AMD Radeon 780M vs AMD Radeon RX 6650 XT Comparison

Where Each One Wins

The AMD Radeon RX 6650 XT is the clear winner for DirectX 12 gaming workloads and Vulkan-based applications. It takes 2 out of 3 head-to-head benchmarks, with its most dominant showing in 3DMark Steel Nomad DX12, where it scores 1859 against the Radeon 780M's 480, a 287.3% advantage. The Vulkan test also goes decisively to the RX 6650 XT, with 81306 points versus 33683, a 141.4% lead. These results indicate that the discrete card is the stronger choice for modern gaming APIs, particularly where raw rasterization throughput and dedicated graphics memory matter.

The AMD Radeon 780M wins the OpenCL compute benchmark, scoring 18602 against the RX 6650 XT's 10111, a 45.6% deficit for the discrete card. This is a notable reversal, as the integrated GPU outperforms the dedicated card in this specific compute workload. The 780M's higher boost clock of 2900 MHz, combined with its RDNA 3.0 architecture, appears to give it an edge in this particular OpenCL test. For users whose primary workload is OpenCL-based compute, the 780M is the better performer. However, it is important to note that the 780M is an integrated graphics processor with system-shared memory, so its real-world compute performance will depend heavily on the host system's memory configuration.

Architecture Differences

The two GPUs come from different generations of AMD's RDNA architecture. The RX 6650 XT uses RDNA 2.0 with the Navi 23 chip, built on TSMC's 7 nm process. The 780M uses RDNA 3.0 with the Phoenix chip, built on TSMC's 4 nm process. The newer process node allows the 780M to pack 25,390 million transistors into a 178 mm² die, while the RX 6650 XT has 11,060 million transistors on a 237 mm² die. This results in a transistor density of 142.6M per mm² for the 780M versus 46.7M per mm² for the RX 6650 XT, a substantial difference in integration efficiency.

The RX 6650 XT has 2048 shading units, 128 texture mapping units, 64 render output units, and 32 ray tracing cores. The 780M has 768 shading units, 48 TMUs, 32 ROPs, and 12 ray tracing cores. This is a significant reduction in raw execution resources, explaining the RX 6650 XT's dominance in most gaming benchmarks. The RX 6650 XT also has dedicated 8 GB of GDDR6 memory on a 128-bit bus with 280.3 GB/s of bandwidth. The 780M uses system-shared memory, with its bandwidth listed as system dependent. This is a critical difference, as dedicated VRAM provides predictable performance, whereas the 780M's performance is contingent on the host system's memory speed and capacity.

Clock speeds also differ markedly. The RX 6650 XT has a base clock of 2055 MHz, a boost clock of 2635 MHz, and a game clock of 2410 MHz. The 780M has a base clock of 800 MHz and a boost clock of 2900 MHz. The 780M's boost clock is higher, but the RX 6650 XT's base clock is more than 2.5 times higher, which helps it sustain performance under load. The RX 6650 XT also has a significantly higher power envelope at 176 W TDP, while the 780M is rated at 15 W TDP. This explains the performance difference: the RX 6650 XT draws far more power to feed its larger array of compute units.

Head-to-Head Benchmarks

The 3DMark Steel Nomad DX12 test is the most lopsided comparison. The RX 6650 XT scores 1859, while the 780M scores 480. This 287.3% delta shows that the discrete card has more than three times the performance in this DirectX 12 workload. The RX 6650 XT's dedicated memory bandwidth of 280.3 GB/s and its 2048 shading units are the dominant factors here. The 780M, with 768 shading units and shared memory, cannot keep pace in this demanding test.

The Geekbench Vulkan test also goes to the RX 6650 XT, with a score of 81306 versus 33683 for the 780M. The 141.4% delta confirms that the RX 6650 XT has a substantial advantage in Vulkan-based applications. The RX 6650 XT's FP32 throughput of 10.79 TFLOPS, compared to the 780M's 8.909 TFLOPS, contributes to this result. The texture rate of 337.3 GTexel/s for the RX 6650 XT versus 139.2 GTexel/s for the 780M also plays a role.

The Geekbench OpenCL test is the sole win for the 780M. It scores 18602 against the RX 6650 XT's 10111, a 45.6% deficit for the discrete card. This is surprising given the RX 6650 XT's higher FP32 rating. The 780M's higher boost clock of 2900 MHz may be more relevant in this workload, as OpenCL compute tasks often respond well to higher clocks rather than wider execution widths. The 780M's RDNA 3.0 architecture may also have more efficient compute units for this specific test.

FAQ

Q: Which GPU is faster in DirectX 12 gaming?

A: The AMD Radeon RX 6650 XT. In the 3DMark Steel Nomad DX12 benchmark, it scores 1859 versus 480 for the AMD Radeon 780M, a 287.3% advantage.

Q: Does the AMD Radeon 780M win any benchmarks?

A: Yes, it wins the Geekbench OpenCL test with 18602 points, while the RX 6650 XT scores 10111. This is a 45.6% lead for the 780M.

Q: How does the RX 6650 XT compare in Vulkan performance?

A: The RX 6650 XT scores 81306 in Geekbench Vulkan, while the 780M scores 33683. The RX 6650 XT leads by 141.4%.

Q: What is the memory configuration difference?

A: The RX 6650 XT has 8 GB of dedicated GDDR6 memory on a 128-bit bus with 280.3 GB/s bandwidth. The 780M uses system-shared memory, with bandwidth listed as system dependent.

Q: Which GPU has more shading units?

A: The RX 6650 XT has 2048 shading units, while the 780M has 768. The RX 6650 XT also has 128 TMUs versus 48, and 64 ROPs versus 32.

Q: What are the power requirements?

A: The RX 6650 XT has a 176 W TDP and requires a 1x 8-pin power connector with a suggested 450 W PSU. The 780M has a 15 W TDP, uses no power connectors, and is an integrated graphics processor.

The Verdict

The data clearly favors the AMD Radeon RX 6650 XT for gaming and most graphics workloads. It wins 2 out of 3 head-to-head benchmarks, with massive margins in DirectX 12 and Vulkan. The 287.3% lead in 3DMark Steel Nomad DX12 and the 141.4% lead in Geekbench Vulkan are decisive. Its dedicated 8 GB of GDDR6 memory with 280.3 GB/s bandwidth provides consistent performance that the 780M cannot match with system-shared memory. The RX 6650 XT also has more than double the shading units, TMUs, and ROPs, giving it a clear hardware advantage.

The AMD Radeon 780M is the better choice only for specific OpenCL compute workloads. Its 45.6% lead in Geekbench OpenCL shows that RDNA 3.0 and the higher 2900 MHz boost clock can outperform the older RDNA 2.0 architecture in this particular test. The 780M also has a far lower 15 W TDP, making it suitable for systems where power efficiency is paramount. It is an active production part, while the RX 6650 XT is end-of-life.

The percentile data reinforces this split. The RX 6650 XT sits at the 65th percentile among all GPUs, with an average benchmark score of 19765. The 780M sits at the 61st percentile, with an average score of 17588. The RX 6650 XT's nearest rivals include the NVIDIA Tesla K40m and AMD FirePro W7000, with scores within 0.7% of its average. The 780M's nearest rivals include the NVIDIA GeForce RTX 4060 and AMD Radeon HD 7790, all within 0.5% of its average. The RX 6650 XT's higher percentile and average score confirm its overall superiority, despite the 780M's OpenCL win.

Specification Differences

| Specification | AMD Radeon RX 6650 XT | AMD Radeon 780M |

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

| Architecture | RDNA 2.0 | RDNA 3.0 |

| Chip | Navi 23 | Phoenix |

| Process Node | 7 nm | 4 nm |

| Transistors | 11,060 million | 25,390 million |

| Die Size | 237 mm² | 178 mm² |

| Transistor Density | 46.7M / mm² | 142.6M / mm² |

| Base Clock | 2055 MHz | 800 MHz |

| Boost Clock | 2635 MHz | 2900 MHz |

| Memory Size | 8 GB | System Shared |

| Memory Type | GDDR6 | System Shared |

| Memory Bus Width | 128 bit | System Shared |

| Memory Bandwidth | 280.3 GB/s | System Dependent |

| Shading Units | 2048 | 768 |

| TMUs | 128 | 48 |

| ROPs | 64 | 32 |

| Ray Tracing Cores | 32 | 12 |

| Pixel Rate | 168.6 GPixel/s | 92.80 GPixel/s |

| Texture Rate | 337.3 GTexel/s | 139.2 GTexel/s |

| FP32 | 10.79 TFLOPS | 8.909 TFLOPS |

| FP16 | 21.59 TFLOPS (2:1) | 8.909 TFLOPS (1:1) |

| TDP | 176 W | 15 W |

| Slot Width | Dual-slot | IGP |

| Power Connectors | 1x 8-pin | None |

| Suggested PSU | 450 W | None |

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

| Release Date | 2022-05-09 | 2024-01-30 |

| Launch MSRP | 399 USD | None |