AMD Radeon Pro W6800X vs NVIDIA Quadro P6000 Comparison

The Verdict

The AMD Radeon Pro W6800X is the clear performance pick. Its average benchmark score of 160,671 puts it far ahead of the NVIDIA Quadro P6000, which averages 69,986. That is a massive gap, and the W6800X sits at the 97th percentile among all GPUs, while the P6000 sits at the 90th. The W6800X also has more than double the memory capacity with 32 GB versus 24 GB, and it is built on a much newer architecture with modern features like ray tracing cores. Anyone choosing between these two for compute-heavy workloads should take the W6800X without hesitation, provided their system can accommodate its quad-slot cooler and Apple MPX power interface.

The NVIDIA Quadro P6000 does have one meaningful advantage: compatibility. It uses a standard PCIe 3.0 x16 interface, a dual-slot cooler, and a single 8-pin power connector, making it dramatically easier to install in a conventional PC workstation. The W6800X uses an Apple MPX bus interface and an Apple MPX power connector, which means it is designed for Apple Mac Pro systems, not standard PC motherboards. If the target machine is a standard PC, the P6000 is the only one of the two that will drop in without custom hardware. The P6000 also has a lower transistor density and older process node, but those are irrelevant to a buyer. The decision comes down to this: if you can physically mount and power the W6800X, buy it. If you need a drop-in card for a classic PC workstation, the P6000 is the practical choice.

Architecture Differences

The AMD Radeon Pro W6800X is built on the Navi 21 chip using the RDNA 2.0 architecture, fabricated on a 7 nm process at TSMC. It packs 26,800 million transistors into a 520 mm² die, giving it a transistor density of 51.5 million per square millimeter. The NVIDIA Quadro P6000 uses the GP102 chip with the Pascal architecture, also from TSMC, but on a much older 16 nm process. It contains 11,800 million transistors on a 471 mm² die, for a density of 25.1 million per square millimeter. The W6800X has more than twice the transistor count in a slightly larger die, evidence of the density advantage of the newer node.

The W6800X features 60 ray tracing cores, while the P6000 has none. That alone makes the W6800X the only choice for any workload that leverages DXR or similar ray-traced rendering paths. The W6800X also supports DirectX 12 Ultimate with feature level 12_2, while the P6000 is limited to DirectX 12 (12_1). Both support OpenGL 4.6 and Vulkan 1.4, so API-level compatibility for those two is equal.

Memory technology differs significantly. The W6800X uses 32 GB of GDDR6 on a 256 bit bus, delivering 512.0 GB/s of bandwidth. The P6000 uses 24 GB of GDDR5X on a 384 bit bus, delivering 432.8 GB/s. The W6800X has more capacity and more bandwidth despite the narrower bus, thanks to the faster memory type. The W6800X memory runs at 2000 MHz with 16 Gbps effective, while the P6000 memory runs at 1127 MHz with 9 Gbps effective.

Shading unit and texture mapping unit counts are identical: both have 3840 shading units and 240 TMUs, and both have 96 ROPs. However, clock speeds differ substantially. The W6800X has a base clock of 1800 MHz and a boost clock of 2087 MHz. The P6000 is far slower in raw clock terms, with a base of 1506 MHz and a boost of 1645 MHz. That clock gap explains much of the performance difference in shader-bound workloads.

Power and physical design are polar opposites. The W6800X has a TDP of 200 W but requires a quad-slot cooler and an Apple MPX power connector, with a suggested PSU of 550 W. The P6000 has a higher TDP of 250 W but fits in a dual-slot cooler and uses a single 8-pin power connector, with a suggested PSU of 600 W. The P6000 is 267 mm long and 111 mm high. The W6800X is also 267 mm long but 120 mm high, 9 mm taller. Display outputs differ as well: the W6800X provides 1x HDMI 2.1 and 4x Thunderbolt, while the P6000 provides 1x DVI and 4x DisplayPort 1.4a.

FAQ

Q: Which card is faster in average benchmark score?

A: The AMD Radeon Pro W6800X averages 160,671 across all recorded benchmarks, compared to 69,986 for the NVIDIA Quadro P6000. That is roughly 2.3 times the average score.

Q: Does the NVIDIA Quadro P6000 support ray tracing?

A: No. The P6000 has no ray tracing cores. The AMD Radeon Pro W6800X has 60 ray tracing cores, so it is the only one of the two capable of hardware-accelerated ray tracing.

Q: Can the AMD Radeon Pro W6800X be installed in a standard PC?

A: The W6800X uses an Apple MPX bus interface and an Apple MPX power connector, so it is not designed for standard PC motherboards. The P6000 uses PCIe 3.0 x16 and a single 8-pin power connector, which is standard in conventional workstations.

Q: Which card has more memory bandwidth?

A: The W6800X has 512.0 GB/s of bandwidth from its 32 GB GDDR6 memory on a 256 bit bus. The P6000 has 432.8 GB/s from 24 GB GDDR5X on a 384 bit bus.

Q: How do the two compare in OpenCL performance?

A: In the head-to-head Geekbench OpenCL test, the W6800X scored 124,498 versus 66,382 for the P6000, a delta of 87.5% in favor of the AMD card.

Q: Are both cards still in production?

A: No. Both are listed as end-of-life products. The W6800X was released on 2021-08-02, and the P6000 was released on 2016-09-30.

Specification Differences

The two cards differ in nearly every major specification category. The process node is 7 nm for the W6800X versus 16 nm for the P6000. Transistor count is 26,800 million versus 11,800 million. Die size is 520 mm² versus 471 mm². Transistor density is 51.5 million per mm² versus 25.1 million per mm².

Clock speeds: the W6800X runs at 1800 MHz base and 2087 MHz boost. The P6000 runs at 1506 MHz base and 1645 MHz boost. Memory clocks: 2000 MHz with 16 Gbps effective for the W6800X, versus 1127 MHz with 9 Gbps effective for the P6000.

Memory configuration: 32 GB GDDR6 on a 256 bit bus with 512.0 GB/s bandwidth for the W6800X, versus 24 GB GDDR5X on a 384 bit bus with 432.8 GB/s bandwidth for the P6000.

Compute resources: both have 3840 shading units, 240 TMUs, and 96 ROPs, but the W6800X adds 60 ray tracing cores, while the P6000 has none. The W6800X delivers 16.03 TFLOPS FP32 and 32.06 TFLOPS FP16 with a 2:1 ratio. The P6000 delivers 12.63 TFLOPS FP32 and only 197.4 GFLOPS FP16 with a 1:64 ratio, making the W6800X dramatically superior in half-precision work.

Pixel and texture rates: the W6800X achieves 200.4 GPixel/s and 500.9 GTexel/s. The P6000 achieves 157.9 GPixel/s and 394.8 GTexel/s.

Power and physical: the W6800X has a 200 W TDP, quad-slot cooler, Apple MPX power connector, and 550 W suggested PSU. The P6000 has a 250 W TDP, dual-slot cooler, single 8-pin power connector, and 600 W suggested PSU.

Interface and outputs: the W6800X uses Apple MPX and offers 1x HDMI 2.1 plus 4x Thunderbolt. The P6000 uses PCIe 3.0 x16 and offers 1x DVI plus 4x DisplayPort 1.4a.

Dimensions: both are 267 mm long, but the W6800X is 120 mm high versus 111 mm for the P6000.

API support: the W6800X supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The P6000 supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4.

Head-to-Head Benchmarks

The database includes only one shared benchmark between the two cards: Geekbench OpenCL. The AMD Radeon Pro W6800X scored 124,498, while the NVIDIA Quadro P6000 scored 66,382. That gives the W6800X a win with a delta of 87.5%. This is not a close contest; the AMD card is nearly twice as fast in raw OpenCL compute.

The W6800X also has its own Geekbench Metal score of 196,844, a test the P6000 cannot run since it lacks Metal support in the recorded data. The P6000 has its own Geekbench Vulkan score of 73,590, a test not recorded for the W6800X. Direct comparison across these different APIs is not possible from the data, but the OpenCL result is the cleanest apples-to-apples metric available.

Looking at average benchmark scores, the W6800X sits at 160,671 overall, which places it in the 97th percentile of all GPUs. Its nearest rivals are the NVIDIA A100 PCIe 40 GB at 162,504, which is 1.1% ahead, the AMD Radeon PRO W7800 at 164,894, which is 2.6% ahead, the NVIDIA RTX A5500 at 165,217, which is 2.8% ahead, and the NVIDIA RTX 4500 Ada Generation at 166,094, which is 3.3% ahead. So the W6800X is competitive with modern flagship workstation cards, sitting only slightly below them.

The P6000, by contrast, averages 69,986, placing it in the 90th percentile. Its nearest rivals are the AMD Radeon Pro WX 8200 at 69,870, which is 0.2% behind, the NVIDIA RTX A3000 Mobile at 70,140, which is 0.2% ahead, the AMD Radeon RX 6600 LE at 70,829, which is 1.2% ahead, and the NVIDIA CMP 90HX at 69,000, which is 1.4% behind. The P6000 is effectively at parity with mid-range cards from its era, not in the same league as the W6800X.

Where Each One Wins

The AMD Radeon Pro W6800X wins in every compute metric recorded. It dominates OpenCL with an 87.5% advantage. It has more memory, 32 GB versus 24 GB, and higher bandwidth, 512.0 GB/s versus 432.8 GB/s. It has 60 ray tracing cores, which the P6000 completely lacks. It delivers more than double the FP16 throughput, 32.06 TFLOPS versus 197.4 GFLOPS, a 162x difference in half-precision capability. It also has a higher pixel rate, 200.4 GPixel/s versus 157.9 GPixel/s, and a higher texture rate, 500.9 GTexel/s versus 394.8 GTexel/s. For any workload that is shader-bound, memory-bound, or ray-tracing-bound, the W6800X is the stronger card.

The NVIDIA Quadro P6000 wins in the practical domain of installability. It uses PCIe 3.0 x16, a universal interface, and a dual-slot cooler with a single 8-pin power connector. The W6800X requires an Apple MPX bus and power connector, limiting it to Apple Mac Pro systems. The P6000 also has a lower TDP of 250 W versus 200 W, which may seem counterintuitive, but the higher power draw is offset by the far simpler physical installation. The P6000 also supports DVI output, which may matter for legacy display setups, and it uses DisplayPort 1.4a outputs, while the W6800X uses HDMI 2.1 and Thunderbolt.

For a Mac Pro user, the W6800X is the obvious choice. For a PC workstation user, the P6000 is the only one of the two that will work without custom adapters. The performance gap is enormous, but it only matters if the card can physically be installed in the target machine. The database shows a clear winner in raw performance, and an equally clear winner in compatibility. The right pick depends entirely on the host system.