AMD Radeon RX 590 vs AMD Radeon RX 6600M Comparison

The AMD Radeon RX 590 and the AMD Radeon RX 6600M are two very different interpretations of a "Radeon" GPU, separated by a generational leap in architecture and intended use case. The RX 590 is a legacy desktop card built for a previous era, while the RX 6600M is a modern mobile solution designed for efficiency. Based strictly on the data, the RX 6600M is the clear performance winner, dominating in both available head-to-head benchmarks with significant margins. However, the RX 590 still holds a niche for those who need a specific, power-hungry desktop component with a particular set of legacy features. The data suggests that if you prioritize raw performance in modern workloads, the RX 6600M is the only logical choice, but if you require a dual-slot desktop card with a specific power connector and a lower PCIe bus interface requirement, the RX 590 remains a viable, albeit older, option.

The Verdict

The benchmark data is unequivocal: the AMD Radeon RX 6600M is the superior performer. In the DirectX 12 test (3dmark_3dmark_steel_nomad_dx12), the RX 6600M scores 1495 points, which is 27.7% higher than the RX 590's score of 1081. This is not a marginal difference; it is a substantial generational leap. The gap widens further in the Geekbench Metal test, where the RX 6600M's score of 92237 is 47.5% higher than the RX 590's 48407. For any user whose primary concern is application performance in these API environments, the RX 6600M is the definitive choice.

The RX 590, however, is not without its reasons to exist. It is a dual-slot, 241 mm desktop card that requires a 1x 8-pin power connector and a 450 W suggested PSU. The RX 6600M, in contrast, is an IGP (Integrated Graphics Processor) with no power connectors and no length specified, making it exclusively a laptop or mobile part. If you are building a desktop tower and need a discrete GPU that slots into a PCIe 3.0 x16 slot, the RX 590 is the only one of the two that fits that physical description. The data shows the RX 590 has an average benchmark score of 24744, placing it in the 70th percentile of all GPUs, while the RX 6600M's average score is 23273, placing it in the 68th percentile. This indicates that despite the RX 6600M's wins in the head-to-head tests, the broader benchmark suite shows the RX 590 holding a slight edge in overall average performance, likely due to the nature of the tests included.

Architecture Differences

The architectural divide between these two GPUs is profound. The RX 590 is built on the GCN 4.0 architecture and uses the Polaris 30 chip, manufactured on a 12 nm process at GlobalFoundries. It contains 5,700 million transistors on a 232 mm² die, resulting in a transistor density of 24.6M per mm². Its feature set is anchored in the past, supporting DirectX 12 (12_0), OpenGL 4.6, and Vulkan 1.3.

In contrast, the RX 6600M is based on the RDNA 2.0 architecture with the Navi 23 chip, fabricated by TSMC on a much more advanced 7 nm process. This node shrink allows it to pack 11,060 million transistors into a similar 237 mm² die, achieving a significantly higher transistor density of 46.7M per mm². The architectural leap brings modern features, including support for DirectX 12 Ultimate (12_2) and a newer Vulkan 1.4. The RX 6600M also introduces dedicated ray tracing hardware with 28 RT cores, a feature entirely absent from the RX 590. The RX 6600M's FP16 performance is also doubled relative to its FP32 output (17.32 TFLOPS vs 8.659 TFLOPS), whereas the RX 590 has a 1:1 ratio (7.119 TFLOPS for both), indicating a design optimized for newer, more parallel workloads.

Head-to-Head Benchmarks

The data provided includes two direct benchmark comparisons, and the RX 6600M wins both decisively. In the 3dmark_3dmark_steel_nomad_dx12 test, the RX 6600M scores 1495 against the RX 590's 1081. This represents a 27.7% advantage for the RX 6600M. This is a significant lead in a modern DirectX 12 workload, showcasing the efficiency of the RDNA 2.0 architecture over the older GCN design.

The second comparison, geekbench_metal, shows an even larger gap. The RX 6600M scores 92237, while the RX 590 manages only 48407. The delta here is 47.5% in favor of the RX 6600M. This benchmark is particularly telling for Apple ecosystem or Metal API users, where the RX 6600M demonstrates a massive performance advantage. With the RX 590 winning zero head-to-head tests and the RX 6600M winning two, the performance hierarchy is clear. The RX 6600M does not just outperform; it outperforms by a wide margin in these specific, modern API tests.

FAQ

Q: Which GPU has higher raw compute performance (FP32)?

A: The AMD Radeon RX 6600M is stronger, with 8.659 TFLOPS of FP32 performance compared to the RX 590's 7.119 TFLOPS.

Q: Do both GPUs support the same version of DirectX?

A: No. The RX 6600M supports DirectX 12 Ultimate (12_2), while the RX 590 is limited to DirectX 12 (12_0).

Q: Which card has a higher memory bandwidth?

A: The RX 590 has a higher memory bandwidth of 256.0 GB/s, compared to the RX 6600M's 224.0 GB/s.

Q: Is the RX 6600M a desktop card?

A: No. The RX 6600M is an IGP (Integrated Graphics Processor) with no power connectors and "Portable Device Dependent" display outputs, indicating it is designed for laptops. The RX 590 is a dual-slot desktop card.

Q: How do their transistor counts compare?

A: The RX 6600M has nearly double the transistors, with 11,060 million, compared to the RX 590's 5,700 million.

Q: Which GPU has a higher average benchmark score?

A: The RX 590 has a higher average benchmark score of 24744, versus the RX 6600M's 23273, despite losing the two head-to-head tests.

Where Each One Wins

Based on the benchmark data, the RX 6600M is the clear winner in performance-critical scenarios. It wins the 3DMark Steel Nomad test by 27.7% and the Geekbench Metal test by 47.5%, making it the superior choice for any gaming or compute workload that utilizes DirectX 12 or the Metal API. Its architecture is also more future-proof, featuring 28 dedicated RT cores for ray tracing and support for DirectX 12 Ultimate. If you are looking for maximum performance in these modern APIs, the RX 6600M is the only choice.

The RX 590, conversely, wins in scenarios defined by its physical form factor and legacy software support. Its higher average benchmark score of 24744 suggests it performs better in a broader set of older or less API-specific tests. It is a dual-slot desktop card with a 241 mm length, a 1x 8-pin power connector, and a PCIe 3.0 x16 interface. This makes it the only option for a user who needs a discrete, desktop-installable GPU with those specific power and slot requirements. Its higher memory bandwidth (256.0 GB/s vs 224.0 GB/s) could also be a deciding factor in memory-bandwidth-sensitive tasks that are not represented in the head-to-head tests. Ultimately, the RX 590 wins on compatibility for older desktop systems, while the RX 6600M wins on raw modern performance.

Specification Differences

The specification sheets reveal a complete redesign between the two GPUs. The most striking difference is the process node: the RX 590 uses a 12 nm process, while the RX 6600M uses a 7 nm process. This leads to a massive difference in power consumption, with the RX 590 rated at 175 W TDP and the RX 6600M at just 100 W TDP. The physical dimensions also differ entirely; the RX 590 is a 241 mm dual-slot card, while the RX 6600M has no specified dimensions and is listed as an IGP. Power connectors follow suit, with the RX 590 requiring a single 8-pin connector and the RX 6600M requiring none.

The memory subsystems are also notably different. While both have 8 GB of memory, the RX 590 uses GDDR5 on a 256-bit bus, achieving 256.0 GB/s bandwidth. The RX 6600M uses faster GDDR6 memory but on a narrower 128-bit bus, resulting in 224.0 GB/s bandwidth. Clock speeds are dramatically higher on the RX 6600M, with a base clock of 2068 MHz and a boost of 2416 MHz, compared to the RX 590's 1469 MHz base and 1545 MHz boost. The bus interface also differs, with the RX 590 using PCIe 3.0 x16 and the RX 6600M using the newer PCIe 4.0 x8. Finally, the RX 6600M features 28 RT cores and a 2:1 FP16 ratio, neither of which is present on the RX 590. The RX 590 has a launch MSRP of 279 USD, while the RX 6600M has no stated MSRP.