ray tracing cores, something the R9 M390X lacks entirely.

Memory is another fundamental split. The R9 M390X uses 4 GB of GDDR5 on a 256 bit bus, delivering 160.0 GB/s of bandwidth. The 780M uses system shared memory, so its capacity, bus width, and bandwidth are all system dependent. This is the classic tradeoff between a discrete GPU with dedicated VRAM and an integrated GPU that borrows from the CPU's memory pool.

The 780M also supports newer API features. It runs DirectX 12 Ultimate (12_2) and Vulkan 1.4, while the R9 M390X is limited to DirectX 12 (12_0) and Vulkan 1.2.170. Both support OpenGL 4.6. The 780M uses PCIe 4.0 x8, the R9 M390X uses PCIe 3.0 x16.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The R9 M390X has an average benchmark score of 20662, placing it in the 66th percentile of all GPUs. The 780M averages 17588, which is the 61st percentile. The R9 M390X is 0.2% behind the AMD Radeon RX 5600 XT, while the 780M is 0.2% ahead of the NVIDIA GeForce RTX 4060.

Q: Is the 780M really faster in Vulkan?

A: Yes, by a massive margin. In Geekbench Vulkan, the 780M scores 33683 versus 16951 for the R9 M390X, a 49.7% difference. This is the only direct head-to-head test in the database, and the 780M wins it outright.

Q: Does the R9 M390X have more shading units?

A: It does. The R9 M390X has 2048 shading units, while the 780M has 768. However, the 780M compensates with a 2900 MHz boost clock and RDNA 3.0 efficiency, resulting in 8.909 TFLOPS of FP32 performance versus 2.961 TFLOPS for the R9 M390X.

Q: Which GPU supports ray tracing?

A: Only the 780M. It has 12 ray tracing cores. The R9 M390X has no ray tracing hardware whatsoever, which is expected given its 2015 release date.

Q: What about power consumption?

A: The R9 M390X is rated at 75 W TDP, while the 780M is rated at 15 W. The 780M is an integrated GPU with no power connectors, just like the R9 M390X's MXM module design, but the 780M delivers far more performance per watt.

Q: Which GPU is still in production?

A: The 780M is listed as active, released in January 2024. The R9 M390X is end-of-life, released in May 2015. The 780M's predecessor is Navi II IGP, and its successor is Navi III IGP.

Specification Differences

| Specification | AMD Radeon R9 M390X | AMD Radeon 780M |

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

| Architecture | GCN 3.0 | RDNA 3.0 |

| Process Node | 28 nm | 4 nm |

| Transistors | 5,000 million | 25,390 million |

| Die Size | 366 mm² | 178 mm² |

| Transistor Density | 13.7M / mm² | 142.6M / mm² |

| Boost Clock | Not listed | 2900 MHz |

| Memory Size | 4 GB GDDR5 | System Shared |

| Memory Bus | 256 bit | System Shared |

| Memory Bandwidth | 160.0 GB/s | System Dependent |

| Shading Units | 2048 | 768 |

| TMUs | 128 | 48 |

| ROPs | 32 | 32 |

| RT Cores | None | 12 |

| Pixel Rate | 23.14 GPixel/s | 92.80 GPixel/s |

| Texture Rate | 92.54 GTexel/s | 139.2 GTexel/s |

| FP32 Performance | 2.961 TFLOPS | 8.909 TFLOPS |

| FP16 Performance | 2.961 TFLOPS (1:1) | 8.909 TFLOPS (1:1) |

| TDP | 75 W | 15 W |

| Slot Width | MXM Module | IGP |

| Bus Interface | PCIe 3.0 x16 | PCIe 4.0 x8 |

| DirectX Support | 12 (12_0) | 12 Ultimate (12_2) |

| Vulkan Support | 1.2.170 | 1.4 |

| Production Status | End-of-life | Active |

| Release Date | May 2015 | January 2024 |

Head-to-Head Benchmarks

There is exactly one shared benchmark between these two GPUs in the database, but it tells the whole story. In Geekbench Vulkan, the 780M scores 33683, while the R9 M390X scores 16951. The delta is 49.7% in favor of the 780M. That is not a marginal improvement, it is a near-doubling of performance.

To put that score in context, the 780M's Vulkan result is roughly double what the R9 M390X achieves. The 780M also has a Geekbench OpenCL score of 18602, which is not directly comparable to the R9 M390X's Metal score of 24373, but it shows the 780M is competitive even in compute workloads despite its integrated nature.

The R9 M390X's best showing is in Geekbench Metal, where it scores 24373. That test is not available for the 780M, so we cannot make a direct comparison there. But the overall average scores tell the story: the R9 M390X averages 20662 across all tests, while the 780M averages 17588. The 780M's average is dragged down by the fact that it includes a 3DMark Steel Nomad DX12 result of 480, which is a very different workload than the Geekbench tests.

The interesting wrinkle is that the 780M sits closer to its nearest rivals than the R9 M390X does. The 780M is within 0.5% of the AMD Radeon Pro 560, NVIDIA GeForce RTX 4060, AMD Radeon HD 7790, and AMD Radeon Pro 460. The R9 M390X is within 0.6% of the AMD Radeon RX 5600 XT, Intel Arc A750, Intel Arc B570, and NVIDIA GeForce RTX 3070 Mobile. Both GPUs are remarkably tightly clustered with their peers, but the 780M's Vulkan dominance is the standout data point.

The Verdict

The data is unambiguous: the AMD Radeon 780M is the faster GPU in the one test that matters most for modern gaming and compute. Its Vulkan score of 33683 crushes the R9 M390X's 16951, and its FP32 throughput of 8.909 TFLOPS is triple the older card's 2.961 TFLOPS. It does all of this at 15 W versus 75 W, with a smaller die, newer architecture, and active production status.

The R9 M390X is not without merits. Its average benchmark score is higher at 20662 versus 17588, and it holds up well in the 66th percentile of all GPUs. Its dedicated 4 GB of GDDR5 memory on a 256 bit bus is a genuine advantage in memory-bound scenarios, and its Metal score of 24373 shows it is still capable in Apple-centric workloads. But it is end-of-life, built on 28 nm, and lacks ray tracing support entirely.

For a builder choosing between these two today, the 780M is the obvious pick for any Vulkan-based gaming or compute workload. It is also the only option if you need ray tracing, DirectX 12 Ultimate, or Vulkan 1.4. The R9 M390X makes sense only if you are constrained to a system that requires an MXM module, or if you specifically need dedicated VRAM and can tolerate 2015-era architecture. The benchmark data does not support choosing the R9 M390X for raw performance, because the 780M wins the only direct comparison by nearly half.