AMD Radeon RX 5500M vs NVIDIA P106-090 Comparison

# NVIDIA P106-090 vs AMD Radeon RX 5500M

The NVIDIA P106-090 and AMD Radeon RX 5500M occupy the same performance percentile (54th vs 54th), yet they are fundamentally different products separated by two years of architectural evolution. The P106-090 is a mining-focused Pascal card with no display outputs, while the RX 5500M is a mobile RDNA 1.0 part. Benchmark results show the RX 5500M dominates in compute workloads, but the P106-090's legacy in the mining GPU generation makes it a niche player. The data reveals a 45% gap in OpenCL performance and a 47.9% gap in Vulkan performance, both favoring AMD, yet the average benchmark scores remain close (13,470 vs 13,356), suggesting the P106-090's strengths lie elsewhere.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The NVIDIA P106-090 has an average benchmark score of 13,470, which is 114 points higher than the AMD Radeon RX 5500M's 13,356. This places both in the 54th percentile of all GPUs, and the P106-090's nearest rival is the NVIDIA GeForce GTX 570 (deltaPct -0.3), while the RX 5500M's closest competitor is the AMD FirePro M6100 (deltaPct 0).

Q: How do the two GPUs compare in OpenCL performance?

A: The AMD Radeon RX 5500M scores 38,725 in Geekbench OpenCL, which is 45% higher than the NVIDIA P106-090's 21,304. This is the largest single benchmark delta between the two, indicating a substantial compute advantage for the AMD part.

Q: What is the memory configuration difference?

A: The P106-090 uses 3 GB of GDDR5 on a 192-bit bus with 192.2 GB/s bandwidth, while the RX 5500M uses 4 GB of GDDR6 on a 128-bit bus with 224.0 GB/s bandwidth. Despite the narrower bus, the RX 5500M achieves higher memory bandwidth due to faster 14 Gbps effective memory speed versus 8 Gbps on the P106-090.

Q: Which GPU consumes more power?

A: The AMD Radeon RX 5500M has a TDP of 85 W, which is 10 W higher than the NVIDIA P106-090's 75 W TDP. The P106-090 requires a 250 W suggested PSU and a 1x 6-pin power connector, while the RX 5500M has no power connectors (portable device dependent).

Q: Are there any display output differences?

A: Yes, the NVIDIA P106-090 has no display outputs at all, reflecting its mining-oriented design. The AMD Radeon RX 5500M's display outputs are "Portable Device Dependent," meaning they vary by laptop implementation.

Q: What is the release timeline for each product?

A: The NVIDIA P106-090 was released on July 30, 2017, while the AMD Radeon RX 5500M launched on October 6, 2019. Both are now end-of-life products, with the RX 5500M succeeding the Polaris Mobile architecture.

Architecture Differences

The architectural divide between these two GPUs is stark. The NVIDIA P106-090 is built on the Pascal architecture, using the GP106 chip manufactured on TSMC's 16 nm process. It contains 4,400 million transistors on a 200 mm² die, yielding a transistor density of 22.0 million per mm². The AMD Radeon RX 5500M, in contrast, uses the RDNA 1.0 architecture with the Navi 14 chip on a 7 nm process, packing 6,400 million transistors into a smaller 158 mm² die for a density of 40.5 million per mm². This means the RX 5500M crams 45% more transistors into 21% less silicon area.

The compute resources differ dramatically. The P106-090 has 768 shading units, 48 texture mapping units (TMUs), and 48 raster operation pipelines (ROPs). The RX 5500M nearly doubles the shading units to 1,408 and increases TMUs to 88, but reduces ROPs to 32. This configuration favors the AMD part for shader-heavy workloads while giving the NVIDIA card an edge in pixel fill rate. The P106-090 achieves 73.49 GPixel/s pixel rate and 73.49 GTexel/s texture rate, whereas the RX 5500M produces 52.64 GPixel/s and 144.8 GTexel/s respectively.

Memory architecture also diverges. The P106-090 uses 3 GB GDDR5 on a 192-bit interface, while the RX 5500M uses 4 GB GDDR6 on a 128-bit bus. Despite the narrower bus, the RX 5500M's 14 Gbps effective memory speed delivers 224.0 GB/s, outpacing the P106-090's 192.2 GB/s. Both support DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4, but the RX 5500M adds a 2:1 FP16 ratio (9.265 TFLOPS) versus the P106-090's severely limited 1:64 FP16 (36.74 GFLOPS). The bus interface also differs: PCIe 1.0 x1 on the P106-090 versus PCIe 4.0 x8 on the RX 5500M.

Head-to-Head Benchmarks

The head-to-head benchmark data shows a clear sweep for the AMD Radeon RX 5500M, winning both recorded tests. In Geekbench OpenCL, the RX 5500M scores 38,725 against the P106-090's 21,304, a delta of -45% from the NVIDIA card's perspective. This means the AMD part is 81.8% faster in raw OpenCL compute, a massive margin that reflects the RDNA 1.0 architecture's higher FP32 throughput (4.632 TFLOPS vs 2.352 TFLOPS) and double the shading units.

The Vulkan gap is even more pronounced. The RX 5500M achieves 35,693 in Geekbench Vulkan, while the P106-090 manages only 18,596, representing a -47.9% delta. This 91.9% performance advantage in Vulkan suggests the AMD architecture handles modern graphics APIs more efficiently, likely due to its newer design and improved driver optimization for RDNA. The RX 5500M also has additional benchmark data points not available for the P106-090, including PassMark scores (DirectX 10: 39, DirectX 11: 35, DirectX 12: 28, DirectX 9: 100, G2D: 414, G3D: 5,848, GPU Compute: 2,316) and a Geekbench Metal score of 50,359, though these cannot be directly compared.

When placed against their respective nearest rivals, both GPUs sit at the same performance tier. The P106-090's closest competitors include the GTX 570 (deltaPct -0.3), Radeon Pro 555 (deltaPct 0.5), and Radeon HD 7770M (deltaPct -0.5). The RX 5500M's rivals include the FirePro M6100 (deltaPct 0), Radeon HD 8950M (deltaPct -0.1), and Radeon Pro 555X (deltaPct 0.3). These deltas are all within 0.5%, indicating both cards are extremely competitive within their immediate peer group despite the dramatic head-to-head difference.

Specification Differences

The specification sheets reveal stark contrasts across nearly every category. The process node differs significantly: 16 nm for the P106-090 versus 7 nm for the RX 5500M, both from TSMC. Transistor count is 4,400 million versus 6,400 million, and die size is 200 mm² versus 158 mm². The transistor density nearly doubles from 22.0M/mm² to 40.5M/mm².

Clock speeds favor the AMD part across the board. The RX 5500M has a base clock of 1375 MHz, a game clock of 1448 MHz, and a boost clock of 1645 MHz, while the P106-090 operates at 1354 MHz base and 1531 MHz boost. Memory clocks also differ: 2002 MHz (8 Gbps effective) for the NVIDIA card versus 1750 MHz (14 Gbps effective) for the AMD card. The memory bus width reverses the trend, with the P106-090 using 192-bit GDDR5 and the RX 5500M using 128-bit GDDR6.

The TDP is 75 W for the P106-090 and 85 W for the RX 5500M. Physical characteristics diverge: the P106-090 is a dual-slot card measuring 250 mm (9.8 inches) with a 1x 6-pin power connector and a 250 W suggested PSU, while the RX 5500M has no specified slot width, dimensions, or PSU requirement, and uses no power connectors. The P106-090 has no display outputs, whereas the RX 5500M's outputs are portable-device dependent. The bus interface is PCIe 1.0 x1 for the P106-090 versus PCIe 4.0 x8 for the RX 5500M. The RX 5500M also lists a predecessor (Polaris Mobile), while the P106-090 has none.

Where Each One Wins

The AMD Radeon RX 5500M wins decisively in compute-heavy and modern API workloads. Its 45% OpenCL advantage and 47.9% Vulkan advantage make it the clear choice for tasks that leverage general-purpose GPU compute, such as machine learning inference, video encoding, or physics simulations. The 4 GB GDDR6 memory also provides more capacity and higher bandwidth (224.0 GB/s vs 192.2 GB/s), which benefits large datasets and high-resolution textures. The 2:1 FP16 ratio (9.265 TFLOPS) is a significant advantage for applications that support half-precision arithmetic, whereas the P106-090's 1:64 FP16 (36.74 GFLOPS) is effectively useless for such workloads. The RX 5500M's PCIe 4.0 x8 interface also offers much higher host bandwidth than the P106-090's PCIe 1.0 x1, reducing data transfer bottlenecks.

The NVIDIA P106-090, despite losing both head-to-head benchmarks, retains specific advantages. Its pixel rate of 73.49 GPixel/s is 39.6% higher than the RX 5500M's 52.64 GPixel/s, which can benefit fill-rate-limited scenarios like MSAA or certain post-processing effects. It also has 48 ROPs versus 32, providing more rasterization throughput. The lower 75 W TDP makes it more power-efficient in absolute terms, though the RX 5500M delivers far more performance per watt given its 81.8% OpenCL advantage at only 13.3% higher power draw. The P106-090's 192-bit memory bus, while yielding lower bandwidth, may offer better latency characteristics in some workloads. Its 16 nm process and older Pascal architecture are less advanced, but the card's mining-focused design with no display outputs means it could be repurposed for headless compute tasks where the RX 5500M's mobile form factor is unavailable.

For gaming and general graphics, the RX 5500M is the stronger performer based on its Vulkan score, which is a better predictor of modern game performance than OpenCL. The P106-090's lack of display outputs disqualifies it from direct gaming use unless paired with a secondary GPU for output. The RX 5500M's portable-device-dependent outputs make it suitable for laptops, while the P106-090's dual-slot, 250 mm card form factor is desktop-bound. In summary, the RX 5500M wins for any application that can use its compute power or modern API support, while the P106-090 wins only in niche fill-rate or legacy scenarios. The 2-0 win count in head-to-head benchmarks tells the story: AMD's newer architecture decisively outperforms NVIDIA's mining-era Pascal part.