AMD Radeon RX 7800 XT vs NVIDIA P106-090 Comparison

Head-to-Head Benchmarks

The benchmark data presents a stark contrast between these two GPUs. The AMD Radeon RX 7800 XT dominates in modern 3D rendering workloads, while the NVIDIA P106-090 shows a surprising edge in one specific compute test. The recorded results are unambiguous: the AMD card wins 2 out of 3 head-to-head comparisons, and its victories are decisive.

In the 3DMark Steel Nomad DX12 test, the RX 7800 XT scores 4,157 points against the P106-090's 509 points. That is an 87.8% deficit for the NVIDIA card, meaning the AMD GPU delivers roughly eight times the performance in this DirectX 12 workload. This is not a close contest; it is a generational gap made visible through raw numbers. The P106-090, built for mining with no display outputs, was never designed for this kind of modern gaming benchmark, and the data confirms its limitations.

The Vulkan results tell a similar story. The RX 7800 XT scores 54,228 points in Geekbench Vulkan, while the P106-090 manages 18,596 points. The delta stands at 65.7% in favor of AMD. Vulkan is a low-level API that benefits from architectural efficiency and raw hardware resources, and the RDNA 3.0 architecture with 3,840 shading units simply overwhelms the Pascal-based card with 768 shading units. The gap is substantial enough that any Vulkan-based workload would show a clear, measurable advantage for the RX 7800 XT.

However, the P106-090 claims one unexpected victory. In Geekbench OpenCL, the NVIDIA card scores 21,304 points versus 18,290 points for the RX 7800 XT. That is a 16.5% lead for the older card. This result is notable because OpenCL compute performance does not always track with gaming performance. The P106-090's Pascal architecture, despite its age, apparently handles this particular OpenCL workload more efficiently. It is a narrow but real win for NVIDIA in this specific test.

Looking at the broader average benchmark scores, the picture becomes more complex. The P106-090 has an average benchmark score of 13,470, while the RX 7800 XT averages 11,627. This counterintuitive result stems from the fact that the RX 7800 XT's benchmark suite includes several PassMark tests with very low scores, which drag down its average. The NVIDIA card, with only three benchmarks in the database, benefits from a smaller and more favorable sample set. The P106-090's percentile ranking is 54 against the RX 7800 XT's 51, so by the database's aggregate measure, the mining GPU sits slightly higher relative to all other GPUs. Yet the head-to-head 3D results tell the more meaningful story for actual graphics workloads.

The Verdict

The data supports a clear conclusion: the AMD Radeon RX 7800 XT is the superior graphics processor for any modern rendering task. Its 87.8% advantage in 3DMark Steel Nomad DX12 and 65.7% lead in Geekbench Vulkan are decisive margins that no amount of architectural nostalgia can bridge. The RX 7800 XT is an active product, built on a 5 nm process with 16 GB of GDDR6 memory, and it delivers the kind of performance that modern games and applications expect.

The NVIDIA P106-090 is a different creature entirely. Its single head-to-head win in Geekbench OpenCL, with a 16.5% margin, shows that it retains some compute capability. But the card has no display outputs, uses a PCIe 1.0 x1 interface, and was designed for mining, not for user-facing graphics work. Its 3 GB GDDR5 memory and 192-bit bus are dwarfed by the RX 7800 XT's 16 GB GDDR6 on a 256-bit bus. For anyone building a system for gaming, content creation, or any visual workload, the RX 7800 XT is the only rational choice based on this data.

That said, the P106-090's OpenCL result and its higher average benchmark score suggest it is not without utility in specific compute scenarios. But those scenarios are narrow, and the card's end-of-life production status and lack of display outputs limit its practical applications. The RX 7800 XT, with its active production status and modern feature set, is the GPU that the benchmark evidence supports for virtually every use case.

FAQ

Q: Which GPU wins in DirectX 12 gaming benchmarks?

A: The AMD Radeon RX 7800 XT wins decisively. In the 3DMark Steel Nomad DX12 test, it scores 4,157 points compared to the NVIDIA P106-090's 509 points, an 87.8% advantage for AMD.

Q: Does the NVIDIA P106-090 win any benchmarks?

A: Yes, the P106-090 wins the Geekbench OpenCL test with a score of 21,304 against the RX 7800 XT's 18,290, a 16.5% lead. This is its only head-to-head victory in the recorded data.

Q: How do the average benchmark scores compare?

A: The NVIDIA P106-090 has a higher average benchmark score of 13,470 versus 11,627 for the RX 7800 XT. However, this aggregate figure includes several low-scoring PassMark tests for the AMD card, and the head-to-head 3D results strongly favor the RX 7800 XT.

Q: What is the memory configuration difference?

A: The RX 7800 XT has 16 GB of GDDR6 memory on a 256-bit bus with 624.1 GB/s bandwidth. The P106-090 has 3 GB of GDDR5 on a 192-bit bus with 192.2 GB/s bandwidth.

Q: Which GPU has a higher percentile ranking?

A: The NVIDIA P106-090 ranks in the 54th percentile among all GPUs, while the RX 7800 XT ranks in the 51st percentile. This is based on the database's aggregate scoring methodology.

Q: What are the power requirements for each card?

A: The RX 7800 XT has a TDP of 263 W and requires a 600 W suggested power supply with 2x 8-pin connectors. The P106-090 has a TDP of 75 W and requires a 250 W suggested power supply with a single 6-pin connector.

Specification Differences

The two GPUs differ across nearly every measurable specification. The process node is a major divider: the P106-090 uses a 16 nm process from TSMC, while the RX 7800 XT uses a 5 nm process from the same foundry. Transistor counts reflect this gap, with the AMD card packing 28,100 million transistors against the NVIDIA card's 4,400 million. Die size also differs, with the RX 7800 XT at 346 mm² versus 200 mm² for the P106-090.

Clock speeds show the architectural evolution. The P106-090 has a base clock of 1354 MHz and a boost clock of 1531 MHz. The RX 7800 XT has a lower base clock of 1295 MHz but a much higher boost clock of 2430 MHz, with a game clock of 2124 MHz. Memory speeds follow suit: the NVIDIA card runs at 2002 MHz (8 Gbps effective), while the AMD card runs at 2438 MHz (19.5 Gbps effective).

The compute resources are heavily skewed toward AMD. The RX 7800 XT features 3,840 shading units, 240 texture mapping units, and 96 render output units. The P106-090 has 768 shading units, 48 TMUs, and 48 ROPs. The RX 7800 XT also includes 60 ray tracing cores, which the P106-090 lacks entirely. Raw throughput numbers confirm the gap: the RX 7800 XT delivers 37.32 TFLOPS of FP32 performance, while the P106-090 manages 2.352 TFLOPS. Texture and pixel rates follow the same pattern, with the AMD card at 583.2 GTexel/s and 233.3 GPixel/s versus 73.49 GTexel/s and 73.49 GPixel/s for NVIDIA.

The bus interface differs radically. The P106-090 uses PCIe 1.0 x1, a severely limited connection, while the RX 7800 XT uses PCIe 4.0 x16. Display outputs are absent on the P106-090, while the RX 7800 XT offers 1x HDMI 2.1a and 3x DisplayPort 2.1. Physical dimensions also vary: the P106-090 is 250 mm long, while the RX 7800 XT is 267 mm long, 111 mm high, and 50 mm wide. Power consumption is a notable difference, with the RX 7800 XT rated at 263 W TDP versus 75 W for the P106-090.

Architecture Differences

The architectural divide between these GPUs is fundamental. The NVIDIA P106-090 is built on the Pascal architecture, specifically the GP106, and belongs to the Mining GPUs generation. It uses the 16 nm process node at TSMC with 4,400 million transistors on a 200 mm² die. Pascal was designed in an era when 16 nm was leading-edge technology, and the architecture lacks dedicated ray tracing hardware. The card supports DirectX 12 (12_1) and OpenGL 4.6 and Vulkan 1.4, with FP16 performance rated at 36.74 GFLOPS at a 1:64 ratio, indicating a heavily reduced rate compared to FP32. Its transistor density is 22.0 million transistors per square millimeter.

The AMD Radeon RX 7800 XT is built on the RDNA 3.0 architecture, specifically Navi 32, and belongs to the Navi III (RX 7000) generation. It uses the 5 nm process with 28,100 million transistors on a 60 mm² die, achieving a density of 81.2 million transistors per square millimeter. RDNA 3.0 is a complete departure from previous AMD architectures, with 60 ray tracing cores integrated directly into the design. The architecture supports DirectX 12 Ultimate (12_2) and FP16 performance at 37.32 TFLOPS at a 1:1 ratio, meaning FP16 and FP32 are processed at equal rates. The RX 7800 XT is a 5 nm part through and through, with 96 ROPs and 240 TMUs, making it a fully modern GPU with the feature set expected of a current-generation product.

The cache hierarchy and feature set differ substantially. The P106-090 has 768 shading units arranged in a configuration that was standard for its era, with 48 texture units and 48 ROPs. The RX 7800 XT has 3,840 shading units, 240 TMUs, and 96 ROPs, a layout optimized for the high throughput demands of RDNA 3.0. The memory architecture also differs in size and speed, with the RX 7800 XT using 16 GB of GDDR6 on a 256-bit bus and the P106-090 using GDDR5 on a 192-bit bus. Pascal was designed for a generation when GDDR5 was standard and GDDR6 was not yet widespread. RDNA 3.0 was designed for the 5 nm era, where GDDR6 has become the standard for high-bandwidth memory solutions.

Where Each One Wins

The AMD Radeon RX 7800 XT wins in every category that matters for modern graphics. Its 3DMark Steel Nomad DX12 score of 4,157 is 87.8% ahead of the P106-090, making it the clear choice for DirectX 12 gaming and any workload that uses modern rendering APIs. Its Geekbench Vulkan score of 54,228, a 65.7% lead, reinforces this position for Vulkan-based titles and applications. The RX 7800 XT also has 60 ray tracing cores, enabling hardware-accelerated ray tracing, 16 GB of GDDR6 memory for high-resolution textures and large datasets, and display outputs for connecting displays. With 37.32 TFLOPS of FP32 throughput, 583.2 GTexel/s texture rate, 233.3 GPixel/s pixel rate, the RX 7800 XT has the raw resources to handle demanding visual workloads. Its 5 nm process, 28,100 million transistors, and 346 mm² die size represent the physical foundation for this performance.

The NVIDIA P106-090 wins in a narrower set of circumstances. Its Geekbench OpenCL score of 21,304 is 16.5% ahead of the RX 7800 XT, indicating that it handles this particular compute workload more efficiently. The card's average benchmark score of 13,470 and 54th percentile ranking are also higher than the RX 7800 XT's 11,627 and 51st percentile, though these aggregate measures are influenced by the different benchmark sets each card runs. The P106-090's 75 W TDP is dramatically lower than the RX 7800 XT's 263 W, making it far more power-efficient in absolute terms. It is also a shorter card at 250 mm versus 267 mm, and its 3 GB GDDR5 memory, while small, may be sufficient for certain compute tasks. The 192-bit bus, 192.2 GB/s bandwidth, and 768 shading units give it a baseline level of capability that its OpenCL result demonstrates is still functional. For tasks that rely specifically on OpenCL compute and do not require display output or modern graphics features, the P106-090 shows a measurable advantage. But for everything else, the RX 7800 XT's benchmark results are decisive.