Intel Arc Pro B65 vs NVIDIA GeForce RTX 5090 Comparison

Intel Arc Pro B65 and NVIDIA GeForce RTX 5090 occupy distant corners of the GPU landscape, one a professional-focused workstation part, the other a flagship consumer card. The data reveals a stark asymmetry: the RTX 5090 holds every benchmark win, but the Arc Pro B65 counters with architectural efficiency and a far lower power footprint. The recorded measurements show the RTX 5090 dominating raw performance, while the Arc Pro B65 delivers a competitive feature set for its intended segment.

Where Each One Wins

The benchmark database shows the RTX 5090 winning every recorded test, with no wins recorded for the Arc Pro B65. The RTX 5090's percentile rank of 92 versus the Arc Pro B65's 50 places them in entirely different performance tiers. The Arc Pro B65 is positioned for professional workloads where power draw and thermal output matter as much as raw throughput, while the RTX 5090 targets maximum compute capability regardless of energy cost.

The RTX 5090 wins decisively across every benchmark category: 3DMark Steel Nomad DX12, Geekbench OpenCL and Vulkan, Passmark DirectX 9 through 12, Passmark G2D, G3D, and GPU compute. The Arc Pro B65 has no recorded benchmark scores in the database, meaning its wins are structural rather than measured. It wins on power efficiency by specification, drawing 200 W versus the RTX 5090's 575 W, and on physical footprint, using a single 8-pin connector versus the RTX 5090's 16-pin.

For use cases, the data suggests the Arc Pro B65 suits environments with strict power budgets or workstation deployments where multiple cards might populate a single chassis. The RTX 5090 occupies the upper performance stratum, as its 92nd percentile indicates, with an average benchmark score of 79,842. The Arc Pro B65's zero benchmark scores and 50th percentile reflect an unmeasured or unmarketed position in the database.

Architecture Differences

Both GPUs use a 5 nm process from TSMC, but the transistor counts diverge dramatically. The Intel Arc Pro B65 packs 19,600 million transistors on a 272 mm² die, yielding a density of 72.1 million transistors per square millimeter. The NVIDIA GeForce RTX 5090 contains 92,200 million transistors across a 750 mm² die, achieving 122.9 million per square millimeter. The RTX 5090's die is nearly three times larger and holds nearly five times the transistors.

The Intel chip, designated BMG-G21, uses the Xe2-HPG architecture from the Battlemage Pro Series. The NVIDIA chip, GB202, employs Blackwell 2.0 from the GeForce 50 generation. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, so API compatibility is identical for software targeting those standards.

Memory architecture differs fundamentally. The Arc Pro B65 uses 32 GB of GDDR6 on a 256-bit bus, reaching 608.0 GB/s bandwidth. The RTX 5090 also has 32 GB, but uses GDDR7 on a 512-bit bus, achieving 1.79 TB/s, nearly three times the bandwidth. The RTX 5090's memory clock runs at 1750 MHz (28 Gbps effective) versus the Arc Pro B65's 2375 MHz (19 Gbps effective), but the wider bus delivers the bandwidth advantage.

Core configurations show a massive scale difference. The RTX 5090 has 21,760 shading units, 680 TMUs, 176 ROPs, 170 RT cores, and 680 tensor cores. The Arc Pro B65 has 2,560 shading units, 160 TMUs, 80 ROPs, and 20 RT cores, with no tensor cores listed. The RTX 5090's FP32 throughput of 104.8 TFLOPS compared to the Arc Pro B65's 12.29 TFLOPS is an 8.5 times gap. FP16 performance is identical at 104.8 TFLOPS for the RTX 5090 (1:1 rate) while the Arc Pro B65 achieves 24.58 TFLOPS at a 2:1 ratio.

Head-to-Head Benchmarks

The head-to-head benchmark list between these two cards is empty in the database, so comparisons rely on the RTX 5090's standalone scores and the Arc Pro B65's absence of any. The RTX 5090's Passmark G3D score of 39,650 and GPU compute score of 26,756 represent its strongest recorded results. The Geekbench Vulkan score of 376,728 exceeds the OpenCL score of 334,370, suggesting Vulkan optimization works slightly better for this architecture.

Passmark results show interesting variance across DirectX versions. The RTX 5090 scores 395 in DirectX 9, 341 in DirectX 11, 226 in DirectX 10, and only 185 in DirectX 12. This inverse relationship between API age and score indicates the card's driver stack may prioritize newer workloads differently, or that legacy APIs run efficiently while modern ones face overhead. The 3DMark Steel Nomad DX12 score of 18,355 provides a modern benchmark reference point.

Without Arc Pro B65 scores, relative deltas cannot be computed directly. The RTX 5090's nearest rivals in the database include the NVIDIA Tesla P100 PCIe 16 GB (average score 79,605, delta 0.3%), the Tesla P100 PCIe 12 GB (79,396, delta 0.6%), the AMD Radeon RX 6850M XT (78,940, delta 1.1%), and the AMD Radeon Pro Vega 64X (80,959, delta -1.4%). These deltas indicate the RTX 5090 sits within 1.4% of its closest competitors, meaning its performance class is well populated.

FAQ

Q: Which GPU has higher memory bandwidth?

A: The NVIDIA GeForce RTX 5090 achieves 1.79 TB/s using GDDR7 on a 512-bit bus, while the Intel Arc Pro B65 reaches 608.0 GB/s with GDDR6 on a 256-bit bus.

Q: Do both cards support the same graphics APIs?

A: Yes, both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4 according to the database.

Q: What are the power consumption differences?

A: The Arc Pro B65 has a TDP of 200 W with a suggested PSU of 550 W and a single 8-pin connector. The RTX 5090 has a TDP of 575 W, a suggested PSU of 950 W, and a single 16-pin connector.

Q: How many display outputs does each card provide?

A: The Intel Arc Pro B65 has 4x DisplayPort 2.1 outputs. The NVIDIA GeForce RTX 5090 has 1x HDMI 2.1b and 3x DisplayPort 2.1b outputs.

Q: What is the transistor density of each GPU?

A: The Arc Pro B65 has 72.1 million transistors per square millimeter, while the RTX 5090 has 122.9 million per square millimeter on the same 5 nm TSMC process.

Q: Which card has a higher boost clock?

A: The NVIDIA GeForce RTX 5090 boosts to 2407 MHz, while the Intel Arc Pro B65 has a base and boost clock of 2400 MHz, which are identical.

The Verdict

The data presents a clear performance hierarchy: the RTX 5090 dominates every measured category with an average benchmark score of 79,842 and a 92nd percentile rank. The Arc Pro B65's zero benchmark scores and 50th percentile mean the database cannot substantiate any performance claim for it. Users requiring maximum compute throughput, as indicated by the RTX 5090's 104.8 TFLOPS FP32 and its 170 RT cores, should select the NVIDIA card.

The Arc Pro B65 makes sense for environments where the RTX 5090's 575 W TDP and 950 W suggested PSU are prohibitive. At 200 W with a 550 W suggested PSU, the Intel card fits into existing workstation power budgets without infrastructure changes. Its 32 GB memory matches the RTX 5090's capacity, and its 4x DisplayPort 2.1 outputs exceed the RTX 5090's mixed HDMI and DisplayPort arrangement for multi-display professional setups.

The RTX 5090's nearest rival data shows it performs within 1.4% of the AMD Radeon Pro Vega 64X and within 0.3% of the Tesla P100 16 GB, placing it in a competitive field rather than an isolated peak. The Arc Pro B65's lack of measured scores leaves its actual performance unquantified, so the database only supports choosing it for power, thermal, or display configuration reasons, not for raw throughput.

Specification Differences

The two GPUs differ across nearly every specification field. The Arc Pro B65 uses the BMG-G21 chip with Xe2-HPG architecture from Intel's Battlemage Pro Series, while the RTX 5090 uses the GB202 chip with Blackwell 2.0 from NVIDIA's GeForce 50 generation. Both use 5 nm TSMC fabrication, but the RTX 5090 has 92,200 million transistors versus 19,600 million, and a 750 mm² die versus 272 mm².

Clock speeds differ slightly: the Arc Pro B65 runs at 2400 MHz base and boost, while the RTX 5090 has a 2017 MHz base and 2407 MHz boost. Memory configurations share 32 GB capacity but diverge in type, bus width, and bandwidth: GDDR6 on 256-bit at 608.0 GB/s for Intel, GDDR7 on 512-bit at 1.79 TB/s for NVIDIA.

Compute resources show the largest gaps. The RTX 5090 has 21,760 shading units, 680 TMUs, 176 ROPs, 170 RT cores, and 680 tensor cores. The Arc Pro B65 has 2,560 shading units, 160 TMUs, 80 ROPs, and 20 RT cores, with no tensor cores listed. Pixel rate is 423.6 GPixel/s for NVIDIA versus 192.0 GPixel/s for Intel, and texture rate is 1,636.8 GTexel/s versus 384.0 GTexel/s.

Power and physical specifications also differ. The RTX 5090 draws 575 W with a 950 W suggested PSU and a 16-pin connector, while the Arc Pro B65 draws 200 W with a 550 W suggested PSU and an 8-pin connector. Both are dual-slot cards, but the RTX 5090 has recorded dimensions of 304 mm length, 137 mm height, and 40 mm width, while the Arc Pro B65 has no recorded dimensions. Display outputs favor Intel with 4x DisplayPort 2.1, while NVIDIA offers 1x HDMI 2.1b and 3x DisplayPort 2.1b. The RTX 5090 has a launch MSRP of 1,999 USD, while the Arc Pro B65 has no recorded launch price.