AMD Radeon PRO W6800 vs NVIDIA GeForce RTX 5090 D Comparison

Head-to-Head Benchmarks

The recorded database contains two direct comparison points between the AMD Radeon PRO W6800 and the NVIDIA GeForce RTX 5090 D, both in Geekbench compute workloads. The results are not close. In Geekbench OpenCL, the NVIDIA GeForce RTX 5090 D scores 310,674 against the AMD Radeon PRO W6800’s 121,808. That is a delta of -60.8% from the AMD card’s perspective, meaning the NVIDIA part is roughly 2.5 times faster in this test. The gap widens further in Geekbench Vulkan: the RTX 5090 D posts 376,915, while the W6800 manages 109,961, a -70.8% delta. The NVIDIA card more than triples the AMD card’s score in the Vulkan workload.

The overall benchmark averages tell a similar story, though the comparison is complicated by the fact that the two cards draw from different test suites. The AMD Radeon PRO W6800 has an average benchmark score of 135,396 across its three recorded tests (Geekbench Metal, OpenCL, and Vulkan). That places it in the 96th percentile of all GPUs in the database. Its nearest rivals include the NVIDIA A10M at 135,230 (delta +0.1%), the NVIDIA RTX 4000 Ada Generation at 135,218 (delta +0.1%), the AMD Radeon Pro W6800X Duo at 135,774 (delta -0.3%), and the AMD Radeon PRO V620 at 136,472 (delta -0.8%). So the W6800 sits in a tight cluster of professional workstation cards, trading fractions of a percent with its immediate peers.

The NVIDIA GeForce RTX 5090 D, meanwhile, has an average benchmark score of 77,712 across ten recorded tests, including 3DMark Steel Nomad DX12, several Passmark DirectX variants, Passmark G2D/G3D and GPU compute, plus Geekbench OpenCL and Vulkan. That score lands it in the 92nd percentile of all GPUs. Its nearest rivals are quite different: the AMD Radeon RX 6650M XT at 76,904 (delta +1.1%), the AMD Radeon RX 6850M XT at 78,940 (delta -1.6%), the NVIDIA Tesla P100 PCIe 12 GB at 79,396 (delta -2.1%), and the NVIDIA Tesla P100 PCIe 16 GB at 79,605 (delta -2.4%). The average score is heavily influenced by the Passmark tests, which are older DirectX workloads. The Passmark DirectX 9 score is 434, DirectX 10 is 231, DirectX 11 is 371, and DirectX 12 is 219. These low scores drag down the average, but the Geekbench results show the true compute headroom.

In direct head-to-head terms, the RTX 5090 D wins both recorded matchups. The W6800 has zero wins in the head-to-head table; the RTX 5090 D has two. The -60.8% and -70.8% deltas are decisive. No benchmark in the database shows the AMD card ahead of the NVIDIA card by any margin. The data indicates a generational and architectural chasm: the W6800 is from 2021, built on a 7 nm process with RDNA 2.0, while the RTX 5090 D is from 2025, built on a 5 nm process with Blackwell 2.0. The compute throughput difference is stark, and the Vulkan result in particular suggests the NVIDIA architecture extracts far more from modern graphics APIs.

The Verdict

The verdict from the recorded data is unambiguous: the NVIDIA GeForce RTX 5090 D is the faster card in every benchmark they share. The Geekbench OpenCL margin of 155% (from 121,808 to 310,674) and the Vulkan margin of 243% (from 109,961 to 376,915) are not marginal improvements; they are generational leaps. If the task is raw compute throughput in OpenCL or Vulkan, the RTX 5090 D is the only rational choice from these two options.

However, the database also shows that the AMD Radeon PRO W6800 holds its own in its own weight class. Its average score of 135,396 is higher than the RTX 5090 D’s average of 77,712, but that is misleading because the test suites do not overlap beyond the two Geekbench entries. The W6800’s average is built from three modern compute tests, all of which score well above 100,000. The RTX 5090 D’s average is dragged down by Passmark legacy tests that score in the hundreds. The percentile rankings reflect this: the W6800 sits at the 96th percentile, the RTX 5090 D at the 92nd. So in the database’s global ranking, the AMD card appears higher, but that is an artifact of which tests were run on each card.

For a buyer choosing between these two specific cards, the data points are the head-to-head results. In those, the RTX 5090 D wins both. The W6800 does not win any shared test. The RTX 5090 D also offers a higher boost clock (2407 MHz vs 2322 MHz), nearly six times the shading units (21,760 vs 3,840), and more than double the memory bandwidth (1.79 TB/s vs 512.0 GB/s). The RTX 5090 D has 170 RT cores and 680 tensor cores, while the W6800 has 60 RT cores and no tensor cores. The conclusion is straightforward: for compute performance in the workloads tested, the RTX 5090 D dominates.

Where Each One Wins

The AMD Radeon PRO W6800 does not win any benchmark in the head-to-head table. The recorded data shows zero wins for the AMD card against the RTX 5090 D. Its strengths lie elsewhere: it is a professional workstation card with 32 GB of GDDR6 memory on a 256-bit bus, delivering 512.0 GB/s of bandwidth. It has six mini-DisplayPort 1.4a outputs, which is a unique feature for multi-display professional setups. Its power draw is 250 W, with a suggested PSU of 600 W, which is far more modest than the RTX 5090 D’s 575 W and 950 W suggested PSU. The W6800 is a dual-slot card measuring 267 mm in length, 120 mm in height, and 50 mm in width, making it shorter and narrower than the RTX 5090 D (304 mm length, 137 mm height, 48 mm width). For a workstation that prioritizes multiple display outputs and lower power consumption, the W6800 has a structural advantage, though not a performance one.

The NVIDIA GeForce RTX 5090 D wins in every compute benchmark where the two are compared. It also wins on raw specifications: 32 GB of GDDR7 memory on a 512-bit bus, delivering 1.79 TB/s of bandwidth. Its pixel rate is 423.6 GPixel/s versus the W6800’s 222.9 GPixel/s. Its texture rate is 1,636.8 GTexel/s versus 557.3 GTexel/s. Its FP32 compute is 104.8 TFLOPS versus 17.83 TFLOPS. Its FP16 is 104.8 TFLOPS (1:1) versus 35.67 TFLOPS (2:1). The RTX 5090 D also has PCIe 5.0 x16 interface, while the W6800 is on PCIe 4.0 x16. The RTX 5090 D supports HDMI 2.1b and three DisplayPort 2.1b outputs, whereas the W6800 only has DisplayPort 1.4a.

For use cases, the data suggests the RTX 5090 D is for compute-heavy workloads: OpenCL and Vulkan rendering, high-resolution texture work, and any task that benefits from massive memory bandwidth and tensor cores. The W6800 is for legacy professional workflows that need many display outputs and lower power draw, but it will not compete in raw compute.

FAQ

Q: Which card has a higher Geekbench OpenCL score?

A: The NVIDIA GeForce RTX 5090 D scores 310,674 in Geekbench OpenCL, while the AMD Radeon PRO W6800 scores 121,808. The RTX 5090 D leads by a delta of -60.8% from the AMD card’s perspective.

Q: What is the difference in Geekbench Vulkan performance?

A: The NVIDIA GeForce RTX 5090 D scores 376,915 in Geekbench Vulkan, versus the AMD Radeon PRO W6800’s 109,961. That is a delta of -70.8%, meaning the NVIDIA card is more than three times faster in this workload.

Q: Do both cards have 32 GB of memory?

A: Yes, both cards have 32 GB of memory, but the types differ. The AMD Radeon PRO W6800 uses GDDR6 on a 256-bit bus with 512.0 GB/s bandwidth. The NVIDIA GeForce RTX 5090 D uses GDDR7 on a 512-bit bus with 1.79 TB/s bandwidth.

Q: What are the percentile rankings for each card?

A: The AMD Radeon PRO W6800 is in the 96th percentile of all GPUs in the database, with an average benchmark score of 135,396. The NVIDIA GeForce RTX 5090 D is in the 92nd percentile, with an average score of 77,712.

Q: Which card has more shading units?

A: The NVIDIA GeForce RTX 5090 D has 21,760 shading units. The AMD Radeon PRO W6800 has 3,840 shading units. The NVIDIA card also has 680 tensor cores, while the AMD card has none.

Q: What is the launch MSRP for each card?

A: The AMD Radeon PRO W6800 has a launch MSRP of 2,249 USD. The NVIDIA GeForce RTX 5090 D has a launch MSRP of 2,299 USD.

Architecture Differences

The two cards represent distinct architectural generations and design philosophies. The AMD Radeon PRO W6800 is built on the Navi 21 chip, using RDNA 2.0 architecture, fabricated on a 7 nm process at TSMC. It contains 26,800 million transistors on a 520 mm² die, giving a transistor density of 51.5 million per mm². The NVIDIA GeForce RTX 5090 D uses the GB202 chip, with Blackwell 2.0 architecture, fabricated on a 5 nm process at TSMC. It contains 92,200 million transistors on a 750 mm² die, yielding a transistor density of 122.9 million per mm². The NVIDIA chip has more than three times the transistor count and more than double the density.

The compute core configurations differ dramatically. The W6800 has 3,840 shading units, 240 texture mapping units, and 96 raster operation units. It also has 60 ray tracing cores and no tensor cores. The RTX 5090 D has 21,760 shading units, 680 TMUs, and 176 ROPs. It has 170 ray tracing cores and 680 tensor cores. The raw throughput numbers follow: the W6800 delivers 17.83 TFLOPS FP32 and 35.67 TFLOPS FP16 (2:1). The RTX 5090 D delivers 104.8 TFLOPS FP32 and 104.8 TFLOPS FP16 (1:1). The FP16 ratio alone indicates a fundamental design choice: the AMD card halves its FP16 rate, while the NVIDIA card runs FP16 at the same rate as FP32.

Memory architecture also diverges. The W6800 uses 32 GB of GDDR6 on a 256-bit bus, reaching 512.0 GB/s bandwidth and a memory clock of 2000 MHz (16 Gbps effective). The RTX 5090 D uses 32 GB of GDDR7 on a 512-bit bus, reaching 1.79 TB/s bandwidth and a memory clock of 1750 MHz (28 Gbps effective). The bus width doubling and the newer memory type explain the bandwidth gap.

Power and physical design differ as well. The W6800 has a TDP of 250 W, uses one 6-pin and one 8-pin power connector, and suggests a 600 W PSU. It is dual-slot, 267 mm long, 120 mm high, and 50 mm wide. The RTX 5090 D has a TDP of 575 W, uses a single 16-pin connector, and suggests a 950 W PSU. It is dual-slot, 304 mm long, 137 mm high, and 48 mm wide. The NVIDIA card is longer and taller but slightly thinner.

The display outputs and bus interfaces also separate the two. The W6800 offers six mini-DisplayPort 1.4a outputs and uses PCIe 4.0 x16. The RTX 5090 D offers one HDMI 2.1b and three DisplayPort 2.1b outputs, and uses PCIe 5.0 x16. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The production status differs: the W6800 is end-of-life, released on 2021-06-07, with a predecessor of Radeon Pro Vega. The RTX 5090 D is active, released on 2025-01-29, with a predecessor of GeForce 40 and a successor of GeForce 60.

The architecture differences explain the benchmark results. The RTX 5090 D’s newer process node, higher transistor density, more than five times the shading units, and memory bandwidth advantage all contribute to its decisive wins in OpenCL and Vulkan. The W6800’s older RDNA 2.0 design, while competent for its era, cannot match the Blackwell 2.0 compute throughput in the recorded tests.