AMD Radeon RX 590 GME vs AMD Radeon RX 7900 GRE Comparison

The RX 590 GME and RX 7900 GRE are two AMD graphics cards from different eras, and the benchmark data reflects a complete generational shift. The 7900 GRE wins every single head-to-head test, but the magnitude of those wins varies dramatically depending on the workload. This analysis breaks down exactly where the 7900 GRE dominates, where the gap narrows, and what that means for your build.

Head-to-Head Benchmarks

The most decisive victory for the RX 7900 GRE comes in the 3DMark Steel Nomad DX12 test, where it scores 4,814 points against the RX 590 GME’s 1,001 points. That is a 79.2% advantage for the newer card, meaning the 7900 GRE delivers nearly five times the raw DX12 performance. This is the kind of result that separates modern GPUs from older architectures, and it is the single largest margin in the entire comparison.

The Geekbench OpenCL test tells a similar story, with the 7900 GRE posting 175,758 points versus 36,294 for the 590 GME. The delta here is 79.4%, essentially identical to the 3DMark result. OpenCL is a compute-heavy workload, and the 7900 GRE’s massive lead shows that its advantage is not just in gaming but also in general-purpose GPU compute tasks. For anyone considering the 590 GME for rendering or scientific workloads, this gap is a dealbreaker.

Interestingly, the Vulkan benchmark shows a much smaller gap. The 7900 GRE scores 99,850 points, while the 590 GME manages 60,508 points. That is a 39.4% lead for the 7900 GRE — still a solid win, but less than half the margin seen in the other two tests. This suggests that the older GCN architecture holds up relatively better in Vulkan workloads, or that the test itself is less sensitive to the raw compute advantages of the newer card. It is also importantly both cards share the same 77th percentile ranking among all GPUs, yet their average benchmark scores are nearly identical: the 590 GME averages 32,601 points, while the 7900 GRE averages 32,456 points. This puts them within 0.4% of each other in average score, with the 590 GME technically ahead, despite losing every head-to-head test. This is a quirk of the different benchmark suites included in each card’s profile.

Architecture Differences

The architectural gap between these two cards is enormous, starting with the manufacturing process. The RX 590 GME uses a 14 nm process from GlobalFoundries, while the RX 7900 GRE uses a 5 nm process from TSMC. That node difference alone explains a significant portion of the performance and efficiency gap. The 590 GME’s chip, Polarus 20, is built on GCN 4.0 architecture, while the 7900 GRE uses Navi 31 based on RDNA 3.0. These are fundamentally different designs separated by two full generations.

The transistor counts tell the story of scale. The 590 GME has 5,700 million transistors on a 232 mm² die, giving a density of 24.6 million transistors per square millimeter. The 7900 GRE packs 57,700 million transistors onto a 529 mm² die, a density of 109.1 million per square millimeter. That is roughly ten times the transistor count and over four times the density. The 7900 GRE also uses a chiplet design with the codename "Plum Bonito," whereas the 590 GME is a monolithic chip.

Memory is another major divergence. The 590 GME comes with 8 GB of GDDR5 on a 256-bit bus, delivering 256.0 GB/s of bandwidth. The 7900 GRE has 16 GB of GDDR6 on the same 256-bit bus, but bandwidth jumps to 576.0 GB/s. The newer card also supports PCIe 4.0 x16, while the older card is limited to PCIe 3.0 x16. Display outputs differ as well: the 590 GME has 1x HDMI 2.0b and 3x DisplayPort 1.4a, while the 7900 GRE offers 1x HDMI 2.1a, 2x DisplayPort 2.1, and 1x USB Type-C.

The compute resources are not even close. The 590 GME has 2,304 shading units, 144 texture mapping units, and 32 ROPs. The 7900 GRE has 5,120 shading units, 320 TMUs, and 160 ROPs. The 7900 GRE also has 80 ray tracing cores, a feature entirely absent from the 590 GME. Pixel rate is 45.44 GPixel/s for the 590 GME versus 359.2 GPixel/s for the 7900 GRE, and texture rate is 204.5 GTexel/s versus 718.4 GTexel/s. FP32 performance is 6.543 TFLOPS for the old card and 45.98 TFLOPS for the new one. The FP16 numbers are even more lopsided: 6.543 TFLOPS (1:1) versus 91.96 TFLOPS (2:1).

Where Each One Wins

The RX 7900 GRE wins in every scenario where raw performance matters. In DX12 gaming, the 3DMark Steel Nomad result of 4,814 versus 1,001 means the 7900 GRE is the only viable choice for modern titles at high settings. The OpenCL compute advantage of 175,758 versus 36,294 makes it the clear pick for any productivity workload, including video encoding, 3D rendering, or machine learning inference. The 16 GB of memory versus 8 GB also gives it a significant edge in texture-heavy games or large datasets.

The RX 590 GME does not win any benchmark outright, but its Vulkan result of 60,508 points shows it is not completely obsolete. For older DX11 or Vulkan titles, or for users running at 1080p with lower detail settings, the 590 GME can still deliver playable frame rates. Its lower 175 W TDP versus the 7900 GRE’s 260 W also makes it easier to cool and requires only a 450 W suggested PSU versus 600 W. The 590 GME is also physically shorter at 241 mm versus 276 mm, which matters in compact cases.

The 7900 GRE’s advantage in the average benchmark score is minimal — it actually trails the 590 GME by 0.4% in that metric. However, this is misleading because the 7900 GRE has a much larger benchmark suite that includes older DirectX 9, 10, and 11 tests where it scores relatively low. For example, it scores 310 in Passmark DX9 and 139 in DX10, but 27,089 in Passmark G3D. The 590 GME does not have those older test scores in its profile, which skews the average.

FAQ

Q: Which card has better ray tracing performance?

A: The RX 7900 GRE has 80 ray tracing cores, while the RX 590 GME has none. The 7900 GRE is the only card of the two that supports hardware-accelerated ray tracing.

Q: How much faster is the 7900 GRE in DX12?

A: In the 3DMark Steel Nomad DX12 test, the 7900 GRE scores 4,814 points versus 1,001 for the 590 GME, a 79.2% difference.

Q: Is the memory bandwidth difference significant?

A: Yes. The 7900 GRE has 576.0 GB/s of bandwidth versus 256.0 GB/s for the 590 GME, a 320 GB/s gap that heavily impacts high-resolution textures and compute workloads.

Q: Do both cards support the same DirectX version?

A: No. The 7900 GRE supports DirectX 12 Ultimate (12_2), while the 590 GME only supports DirectX 12 (12_0).

Q: Which card is more power efficient?

A: The 590 GME has a 175 W TDP versus 260 W for the 7900 GRE. However, the 7900 GRE delivers far more performance per watt given its massive lead in every benchmark.

Q: Are the average scores really that close?

A: Yes, the 590 GME averages 32,601 points and the 7900 GRE averages 32,456 points, a 0.4% difference. This is because the 7900 GRE includes older DirectX 9/10/11 tests in its benchmark profile that drag down its average.

The Verdict

The data is unambiguous: the RX 7900 GRE is the superior card in every head-to-head test. Its 79.2% lead in 3DMark Steel Nomad and 79.4% lead in Geekbench OpenCL make it the only option for anyone building a modern gaming or workstation PC. The 16 GB memory buffer, 576.0 GB/s bandwidth, and 80 ray tracing cores are all features the 590 GME simply cannot match. If you are buying new, the 7900 GRE is the obvious choice, and its launch MSRP of 549 USD places it in a different class entirely.

The RX 590 GME is not without merit, but its strengths are limited to legacy scenarios. Its Vulkan score of 60,508 is respectable, and its lower power draw and shorter length make it easier to fit into small form factor builds. For a budget-conscious builder who only plays older titles or does light productivity work, the 590 GME could still serve. However, its 8 GB memory and 256.0 GB/s bandwidth are severe bottlenecks for anything modern. The 7900 GRE’s average score being lower than the 590 GME’s is a statistical artifact, not a real performance indicator — the head-to-head results are what matter, and they are a total sweep.

Specification Differences

| Specification | AMD Radeon RX 590 GME | AMD Radeon RX 7900 GRE |

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

| Architecture | GCN 4.0 | RDNA 3.0 |

| Process Node | 14 nm | 5 nm |

| Foundry | GlobalFoundries | TSMC |

| Transistors | 5,700 million | 57,700 million |

| Die Size | 232 mm² | 529 mm² |

| Transistor Density | 24.6M / mm² | 109.1M / mm² |

| Base Clock | 1257 MHz | 1287 MHz |

| Boost Clock | 1420 MHz | 2245 MHz |

| Game Clock | — | 1880 MHz |

| Memory Clock | 2000 MHz (8 Gbps effective) | 2250 MHz (18 Gbps effective) |

| Memory Size | 8 GB | 16 GB |

| Memory Type | GDDR5 | GDDR6 |

| Bus Width | 256 bit | 256 bit |

| Memory Bandwidth | 256.0 GB/s | 576.0 GB/s |

| Shading Units | 2304 | 5120 |

| TMUs | 144 | 320 |

| ROPs | 32 | 160 |

| RT Cores | — | 80 |

| Pixel Rate | 45.44 GPixel/s | 359.2 GPixel/s |

| Texture Rate | 204.5 GTexel/s | 718.4 GTexel/s |

| FP32 Performance | 6.543 TFLOPS | 45.98 TFLOPS |

| FP16 Performance | 6.543 TFLOPS (1:1) | 91.96 TFLOPS (2:1) |

| TDP | 175 W | 260 W |

| Power Connectors | 1x 8-pin | 2x 8-pin |

| Suggested PSU | 450 W | 600 W |

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

| Display Outputs | 1x HDMI 2.0b, 3x DisplayPort 1.4a | 1x HDMI 2.1a, 2x DisplayPort 2.1, 1x USB Type-C |

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

| Vulkan Support | 1.3 | 1.4 |

| Length | 241 mm (9.5 inches) | 276 mm (10.9 inches) |

| Height | — | 110 mm (4.3 inches) |

| Width | — | 51 mm (2 inches) |

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

| Release Date | 2020-03-08 | 2023-07-26 |