Intel Arc B580 vs NVIDIA GeForce RTX 5080 SUPER Comparison

Intel Arc B580 and NVIDIA GeForce RTX 5080 SUPER occupy very different positions in the database, and the recorded measurements reflect that gap clearly. The RTX 5080 SUPER carries 4.1x the shading units, 4.2x the RT cores, and 4.1x the FP32 throughput of the Arc B580, yet in the single shared benchmark, the 3DMark Steel Nomad DX12 test, the NVIDIA card wins by just 0.2%. That narrow margin is the headline result here, though it comes with important context about how each GPU is positioned in the overall rankings.

Head-to-Head Benchmarks

The only direct comparison available in the database is the 3DMark Steel Nomad DX12 test. The Intel Arc B580 scores 3068, while the NVIDIA GeForce RTX 5080 SUPER scores 3075. The delta is a mere 0.2% in favor of NVIDIA, making this effectively a statistical tie in this particular workload. The RTX 5080 SUPER is recorded as the winner, but the margin is so small that the data does not support a meaningful performance advantage for either card in this test.

Looking at the broader average benchmark scores reveals a more complex picture. The Arc B580 has an average benchmark score of 23021 across ten recorded tests, while the RTX 5080 SUPER has an average score of 3075 based on a single test entry. This discrepancy is not a direct performance comparison, since the averages come from different test suites and different numbers of runs. The Arc B580's average is heavily influenced by non-3D workloads, including Geekbench OpenCL, Geekbench Vulkan, and Passmark compute tests, where its recorded scores are substantial.

The percentile data reinforces how differently these cards are classified. The Arc B580 sits at the 68th percentile of all GPUs in the database, placing it above the median. The RTX 5080 SUPER, despite its far larger hardware resources, sits at the 19th percentile. This is a direct consequence of the benchmark data available: the RTX 5080 SUPER has only one recorded test, the Steel Nomad result, while the Arc B580 has ten. The percentile field reflects the full set of recorded measurements, not raw hardware capability.

The nearest rival data further illustrates the gap. The Arc B580's closest competitors are the AMD Radeon RX 580 2048SP with an average score of 23061 and a delta of 0.2% in favor of AMD, the NVIDIA GeForce RTX 3080 at 23172 with a 0.7% delta in favor of the RTX 3080, the NVIDIA P106-100 at 23249 with a 1% delta, and the NVIDIA GeForce RTX 2080 at 22895 with a 0.6% delta in favor of the Arc B580. These are all older or lower-tier cards, and the Arc B580 trades blows with them within a narrow band of about 1%.

The RTX 5080 SUPER's nearest rivals are a different story entirely. The NVIDIA Quadro P1000 averages 3163, which is 2.8% higher than the RTX 5080 SUPER. The Intel Arc Pro B60 averages 3182, 3.4% higher. The NVIDIA GeForce 820A averages 2983, 3.1% lower, and the NVIDIA GeForce GTX 860M averages 2967, 3.6% lower. These rivals are professional or legacy mobile parts, not modern desktop gaming GPUs, which indicates that the database currently has very limited benchmark coverage for the RTX 5080 SUPER.

The Verdict

The data supports a straightforward conclusion: the RTX 5080 SUPER is the faster card on paper by a wide margin in nearly every hardware specification, but the recorded benchmark results do not yet reflect that advantage. The single Steel Nomad score shows a 0.2% win for NVIDIA, which is not a meaningful difference. Anyone looking at the database today would see two cards that perform almost identically in that one test, despite the RTX 5080 SUPER having 24 GB of GDDR7 memory versus 12 GB of GDDR6, 1.02 TB/s of bandwidth versus 456.0 GB/s, and 56.28 TFLOPS of FP32 throughput versus 13.67 TFLOPS.

The RTX 5080 SUPER is the card for users who need the raw throughput and memory capacity that its specifications describe. It delivers 24 GB of VRAM, which is double the Arc B580's 12 GB, and its 256 bit memory bus with 1.02 TB/s bandwidth is more than double the Arc B580's 192 bit bus and 456.0 GB/s. It also supports PCIe 5.0 x16, whereas the Arc B580 uses PCIe 4.0 x8. The Arc B580 is the card for users who want a lower-power, smaller-footprint GPU with a simpler power connector requirement.

The launch MSRP of the Arc B580 is 249 USD, and the launch MSRP of the RTX 5080 SUPER is 999 USD. The database does not provide enough benchmark data on the RTX 5080 SUPER to justify that price difference from measured performance alone. The Arc B580 has a much richer benchmark history, including ten separate tests, so its performance profile is far better understood.

Architecture Differences

The two GPUs come from different architectural families. The Arc B580 uses the Xe2-HPG architecture on the BMG-G21 chip, part of the Battlemage generation. The RTX 5080 SUPER uses the Blackwell 2.0 architecture on the GB203 chip, part of the GeForce 50 series. Both are fabricated on a 5 nm process at TSMC, so the manufacturing node is identical.

The transistor counts differ substantially. The Arc B580 packs 19,600 million transistors on a 272 mm² die, giving a transistor density of 72.1 million per square millimeter. The RTX 5080 SUPER packs 45,600 million transistors on a 378 mm² die, giving a density of 120.6 million per square millimeter. The NVIDIA chip has 2.3x the transistors on a die that is only 1.4x the area, which explains its higher density.

The compute layout is where the gap becomes obvious. The Arc B580 has 2560 shading units, 160 texture mapping units, and 80 render output units. The RTX 5080 SUPER has 10752 shading units, 336 TMUs, and 112 ROPs. The NVIDIA card has 4.2x the shading units, 2.1x the TMUs, and 1.4x the ROPs. Ray tracing hardware follows a similar pattern: 20 RT cores on the Arc B580 versus 84 on the RTX 5080 SUPER. The RTX 5080 SUPER also has 336 tensor cores, while the Arc B580 has no tensor core count recorded in the database.

Clock behavior differs as well. The Arc B580 has a base clock of 2670 MHz and a boost clock of 2670 MHz, meaning it runs at a flat frequency. The RTX 5080 SUPER has a base clock of 2295 MHz and a boost clock of 2617 MHz, so it starts lower and boosts higher. The memory clocks also differ: the Arc B580 uses 2375 MHz with 19 Gbps effective, while the RTX 5080 SUPER uses 2000 MHz with 32 Gbps effective.

The FP32 and FP16 throughput figures show a fundamental design difference. The Arc B580 delivers 13.67 TFLOPS FP32 and 27.34 TFLOPS FP16 with a 2:1 ratio. The RTX 5080 SUPER delivers 56.28 TFLOPS FP32 and 56.28 TFLOPS FP16 with a 1:1 ratio. The NVIDIA card does not halve its throughput when switching to FP16, which indicates a different execution path for that precision.

Specification Differences

The memory subsystem is a major differentiator. The Arc B580 has 12 GB of GDDR6 on a 192 bit bus, producing 456.0 GB/s of bandwidth. The RTX 5080 SUPER has 24 GB of GDDR7 on a 256 bit bus, producing 1.02 TB/s of bandwidth. That is double the capacity and more than double the bandwidth.

The power requirements are equally different. The Arc B580 has a TDP of 190 W and uses a single 8-pin power connector, with a suggested PSU of 450 W. The RTX 5080 SUPER has a TDP of 415 W and uses a single 16-pin power connector, with no suggested PSU recorded in the database.

Physical dimensions differ. The Arc B580 measures 272 mm in length, 115 mm in height, and 45 mm in width, making it a dual-slot card. The RTX 5080 SUPER measures 304 mm in length, 137 mm in height, and 40 mm in width, also dual-slot but longer and taller while being slightly thinner.

The bus interface is another clear split. The Arc B580 uses PCIe 4.0 x8, while the RTX 5080 SUPER uses PCIe 5.0 x16. Display outputs also differ: the Arc B580 has 1x HDMI 2.1a and 3x DisplayPort 2.1, while the RTX 5080 SUPER has 1x HDMI 2.1b and 3x DisplayPort 2.1b. Both support DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4.

The pixel and texture rates reflect the raw output capacity. The Arc B580 produces 213.6 GPixel/s and 427.2 GTexel/s. The RTX 5080 SUPER produces 293.1 GPixel/s and 879.3 GTexel/s. The NVIDIA card is 1.4x faster in pixel fill and 2.1x faster in texture fill.

FAQ

Q: Which card wins the 3DMark Steel Nomad DX12 test?

A: The NVIDIA GeForce RTX 5080 SUPER wins with a score of 3075, versus 3068 for the Intel Arc B580, a margin of 0.2%.

Q: How much memory does each card have?

A: The Intel Arc B580 has 12 GB of GDDR6 memory, while the NVIDIA GeForce RTX 5080 SUPER has 24 GB of GDDR7 memory.

Q: What is the power draw difference?

A: The Arc B580 has a TDP of 190 W and uses a single 8-pin power connector. The RTX 5080 SUPER has a TDP of 415 W and uses a single 16-pin power connector.

Q: How do their bandwidth figures compare?

A: The Arc B580 delivers 456.0 GB/s over a 192 bit bus, while the RTX 5080 SUPER delivers 1.02 TB/s over a 256 bit bus.

Q: Which card has more shading units?

A: The RTX 5080 SUPER has 10752 shading units, compared to 2560 on the Arc B580, a 4.2x difference.

Q: Are both cards on the same manufacturing process?

A: Yes, both are fabricated on a 5 nm process at TSMC, though the RTX 5080 SUPER has a larger die at 378 mm² versus 272 mm².

Where Each One Wins

The Arc B580 wins in efficiency-oriented scenarios. Its 190 W TDP is less than half the RTX 5080 SUPER's 415 W. It uses a standard 8-pin power connector rather than a 16-pin connector, and its 272 mm length fits in shorter cases. Its suggested PSU requirement of 450 W is recorded in the database, while the RTX 5080 SUPER has no suggested PSU listed. The Arc B580 also has a far more complete benchmark record, with ten tests covering DirectX 9 through DirectX 12, OpenCL, Vulkan, and compute workloads.

The RTX 5080 SUPER wins in raw specification categories. It has more than double the memory capacity, more than double the memory bandwidth, 4.2x the shading units, 4.2x the RT cores, and 4.1x the FP32 throughput. Its 336 tensor cores provide AI acceleration hardware that the Arc B580 does not list. Its 1.02 TB/s bandwidth and 24 GB capacity place it in a different performance class for memory-bound workloads. The PCIe 5.0 x16 interface provides double the lane width and a newer generation compared to the Arc B580's PCIe 4.0 x8.

The single shared benchmark does not separate these cards meaningfully. The 0.2% delta in Steel Nomad is within the noise of any GPU comparison. The database currently records one test for the RTX 5080 SUPER and ten for the Arc B580, so the NVIDIA card's performance profile is incomplete. The Arc B580's 68th percentile ranking versus the RTX 5080 SUPER's 19th percentile ranking reflects that data imbalance, not a true performance hierarchy. The hardware specifications describe a much larger gap than the recorded measurements currently show.