AMD Radeon RX 470 vs AMD Radeon RX 590 GME Comparison

The Verdict

The AMD Radeon RX 590 GME is the clear performance winner in this comparison, taking all three head-to-head benchmarks. The data shows it leads by 18.9% in 3DMark Steel Nomad DX12, 8.1% in Geekbench OpenCL, and a massive 51.7% in Geekbench Vulkan. Its average benchmark score of 32601 sits 12.4% above the RX 470's 28996.

The RX 590 GME is the pick for anyone who wants higher frame rates in modern APIs like Vulkan and DX12, where its advantage is substantial. The RX 470, despite being the older and slower card, still holds its own in compute workloads, trailing by only 8.1% in OpenCL. It also carries a lower power draw of 120 W versus 175 W, making it a reasonable choice for systems with a 300 W suggested PSU instead of 450 W.

Neither card is current production; both are end-of-life. The RX 590 GME launched on March 8, 2020, while the RX 470 arrived on August 3, 2016. If the choice is purely about performance, the RX 590 GME wins decisively. If the choice is about fitting into an existing power envelope with a 6-pin connector, the RX 470 is the only option. For a new build, the data favors the RX 590 GME without question.

Architecture Differences

Both cards share the same fundamental architecture: GCN 4.0 on a 14 nm process at GlobalFoundries. Both use 5,700 million transistors on a 232 mm² die, giving both a transistor density of 24.6M per mm². The silicon heritage is identical, but the implementations differ.

The RX 590 GME uses the Polaris 20 chip, while the RX 470 uses Ellesmere. Both are part of the Polaris family, but they belong to different generations: the RX 590 GME is in the Polar is (RX 500) generation, and the RX 470 is in the Arctic Islands (RX 400) generation. The predecessor/successor lineage also differs: the RX 590 GME follows Arctic Islands and leads to Vega, while the RX 470 follows Pirate Islands and leads to Polaris.

The RX 590 GME has more execution resources. It packs 2304 shading units, 144 texture mapping units, and 32 ROPs. The RX 470 has 2048 shading units, 128 TMUs, and the same 32 ROPs. That means the RX 590 GME has 12.5% more shaders and 12.5% more TMUs, which directly translates into higher fill rates and compute throughput.

Clock speeds are significantly higher on the RX 590 GME. Its base clock is 1257 MHz and boost reaches 1420 MHz, versus 926 MHz base and 1206 MHz boost on the RX 470. The memory subsystem also differs: the RX 590 GME runs 8 GB of GDDR5 at 2000 MHz (8 Gbps effective) for 256.0 GB/s bandwidth, while the RX 470 runs 4 GB of GDDR5 at 1650 MHz (6.6 Gbps effective) for 211.2 GB/s. Both use a 256-bit bus.

The RX 590 GME has a higher TDP of 175 W and requires a single 8-pin power connector, with a suggested PSU of 450 W. The RX 470 draws 120 W, uses a single 6-pin connector, and needs only a 300 W PSU. Both are dual-slot cards with identical display outputs: 1x HDMI 2.0b and 3x DisplayPort 1.4a. Both support DirectX 12 (12_0), OpenGL 4.6, and Vulkan 1.3.

Physical dimensions are nearly identical: the RX 590 GME is 241 mm (9.5 inches) long, and the RX 470 is 240 mm (9.4 inches) long. The RX 470 also has recorded height of 95 mm (3.7 inches) and width of 35 mm (1.4 inches); the RX 590 GME lacks those measurements in the database.

FAQ

Q: Which card is faster in Vulkan workloads?

A: The RX 590 GME is dramatically faster. In Geekbench Vulkan, it scores 60508 versus 39884 for the RX 470, a 51.7% advantage. This is the largest single benchmark gap between the two cards.

Q: How do the cards compare in DirectX 12 performance?

A: The RX 590 GME wins in 3DMark Steel Nomad DX12 with a score of 1001 versus 842 for the RX 470, an 18.9% lead. This is a solid, but not overwhelming, advantage in a modern DX12 test.

Q: Is the RX 470 still competitive in compute tasks?

A: Yes, relatively speaking. In Geekbench OpenCL, the RX 470 scores 33568 against the RX 590 GME's 36294, a difference of only 8.1%. This is the closest margin between the two cards in any recorded test.

Q: What is the power requirement difference?

A: The RX 590 GME has a 175 W TDP and needs a 450 W PSU with a single 8-pin connector. The RX 470 has a 120 W TDP, requires only a 300 W PSU, and uses a single 6-pin connector. That is a 55 W TDP gap and a 150 W PSU gap.

Q: Do both cards support the same modern APIs?

A: Yes. Both support DirectX 12 (12_0), OpenGL 4.6, and Vulkan 1.3. There is no difference in API feature level between the two.

Q: How much more memory bandwidth does the RX 590 GME have?

A: The RX 590 GME delivers 256.0 GB/s, versus 211.2 GB/s for the RX 470, a 21.2% increase. It also has double the memory capacity at 8 GB versus 4 GB.

Specification Differences

The following specifications differ between the AMD Radeon RX 590 GME and the AMD Radeon RX 470:

• Chip: Polaris 20 (RX 590 GME) versus Ellesmere (RX 470)
• Generation: Polaris (RX 500) versus Arctic Islands (RX 400)
• Base clock: 1257 MHz versus 926 MHz
• Boost clock: 1420 MHz versus 1206 MHz
• Memory clock: 2000 MHz (8 Gbps effective) versus 1650 MHz (6.6 Gbps effective)
• Memory size: 8 GB versus 4 GB
• Memory bandwidth: 256.0 GB/s versus 211.2 GB/s
• Shading units: 2304 versus 2048
• Texture mapping units: 144 versus 128
• Pixel rate: 45.44 GPixel/s versus 38.59 GPixel/s
• Texture rate: 204.5 GTexel/s versus 154.4 GTexel/s
• FP32 performance: 6.543 TFLOPS versus 4.940 TFLOPS
• TDP: 175 W versus 120 W
• Power connectors: 1x 8-pin versus 1x 6-pin
• Suggested PSU: 450 W versus 300 W
• Dimensions: Length 241 mm (9.5 inches) versus 240 mm (9.4 inches); height and width only recorded for the RX 470 (95 mm x 35 mm)
• Release date: March 8, 2020 versus August 3, 2016
• Launch MSRP: The RX 470 had a launch MSRP of 179 USD; the RX 590 GME has no recorded launch MSRP
• Predecessor/successor: Arctic Islands to Vega versus Pirate Islands to Polaris

Identical specifications include: architecture (GCN 4.0), process node (14 nm), foundry (GlobalFoundries), transistor count (5,700 million), die size (232 mm²), transistor density (24.6M / mm²), ROPs (32), memory type (GDDR5), bus width (256 bit), slot width (Dual-slot), bus interface (PCIe 3.0 x16), display outputs (1x HDMI 2.0b, 3x DisplayPort 1.4a), DirectX version (12_0), OpenGL version (4.6), Vulkan version (1.3), and production status (End-of-life).

Head-to-Head Benchmarks

The database records three direct comparisons between these cards, and the RX 590 GME wins all three.

The largest margin is in Geekbench Vulkan. The RX 590 GME scores 60508, while the RX 470 scores 39884. That is a 51.7% difference, which is more than half again as fast. This is a decisive win that indicates the RX 590 GME handles Vulkan titles and workloads far better, likely due to its higher clocks and additional shader units.

In 3DMark Steel Nomad DX12, the RX 590 GME scores 1001 against 842 for the RX 470, an 18.9% advantage. This test represents a modern DX12 gaming load, and the gap is meaningful but not as extreme as Vulkan. The RX 590 GME's higher pixel rate of 45.44 GPixel/s versus 38.59 GPixel/s and texture rate of 204.5 GTexel/s versus 154.4 GTexel/s contribute to this result.

The closest contest is in Geekbench OpenCL. The RX 590 GME scores 36294, and the RX 470 scores 33568, a delta of 8.1%. This smaller gap suggests that raw compute throughput, while higher on the RX 590 GME (6.543 TFLOPS versus 4.940 TFLOPS), does not translate into an equally large advantage in all compute tasks. The RX 470's 4 GB of memory and lower bandwidth (211.2 GB/s) may limit it, but the 8.1% difference is still a clear win for the RX 590 GME.

Overall, the RX 590 GME has 3 wins and the RX 470 has 0 wins in head-to-head testing. The average benchmark scores confirm the trend: 32601 for the RX 590 GME versus 28996 for the RX 470.

Where Each One Wins

The RX 590 GME wins outright in every recorded benchmark category. Its strongest showing is in Vulkan, where the 51.7% lead makes it the obvious choice for any user prioritizing Vulkan-based games or applications. It also leads in DX12 by 18.9%, making it the better option for modern gaming. With 8 GB of memory, it is also better suited for texture-heavy workloads at higher resolutions, though the database does not directly measure that.

The RX 470's only advantage lies in power efficiency and system integration. It draws 55 W less (120 W versus 175 W), requires only a 300 W PSU, and uses a 6-pin connector instead of an 8-pin. For a user upgrading an older system with a limited power supply, the RX 470 is the only viable choice. Its 240 mm length and 95 mm height make it a compact card, and the recorded dimensions show it fits in tight cases.

In compute workloads, the RX 470 is not far behind. The 8.1% OpenCL gap means that for OpenCL-based tasks, the RX 470 delivers nearly the same performance while drawing significantly less power. This makes it a reasonable pick for a secondary compute card or a low-power build.

For gaming, the choice is unambiguous: the RX 590 GME is faster in both DX12 and Vulkan. The RX 470's 74th percentile versus the RX 590 GME's 77th percentile among all GPUs reflects the overall performance gap. The RX 590 GME also sits close to much newer cards in the database, with a 0.4% delta to the AMD Radeon RX 7900 GRE and a 0.2% delta to the AMD FirePro S9300 X2, showing it remains competitive despite its age. The RX 470 sits near the AMD Radeon RX 6800M (0.4% delta) and Intel Arc A370M (-0.6% delta), placing it in a lower performance tier.

In summary, the RX 590 GME is the performance pick, the RX 470 is the efficiency pick, and there is no scenario in the recorded data where the RX 470 wins a benchmark.