AMD Radeon RX 580 vs NVIDIA GeForce GTX 950 Comparison

The benchmark data is unambiguous: the AMD Radeon RX 580 decisively outperforms the NVIDIA GeForce GTX 950 across every shared test, making it a fundamentally higher-performing GPU. While both cards occupy the same 53rd percentile among all GPUs, their average benchmark scores tell a story of separation, with the RX 580’s 12,928 average edging out the GTX 950’s 13,189 by a narrow margin — a statistic that belies the massive individual test deltas. The GTX 950’s nearest rivals include the GeForce GTX 480 (0.8% behind) and the Radeon Pro 555X (1% ahead), while the RX 580 sits within 0.5% of the Radeon 740M and 0.6% of the FirePro W5100, indicating that the 580 competes in a different performance tier entirely.

Head-to-Head Benchmarks

The Radeon RX 580 wins all three head-to-head comparisons, and it does so by enormous margins. In Geekbench Metal, the RX 580 scores 45,235 against the GTX 950’s 7,172, a delta of -84.1% from the NVIDIA card’s perspective — meaning the RX 580 is roughly 5.3 times faster in this API. This is not a marginal generational improvement; it is a complete overhaul of compute capability.

The OpenCL results reinforce the pattern. The RX 580 posts 37,453 points versus the GTX 950’s 15,662, a -58.2% delta. This 2.4x advantage in OpenCL suggests the AMD card’s architecture is far better suited to general-purpose compute workloads, which is consistent with its much larger shader array and higher throughput figures. The GTX 950’s 1.825 TFLOPS of FP32 performance is dwarfed by the RX 580’s 6.175 TFLOPS, a 3.4x gap that shows up directly in these synthetic tests.

Vulkan benchmarks show a similar chasm. The RX 580 achieves 45,173 while the GTX 950 manages 16,734, a -63% delta. This 2.7x improvement in Vulkan is critical for modern gaming, as Vulkan-based titles can leverage the RX 580’s extra shading units and texture units far more effectively. The data shows zero wins for the GTX 950 across all three tests — the NVIDIA card simply cannot compete in any compute or graphics API where both are measured.

Architecture Differences

The two GPUs come from different architectural generations and manufacturing processes. The GTX 950 uses the GM206 chip based on Maxwell 2.0 architecture, fabricated on a 28 nm process at TSMC. The RX 580 employs the Polaris 20 chip with GCN 4.0 architecture, built on a 14 nm process at GlobalFoundries. This node transition is significant: the RX 580 packs 5,700 million transistors into a 232 mm² die, achieving a transistor density of 24.6 million per mm², while the GTX 950 has 2,940 million transistors on a larger 228 mm² die, yielding just 12.9 million per mm².

The compute resources differ dramatically. The RX 580 has 2,304 shading units, 144 texture mapping units, and 32 ROPs, compared to the GTX 950’s 768 shading units, 48 TMUs, and 32 ROPs. This 3x increase in shading units and TMUs directly explains the RX 580’s superior texture rate of 193.0 GTexel/s versus the GTX 950’s 57.02 GTexel/s. Pixel rates are closer — 42.88 GPixel/s for the AMD part versus 38.02 GPixel/s for NVIDIA — because both cards have 32 ROPs, though the RX 580’s higher clock speeds push its pixel throughput higher.

Memory configurations are also generationally apart. The RX 580 offers 8 GB of GDDR5 on a 256-bit bus, delivering 256.0 GB/s of bandwidth, while the GTX 950 has 2 GB of GDDR5 on a 128-bit bus with 105.8 GB/s. Effective memory speed is 8 Gbps on the AMD card versus 6.6 Gbps on the NVIDIA card. The RX 580 also supports FP16 at a 1:1 ratio with FP32, both rated at 6.175 TFLOPS, whereas the GTX 950 lists no FP16 capability.

API support shows minor differences. Both support DirectX 12, but the GTX 950 is rated at feature level 12_1 while the RX 580 is at 12_0. OpenGL is 4.6 for both, and Vulkan support is 1.4 on NVIDIA versus 1.3 on AMD. The RX 580’s display outputs include HDMI 2.0b and three DisplayPort 1.4a connectors, while the GTX 950 offers DVI, HDMI 2.0, and three DisplayPort 1.2 ports.

FAQ

Q: Which GPU has higher raw compute performance?

A: The Radeon RX 580 has 6.175 TFLOPS of FP32 performance, which is 3.4 times the GTX 950’s 1.825 TFLOPS. This aligns with its 2,304 shading units versus 768 on the NVIDIA card.

Q: How do memory bandwidth figures compare?

A: The RX 580 provides 256.0 GB/s of bandwidth over a 256-bit bus with 8 GB of GDDR5, while the GTX 950 offers 105.8 GB/s over a 128-bit bus with 2 GB. The AMD card has 2.4x the bandwidth and 4x the memory capacity.

Q: Are these cards in the same performance percentile?

A: Both GPUs are in the 53rd percentile among all GPUs. However, the average benchmark scores are close (13,189 for GTX 950, 12,928 for RX 580), which masks the RX 580’s decisive wins in individual tests.

Q: What is the power draw difference?

A: The GTX 950 has a TDP of 90 W with a suggested PSU of 250 W, while the RX 580 has a TDP of 185 W with a suggested PSU of 450 W. The RX 580 requires one 8-pin power connector versus the GTX 950’s single 6-pin.

Q: Which card has better Vulkan performance?

A: The RX 580 scores 45,173 in Geekbench Vulkan, which is 2.7 times the GTX 950’s 16,734. The AMD card also supports Vulkan 1.3, while the NVIDIA card supports Vulkan 1.4.

Q: How do their physical sizes compare?

A: The RX 580 is longer at 241 mm (9.5 inches) versus the GTX 950’s 202 mm (8 inches). Both are dual-slot cards, but the RX 580 requires more case clearance.

Specification Differences

• Process Node: GTX 950 is 28 nm; RX 580 is 14 nm
• Transistors: 2,940 million vs 5,700 million
• Die Size: 228 mm² vs 232 mm²
• Transistor Density: 12.9M/mm² vs 24.6M/mm²
• Base Clock: 1024 MHz vs 1257 MHz
• Boost Clock: 1188 MHz vs 1340 MHz
• Memory Clock: 6.6 Gbps effective vs 8 Gbps effective
• Memory Size: 2 GB vs 8 GB
• Memory Bus: 128-bit vs 256-bit
• Memory Bandwidth: 105.8 GB/s vs 256.0 GB/s
• Shading Units: 768 vs 2304
• TMUs: 48 vs 144
• Pixel Rate: 38.02 GPixel/s vs 42.88 GPixel/s
• Texture Rate: 57.02 GTexel/s vs 193.0 GTexel/s
• FP32: 1.825 TFLOPS vs 6.175 TFLOPS
• FP16: Not listed vs 6.175 TFLOPS (1:1)
• TDP: 90 W vs 185 W
• Power Connectors: 1x 6-pin vs 1x 8-pin
• Suggested PSU: 250 W vs 450 W
• Display Outputs: DVI, HDMI 2.0, 3x DP 1.2 vs HDMI 2.0b, 3x DP 1.4a
• DirectX: 12 (12_1) vs 12 (12_0)
• Vulkan: 1.4 vs 1.3
• Length: 202 mm vs 241 mm
• Release Date: 2015-08-19 vs 2017-04-17

Where Each One Wins

The Radeon RX 580 wins every benchmark where both cards are measured, so the use-case split is heavily lopsided. In Geekbench Metal, the RX 580’s 45,235 score versus 7,172 makes it the clear choice for Apple ecosystem compute workloads. The OpenCL result (37,453 vs 15,662) and Vulkan result (45,173 vs 16,734) similarly favor the AMD card for cross-platform compute and modern graphics APIs.

For gaming at higher resolutions or with texture-heavy settings, the RX 580’s 8 GB memory and 256-bit bus are decisive advantages. The GTX 950’s 2 GB frame buffer will bottleneck in modern titles, while the RX 580’s higher texture rate (193.0 GTexel/s) and shading unit count (2304) handle complex shaders more effectively. The RX 580’s 3.4x FP32 throughput also makes it superior for content creation, video encoding, and any GPU-accelerated productivity tasks.

The GTX 950’s only advantages are operational rather than performance-based. Its 90 W TDP means it can run in systems with a 250 W PSU, fitting into compact builds where the RX 580’s 185 W TDP and 450 W PSU requirement would be impractical. It is also shorter at 202 mm, suiting smaller cases. The GTX 950 supports Vulkan 1.4 versus the RX 580’s 1.3, though the performance gap in Vulkan tests makes this a minor point. For users with legacy 6-pin power supplies or tight thermal constraints, the GTX 950 remains a viable low-power option, but it loses on every measurable performance metric.

The Verdict

The data points to one conclusion: the Radeon RX 580 is the superior GPU for anyone prioritizing performance. It wins all three head-to-head benchmarks by margins ranging from 58.2% to 84.1%, and its architectural advantages are overwhelming — 3x the shading units, 2.4x the memory bandwidth, 3.4x the FP32 throughput. The RX 580’s 8 GB memory capacity is four times the GTX 950’s, making it suitable for modern workloads that the older card cannot handle.

The GTX 950 is only defensible in scenarios where power consumption is the primary constraint. Its 90 W TDP allows operation with a 250 W PSU, and its smaller 202 mm length fits in compact enclosures. However, benchmark results show it trails the RX 580 by 84.1% in Metal, 58.2% in OpenCL, and 63% in Vulkan — gaps that no efficiency advantage can offset for gaming or compute.

Users should choose the RX 580 for any GPU-intensive task, from gaming to rendering to machine learning inference. The GTX 950 is a legacy part that remains functional for basic 2D workloads or low-power HTPC builds, but the benchmark database shows it is outclassed by the AMD card in every measurable way. For those who must run on a 250 W PSU, the GTX 950 is the only option of the two; for everyone else, the RX 580 is the verdict-driven pick.