AMD Radeon AI PRO 9600D vs NVIDIA GeForce RTX 5090 D Comparison
AMD Radeon AI PRO 9600D
GeForce RTX 5090 D
PERFORMANCE BENCHMARKS
Analysis: AMD Radeon AI PRO 9600D vs NVIDIA GeForce RTX 5090 D
Head-to-Head Benchmarks
The recorded data does not contain any shared benchmark tests between the AMD Radeon AI PRO 9600D and the NVIDIA GeForce RTX 5090 D. The head-to-head benchmark table is empty, and the AMD card has no benchmark scores or average score listed in the database. The NVIDIA card, by contrast, has a full set of recorded measurements.
The RTX 5090 D delivers a 3DMark Steel Nomad DX12 score of 14,326 points. In Geekbench compute tests, it records 310,674 in OpenCL and 376,915 in Vulkan. Passmark results show a G3D score of 44,065, a GPU compute score of 28,396, and DirectX 9, 10, 11, and 12 scores of 434, 231, 371, and 219 respectively. The G2D score sits at 1,487. These figures place the RTX 5090 D at the 92nd percentile of all GPUs in the database, with an average benchmark score of 77,712.
Without comparable measurements for the Radeon AI PRO 9600D, direct numerical comparison is impossible from the provided data. The AMD card holds the 50th percentile position among all GPUs, but this ranking is not supported by any listed benchmark scores in the record.
The closest recorded rivals to the RTX 5090 D provide context for its standing. The AMD Radeon RX 6650M XT scores 76,904, which is 1.1% lower than the RTX 5090 D's average. The AMD Radeon RX 6850M XT scores 78,940, which is 1.6% higher. The NVIDIA Tesla P100 PCIe 12 GB scores 79,396, 2.1% higher, and the Tesla P100 PCIe 16 GB scores 79,605, 2.4% higher. These deltas indicate the RTX 5090 D sits in a tight cluster with those four cards, all within roughly 2.5% of each other.
Architecture Differences
The two GPUs come from different design families and manufacturing processes. The AMD Radeon AI PRO 9600D uses the Navi 48 chip based on RDNA 4.0 architecture, built on a 4 nm process at TSMC. It belongs to the Radeon Pro Navi generation. The NVIDIA GeForce RTX 5090 D uses the GB202 chip based on Blackwell 2.0 architecture, also at TSMC but on a 5 nm process, and belongs to the GeForce 50 series.
Transistor counts differ substantially. The AMD chip contains 53,900 million transistors on a 357 mm² die, giving a transistor density of 151.0 million per mm². The NVIDIA chip packs 92,200 million transistors on a 750 mm² die, with a density of 122.9 million per mm². The AMD part achieves higher density on a much smaller die, while NVIDIA uses a physically larger chip with roughly 71% more transistors.
The compute resources are in different leagues. The Radeon AI PRO 9600D has 3,072 shading units, 192 texture mapping units, 96 render output units, and 48 ray tracing cores. It has no tensor core count listed. The RTX 5090 D has 21,760 shading units, 680 TMUs, 176 ROPs, 170 ray tracing cores, and 680 tensor cores. The NVIDIA card carries roughly seven times the shading units and more than triple the ray tracing cores.
Clock speeds also differ. The AMD card runs a base clock of 1080 MHz, a boost clock of 2020 MHz, and a game clock of 1080 MHz. The NVIDIA card runs a base clock of 2017 MHz and a boost clock of 2407 MHz. The NVIDIA chip boosts about 19% higher and starts from a much higher base.
Memory configurations show a shared capacity but different implementations. Both cards have 32 GB of video memory. The AMD card uses GDDR6 on a 256-bit bus with 576.0 GB/s of bandwidth, with memory clocked at 2250 MHz or 18 Gbps effective. The NVIDIA card uses GDDR7 on a 512-bit bus with 1.79 TB/s of bandwidth, more than triple the AMD figure, with memory at 1750 MHz or 28 Gbps effective.
The API support is identical: both list DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Both use PCIe 5.0 x16 interfaces. The power and physical profiles diverge sharply. The AMD card draws 150 W TDP with a suggested 450 W power supply, while the NVIDIA card draws 575 W TDP with a suggested 950 W power supply. The AMD card is single-slot with a 16-pin connector, while the NVIDIA card is dual-slot with a 16-pin connector.
FAQ
Q: Which GPU has more memory bandwidth?
A: The NVIDIA GeForce RTX 5090 D has 1.79 TB/s of bandwidth over a 512-bit bus using GDDR7. The AMD Radeon AI PRO 9600D has 576.0 GB/s over a 256-bit bus using GDDR6.
Q: How do the transistor counts compare?
A: The RTX 5090 D has 92,200 million transistors on a 750 mm² die. The Radeon AI PRO 9600D has 53,900 million transistors on a 357 mm² die.
Q: What is the power draw difference?
A: The Radeon AI PRO 9600D has a 150 W TDP and a suggested 450 W power supply. The RTX 5090 D has a 575 W TDP and a suggested 950 W power supply.
Q: Which card has more ray tracing hardware?
A: The RTX 5090 D has 170 ray tracing cores. The Radeon AI PRO 9600D has 48 ray tracing cores.
Q: Are both cards still in production?
A: Yes, both are listed with an Active production status. The AMD card was released on December 10, 2025, and the NVIDIA card was released on January 29, 2025.
Q: What is the launch MSRP of the RTX 5090 D?
A: The launch MSRP is 2,299 USD. No launch MSRP is recorded for the Radeon AI PRO 9600D.
The Verdict
The recorded data makes the positioning clear. The RTX 5090 D is the only one of the two with benchmark measurements, and those measurements place it at the 92nd percentile of all GPUs. Its average benchmark score of 77,712 puts it within 1.1% to 2.4% of its nearest listed rivals, all of which are high-end cards. The Radeon AI PRO 9600D has no recorded benchmark scores and sits at the 50th percentile, a ranking that cannot be validated against any listed test results.
For raw compute throughput, the NVIDIA card dominates on paper. Its FP32 performance of 104.8 TFLOPS is more than four times the AMD card's 24.82 TFLOPS. Its FP16 output matches its FP32 at 104.8 TFLOPS, while the AMD card also runs FP16 at 1:1 ratio with 24.82 TFLOPS. Pixel fill rate favors NVIDIA at 423.6 GPixel/s versus 193.9 GPixel/s, and texture fill rate favors NVIDIA at 1,636.8 GTexel/s versus 387.8 GTexel/s.
The AMD card does offer a lower power envelope at 150 W versus 575 W, a single-slot form factor instead of dual-slot, and a smaller physical footprint at 241 mm length versus 304 mm. Those traits suit constrained chassis designs. But in terms of measured performance, the database only contains results for the NVIDIA card, and those results are strong.
Specification Differences
The two cards differ across nearly every major specification field.
The AMD Radeon AI PRO 9600D uses the Navi 48 chip on RDNA 4.0 architecture, built on a 4 nm process. The NVIDIA GeForce RTX 5090 D uses the GB202 chip on Blackwell 2.0 architecture, built on a 5 nm process. Both are TSMC parts.
Transistors: 53,900 million for AMD, 92,200 million for NVIDIA. Die size: 357 mm² for AMD, 750 mm² for NVIDIA. Transistor density: 151.0M per mm² for AMD, 122.9M per mm² for NVIDIA.
Clocks: AMD base 1080 MHz, boost 2020 MHz, game 1080 MHz. NVIDIA base 2017 MHz, boost 2407 MHz, no game clock listed.
Memory: AMD uses 32 GB GDDR6 on a 256-bit bus at 576.0 GB/s. NVIDIA uses 32 GB GDDR7 on a 512-bit bus at 1.79 TB/s.
Compute units: AMD has 3,072 shading units, 192 TMUs, 96 ROPs, and 48 RT cores. NVIDIA has 21,760 shading units, 680 TMUs, 176 ROPs, 170 RT cores, and 680 tensor cores.
Rates: AMD pixel rate is 193.9 GPixel/s, texture rate is 387.8 GTexel/s, FP32 and FP16 are both 24.82 TFLOPS. NVIDIA pixel rate is 423.6 GPixel/s, texture rate is 1,636.8 GTexel/s, FP32 and FP16 are both 104.8 TFLOPS.
Power: AMD TDP is 150 W with a 450 W suggested PSU. NVIDIA TDP is 575 W with a 950 W suggested PSU.
Physical: AMD is single-slot, 241 mm long, 111 mm high, 19 mm wide. NVIDIA is dual-slot, 304 mm long, 137 mm high, 48 mm wide. Both use one 16-pin power connector.
Display outputs: AMD has 1x DisplayPort 2.1a. NVIDIA has 1x HDMI 2.1b and 3x DisplayPort 2.1b.
Release dates: AMD released December 10, 2025. NVIDIA released January 29, 2025. NVIDIA lists a predecessor (GeForce 40) and successor (GeForce 60); AMD lists a predecessor (Radeon Pro Vega) and no successor.
Where Each One Wins
Based strictly on the recorded data, the NVIDIA GeForce RTX 5090 D wins in every category that has a measured or specified performance figure. It has higher FP32 and FP16 throughput at 104.8 TFLOPS versus 24.82 TFLOPS. It has higher pixel rate at 423.6 GPixel/s versus 193.9 GPixel/s. It has higher texture rate at 1,636.8 GTexel/s versus 387.8 GTexel/s. It has more memory bandwidth at 1.79 TB/s versus 576.0 GB/s. It has more shading units, TMUs, ROPs, ray tracing cores, and tensor cores. It has the only benchmark scores in the database, with an average of 77,712 and a 92nd percentile ranking.
The AMD Radeon AI PRO 9600D wins in efficiency and physical integration. Its 150 W TDP is far lower than the NVIDIA card's 575 W. Its suggested power supply of 450 W is less than half of the 950 W suggested for the NVIDIA card. It occupies a single slot rather than two, and it is shorter, lower, and thinner at 241 mm by 111 mm by 19 mm compared to 304 mm by 137 mm by 48 mm. It also has a higher transistor density at 151.0M per mm² versus 122.9M per mm², though that does not translate into higher measured performance.
For workloads involving computational throughput, ray tracing, or memory bandwidth, the database points squarely at the RTX 5090 D. For installations where power draw, chassis space, and cooling clearance are the priority, the Radeon AI PRO 9600D presents the more accommodating specifications. The AMD card's 50th percentile standing, without any recorded benchmark results to support it, leaves its real-world positioning unclear. The NVIDIA card's 92nd percentile standing, backed by a full suite of benchmark scores, gives it a clear measured advantage in the database.