AMD Radeon RX 590 vs NVIDIA GeForce GTX 1630 Comparison

The AMD Radeon RX 590 and NVIDIA GeForce GTX 1630 represent two very different approaches to the entry-level and mid-range GPU market, separated by nearly four years of release dates. The data shows a fascinating clash: the RX 590, built on a mature 12 nm Polaris 30 process with a massive 232 mm² die, goes head-to-head with the GTX 1630, a compact 200 mm² Turing-based part. Their average benchmark scores are remarkably close—24,744 for the RX 590 versus 24,277 for the GTX 1630—placing both in the 70th percentile of all GPUs. However, the specification sheets reveal starkly different philosophies in memory configuration, power delivery, and compute resources, making this comparison less about raw speed and more about how each card allocates its limited resources.

Head-to-Head Benchmarks

The overall average benchmark scores tell a story of near-parity, but the individual test results reveal distinct strengths. The AMD Radeon RX 590 posts an average benchmark score of 24,744, while the NVIDIA GeForce GTX 1630 trails at 24,277. This translates to a delta of 1.9% in favor of the RX 590 when comparing directly against the GTX 1630 as listed in its nearest rivals. In the broader context, the RX 590 sits just 0.1% behind the NVIDIA RTX A5000 Mobile (24,763) and 0.4% ahead of the Intel Arc A350M (24,647), showing it is firmly anchored in a competitive mid-range cluster. The GTX 1630, meanwhile, is 0.2% ahead of the NVIDIA GeForce GTX 780 Ti (24,236) and 0.4% ahead of the NVIDIA GeForce RTX 2080 SUPER (24,170), which is a surprising result given the RTX 2080 SUPER's higher-tier positioning in other contexts.

Looking at specific workloads, the RX 590 achieves a 3DMark Steel Nomad DX12 score of 1,081, which is its only DirectX 12 benchmark listed. The GTX 1630 counters with two Geekbench results: 24,858 in OpenCL and 23,695 in Vulkan. The OpenCL score for the GTX 1630 is actually higher than its own average benchmark score, suggesting the card performs particularly well in compute-oriented OpenCL tasks. The Vulkan score of 23,695 is notably lower than its OpenCL result, indicating a potential weakness in Vulkan API optimization. For the RX 590, the Geekbench Metal score of 48,407 is nearly double its average benchmark score, which reflects its strong performance in Apple's Metal API—a metric that the GTX 1630 does not have a comparable result for in this data.

The delta percentages against shared rivals are telling. Both cards are compared against the AMD Radeon RX 6600 XT: the RX 590 is 1.2% ahead of it (24,744 vs 24,442), while the GTX 1630 is 0.7% behind it (24,277 vs 24,442). This means the RX 590 holds a 1.9% advantage over the GTX 1630 in average score, and that advantage nearly doubles when both are measured against the RX 6600 XT as a reference point. The performance gap is consistent but narrow—not a landslide victory, but a measurable edge for the older AMD part in aggregate compute throughput.

Architecture Differences

The architectural divide between these two cards is substantial. The RX 590 uses the Polaris 30 chip based on GCN 4.0 architecture, manufactured by GlobalFoundries on a 12 nm process. It packs 5,700 million transistors into a 232 mm² die, yielding a transistor density of 24.6 million per square millimeter. The GTX 1630, by contrast, uses the TU117 chip based on Turing architecture, built by TSMC also on a 12 nm process, but with 4,700 million transistors in a smaller 200 mm² die, giving a density of 23.5 million per square millimeter. The RX 590 has a 21% transistor count advantage and a 16% die size advantage, which translates into significantly more raw compute hardware.

The compute resources diverge sharply. The RX 590 features 2,304 shading units, 144 texture mapping units, and 32 raster output pipelines. The GTX 1630 is far leaner with 512 shading units, 32 TMUs, and 16 ROPs. This 4.5x difference in shading units explains why the RX 590 delivers 7.119 TFLOPS of FP32 performance versus 1.828 TFLOPS for the GTX 1630—a 3.9x gap in raw floating-point throughput. The texture fill rate tells a similar story: 222.5 GTexel/s for the RX 590 versus 57.12 GTexel/s for the GTX 1630. Pixel rate favors the RX 590 at 49.44 GPixel/s versus 28.56 GPixel/s for the GTX 1630.

Memory architecture is where the GTX 1630 takes a different, more modern approach. The RX 590 uses 8 GB of GDDR5 on a 256-bit bus, delivering 256.0 GB/s of bandwidth. The GTX 1630 uses 4 GB of GDDR6 on a 64-bit bus, yielding only 96.00 GB/s. Despite having half the memory capacity and one-quarter of the bus width, the GTX 1630's GDDR6 runs at 12 Gbps effective versus 8 Gbps for the RX 590's GDDR5, which partially compensates but still leaves the RX 590 with 2.7x more memory bandwidth. In FP16 compute, the GTX 1630 shows a 2:1 ratio (3.656 TFLOPS) versus the RX 590's 1:1 ratio (7.119 TFLOPS), indicating the Turing card can double its throughput on half-precision workloads.

FAQ

Q: Which card has the higher average benchmark score?

A: The AMD Radeon RX 590 records an average benchmark score of 24,744, which is 1.9% higher than the NVIDIA GeForce GTX 1630's 24,277.

Q: Why does the RX 590 have such a large advantage in shading units?

A: The RX 590 contains 2,304 shading units compared to 512 on the GTX 1630, a 4.5x difference. This is reflected in FP32 performance, where the RX 590 delivers 7.119 TFLOPS versus 1.828 TFLOPS for the GTX 1630.

Q: Does the GTX 1630 have any performance advantage in API-specific tests?

A: Yes, in Geekbench OpenCL the GTX 1630 scores 24,858, which exceeds its average benchmark score of 24,277. Its Geekbench Vulkan score is lower at 23,695, while the RX 590's Geekbench Metal score of 48,407 is far above its average.

Q: How do the memory configurations affect real-world bandwidth?

A: The RX 590 offers 256.0 GB/s of bandwidth from 8 GB of GDDR5 on a 256-bit bus, versus 96.00 GB/s for the GTX 1630 from 4 GB of GDDR6 on a 64-bit bus. The RX 590's bandwidth is 2.7x higher.

Q: What is the power consumption difference?

A: The RX 590 has a TDP of 175 W and requires a 1x 8-pin power connector with a suggested 450 W PSU. The GTX 1630 has a TDP of 75 W, requires no power connectors, and uses a suggested 250 W PSU.

Q: Which card supports a newer Vulkan version?

A: The GTX 1630 supports Vulkan 1.4, while the RX 590 supports Vulkan 1.3. Both cards support DirectX 12, but the GTX 1630 supports feature level 12_1 versus 12_0 for the RX 590.

The Verdict

The data suggests a clear split based on workload priorities rather than a definitive overall winner. The RX 590 wins on aggregate compute performance, memory bandwidth, and raw throughput metrics—its 2,304 shading units, 7.119 TFLOPS FP32, and 256.0 GB/s bandwidth are decisive advantages that show up in its 1.9% higher average benchmark score. For users prioritizing DirectX 12 performance, the RX 590's 3DMark Steel Nomad score of 1,081 provides a concrete data point. The GTX 1630, however, offers superior efficiency with a 75 W TDP versus 175 W, no external power connector, and a single-slot form factor, making it suitable for systems with minimal power delivery. The GTX 1630 also supports newer API features: Vulkan 1.4 versus 1.3 and DirectX 12_1 versus 12_0.

The percentile ranking is identical at 70 for both cards, meaning they occupy the same tier among all GPUs. The nearest rival data reinforces this: the RX 590's closest competitor is the NVIDIA RTX A5000 Mobile (0.1% higher score), while the GTX 1630's closest is the NVIDIA GeForce GTX 780 Ti (0.2% higher). Neither card dominates its peer group, but the RX 590 consistently edges out the GTX 1630 in aggregate metrics. The GTX 1630's 4 GB memory capacity is half the RX 590's 8 GB, which could impact texture-heavy workloads, while its 64-bit bus limits memory bandwidth to 96.00 GB/s—a bottleneck that no amount of GDDR6 speed can fully overcome.

Specification Differences

| Specification | AMD Radeon RX 590 | NVIDIA GeForce GTX 1630 |

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

| Chip | Polaris 30 | TU117 |

| Architecture | GCN 4.0 | Turing |

| Foundry | GlobalFoundries | TSMC |

| Transistors | 5,700 million | 4,700 million |

| Die Size | 232 mm² | 200 mm² |

| Transistor Density | 24.6M / mm² | 23.5M / mm² |

| Base Clock | 1469 MHz | 1740 MHz |

| Boost Clock | 1545 MHz | 1785 MHz |

| Memory Clock | 2000 MHz / 8 Gbps effective | 1500 MHz / 12 Gbps effective |

| Memory Size | 8 GB | 4 GB |

| Memory Type | GDDR5 | GDDR6 |

| Memory Bus | 256 bit | 64 bit |

| Memory Bandwidth | 256.0 GB/s | 96.00 GB/s |

| Shading Units | 2304 | 512 |

| TMUs | 144 | 32 |

| ROPs | 32 | 16 |

| Pixel Rate | 49.44 GPixel/s | 28.56 GPixel/s |

| Texture Rate | 222.5 GTexel/s | 57.12 GTexel/s |

| FP32 | 7.119 TFLOPS | 1.828 TFLOPS |

| FP16 | 7.119 TFLOPS (1:1) | 3.656 TFLOPS (2:1) |

| TDP | 175 W | 75 W |

| Slot Width | Dual-slot | Single-slot |

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

| Suggested PSU | 450 W | 250 W |

| Display Outputs | 1x HDMI 2.0b, 3x DisplayPort 1.4a | 1x DVI, 1x HDMI 2.0, 1x DisplayPort 1.4a |

| DirectX | 12 (12_0) | 12 (12_1) |

| Vulkan | 1.3 | 1.4 |

| Length | 241 mm (9.5 inches) | 145 mm (5.7 inches) |

| Release Date | 2018-11-14 | 2022-06-27 |

Where Each One Wins

The RX 590 wins decisively in any metric involving raw compute throughput. Its texture rate of 222.5 GTexel/s is 3.9x higher than the GTX 1630's 57.12 GTexel/s, making it better suited for games and applications that are fill-rate limited. The 2.7x memory bandwidth advantage (256.0 GB/s versus 96.00 GB/s) gives it a clear edge in high-resolution textures and large data sets. The 8 GB memory capacity is double the GTX 1630's 4 GB, reducing the likelihood of running out of VRAM in modern titles. Its 3DMark Steel Nomad DX12 score of 1,081 is the only direct DX12 benchmark available, and its Geekbench Metal score of 48,407 suggests strong Apple ecosystem performance.

The GTX 1630 wins on efficiency and compactness. Its 75 W TDP is less than half the RX 590's 175 W, and it requires no power connectors, making it drop-in compatible with older or low-wattage systems. The single-slot design and 145 mm length (versus 241 mm for the RX 590) allow it to fit in small form factor cases. Its higher base and boost clocks (1740 MHz and 1785 MHz versus 1469 MHz and 1545 MHz) indicate better clock-for-clock efficiency. The GTX 1630's Vulkan 1.4 support and DirectX 12_1 feature level are more modern than the RX 590's Vulkan 1.3 and DirectX 12_0, which may matter for future API adoption. Its Geekbench OpenCL score of 24,858 exceeds its own average, suggesting OpenCL compute tasks are a relative strength, and its GDDR6 memory operates at a higher effective speed of 12 Gbps even though the bus is narrow.