AMD FirePro S9300 X2 vs NVIDIA GeForce RTX 5080 Mobile Comparison

The database pits two GPUs from entirely different eras and markets against each other: NVIDIA's GeForce RTX 5080 Mobile, a Blackwell 2.0 laptop part built on TSMC's 5 nm process, and AMD's FirePro S9300 X2, a dual-GPU GCN 3.0 server card from 2016. The recorded data shows a landslide for the modern mobile part, which wins both shared head-to-head benchmarks while drawing a fraction of the power.

Head-to-Head Benchmarks

Only two tests overlap in the database for both cards, and the RTX 5080 Mobile sweeps them.

In Geekbench OpenCL, the RTX 5080 Mobile scores 166986 against 27971 for the FirePro S9300 X2, a delta of 497 percent. That is not a close contest; the Blackwell card delivers roughly six times the compute throughput as measured through the OpenCL API. Geekbench Vulkan shows the same pattern with a slightly narrower gap: 169754 versus 37109, a 357.4 percent advantage. Even in the test where the S9300 X2 performs relatively best, it trails by more than three and a half times.

Context from the rest of the database reinforces the gap. The RTX 5080 Mobile posts 12134 in PassMark GPU Compute and 27711 in PassMark G3D, along with 4952 in 3DMark Steel Nomad, tests the FirePro has no recorded entries for. Its percentile against all GPUs in the database is 81, versus 77 for the FirePro. Average benchmark scores are 38349 for the NVIDIA card and 32540 for the AMD card.

The rivalry picture differs sharply too. The RTX 5080 Mobile sits within one percent of the GeForce MX570 (0.1 percent), RTX 4080 Mobile (0.6 percent), MX570 A (-0.9 percent), and Radeon Pro 580X (-0.9 percent). The FirePro S9300 X2 clusters near the Radeon RX 590 GME (-0.2 percent), RX 7900 GRE (0.3 percent), FirePro S10000 (0.5 percent), and NVIDIA T600 Mobile (-0.9 percent).

FAQ

Q: Which GPU wins the head-to-head benchmarks?

A: The RTX 5080 Mobile wins both shared tests. It leads Geekbench OpenCL by 497 percent (166986 vs 27971) and Geekbench Vulkan by 357.4 percent (169754 vs 37109).

Q: How do their power profiles compare?

A: The RTX 5080 Mobile has a TDP of 80 W, while the FirePro S9300 X2 is rated at 300 W and requires a suggested 700 W PSU with 2x 8-pin connectors. The mobile part needs no power connectors at all.

Q: How much memory does each card have, and what type?

A: The RTX 5080 Mobile carries 16 GB of GDDR7 on a 256-bit bus with 896.0 GB/s of bandwidth. The FirePro S9300 X2 has 4 GB of HBM on a very wide 4096-bit bus, delivering 512.0 GB/s.

Q: Are both cards still in production?

A: No. The RTX 5080 Mobile is listed as Active, released 2025-04-01. The FirePro S9300 X2 is End-of-life, released 2016-03-30, with the Radeon Pro GCN line as its successor.

Q: Which card supports ray tracing and AI acceleration?

A: Only the RTX 5080 Mobile. It has 60 RT cores and 240 tensor cores. The FirePro S9300 X2 has neither, consistent with its pre-ray-tracing GCN 3.0 architecture.

Q: How do their API support levels differ?

A: The RTX 5080 Mobile supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The FirePro S9300 X2 supports DirectX 12 (12_0), OpenGL 4.6, and Vulkan 1.2.170, so it lacks the Ultimate feature level.

Architecture Differences

These GPUs come from opposite ends of a decade of silicon development. The RTX 5080 Mobile uses the GB203 chip on Blackwell 2.0, fabricated at TSMC on a 5 nm node. It packs 45,600 million transistors into a 378 mm² die, yielding a density of 120.6M transistors per mm². The FirePro S9300 X2 uses the Capsaicin chip on GCN 3.0, built at TSMC on a 28 nm process, with 8,900 million transistors across a much larger 596 mm² die and a density of just 14.9M per mm². That is roughly an eightfold density gap in favor of Blackwell.

The functional blocks reflect that generational divide. NVIDIA's part includes 60 RT cores for ray tracing and 240 tensor cores for matrix and AI workloads, hardware categories that simply did not exist on GCN 3.0. The FirePro counters with a dual-GPU server design carrying 4096 shading units, 256 TMUs, and 64 ROPs, versus 7680 shading units, 240 TMUs, and 96 ROPs on the RTX 5080 Mobile.

Peak rates follow the hardware counts. The RTX 5080 Mobile manages 144.0 GPixel/s and 360.0 GTexel/s with 23.04 TFLOPS FP32 and 23.04 TFLOPS FP16 at 1:1. The FirePro S9300 X2 records 62.40 GPixel/s, 249.6 GTexel/s, and 7.987 TFLOPS FP32, with no FP16 figure listed.

Specification Differences

• Process node: 5 nm (RTX 5080 Mobile) vs 28 nm (FirePro S9300 X2)
• Transistors and die: 45,600 million on 378 mm² vs 8,900 million on 596 mm²
• Memory: 16 GB GDDR7, 256-bit, 896.0 GB/s vs 4 GB HBM, 4096-bit, 512.0 GB/s; memory clocks of 1750 MHz (28 Gbps effective) vs 500 MHz (1000 Mbps effective)
• Clocks: 975 MHz base and 1500 MHz boost on the NVIDIA card; no base or boost clocks recorded for the AMD card
• Shading hardware: 7680 shading units, 240 TMUs, 96 ROPs vs 4096 shading units, 256 TMUs, 64 ROPs
• Accelerators: 60 RT cores and 240 tensor cores vs none
• Interface: PCIe 5.0 x16 vs PCIe 3.0 x16
• Power and form factor: 80 W TDP, IGP slot, no power connectors vs 300 W TDP, dual-slot, 2x 8-pin, 700 W suggested PSU, 267 mm long and 111 mm tall
• Display: portable-device-dependent outputs vs no outputs at all
• APIs: DirectX 12 Ultimate (12_2) and Vulkan 1.4 vs DirectX 12 (12_0) and Vulkan 1.2.170
• Status: Active vs End-of-life
• Launch MSRP: the FirePro S9300 X2 carried a launch MSRP of 5,999 USD; no launch MSRP is recorded for the RTX 5080 Mobile

Where Each One Wins

The RTX 5080 Mobile wins every measured contest: both Geekbench OpenCL and Vulkan by enormous margins, 497 percent and 357.4 percent respectively. It is also the only card of the two with recorded results in 3DMark Steel Nomad, PassMark G3D, PassMark GPU Compute, and the PassMark DirectX runs, so every comparison beyond the two shared tests is territory where the FirePro has no data at all. On paper it also wins memory bandwidth (896.0 GB/s vs 512.0 GB/s), memory capacity (16 GB vs 4 GB), pixel and texture rates, FP32 throughput, interface bandwidth, and efficiency, delivering several times the performance at 80 W versus 300 W.

The FirePro S9300 X2 holds no benchmark wins. Its residual strengths are qualitative: the very wide 4096-bit HBM bus was an engineering signature of its generation, its dual-slot server form factor with no display outputs reflects its headless compute orientation, and its 256 TMUs actually exceed the RTX 5080 Mobile's 240. But the recorded data shows nothing where those traits translate into a victory. For any workload in the database, modern OpenCL and Vulkan compute included, the Blackwell mobile part is the clear choice, and it does so in a laptop-class power envelope while the FirePro demands a 700 W PSU.