Intel Arc Pro B60 vs NVIDIA RTX PRO 4000 Blackwell SFF Comparison

The Intel Arc Pro B60 and NVIDIA RTX PRO 4000 Blackwell SFF are two professional workstation graphics cards that target the same compact, dual-slot market segment. Both cards feature 24 GB of memory, identical physical dimensions, and share the same PCIe 5.0 x8 bus interface. However, the data reveals starkly different design philosophies: the Intel card consumes nearly three times the power of the NVIDIA card, while the NVIDIA card delivers a 9.1% higher score in the single available head-to-head benchmark, the 3DMark Steel Nomad DX12 test. The benchmark results indicate that these are closely matched competitors in raw performance, but their architectural and efficiency profiles diverge dramatically.

FAQ

Q: Which card is faster in the 3DMark Steel Nomad DX12 benchmark?

A: The NVIDIA RTX PRO 4000 Blackwell SFF wins this test with a score of 2910, compared to the Intel Arc Pro B60's score of 2646. This represents a 9.1% advantage for the NVIDIA card.

Q: Do both cards have the same amount of memory?

A: Yes, both the Intel Arc Pro B60 and the NVIDIA RTX PRO 4000 Blackwell SFF feature 24 GB of memory. However, the Intel card uses GDDR6 memory, while the NVIDIA card uses the newer GDDR7 type.

Q: What is the power consumption difference between these two cards?

A: The Intel Arc Pro B60 has a TDP of 200 W and requires a 550 W power supply, whereas the NVIDIA RTX PRO 4000 Blackwell SFF has a significantly lower TDP of 70 W and only needs a 250 W power supply. The NVIDIA card also draws power entirely from the PCIe slot, requiring no external power connectors.

Q: How do these cards compare to their nearest rivals in the database?

A: The Intel card's average benchmark score of 3182 puts it 0.6% ahead of the NVIDIA Quadro P1000, while the NVIDIA RTX PRO 4000's average score of 2910 places it 0.1% ahead of the GeForce RTX 4060 Ti 16 GB.

Q: Which card has a higher transistor count?

A: The NVIDIA RTX PRO 4000 Blackwell SFF has 45,600 million transistors on a 378 mm² die, while the Intel Arc Pro B60 has 19,600 million transistors on a smaller 272 mm² die.

Q: Are these cards the same physical size?

A: Yes, both cards share identical dimensions: 167 mm in length, 69 mm in height, and 40 mm in width, and both are dual-slot designs.

Architecture Differences

The architectural gap between these two cards is substantial. The Intel Arc Pro B60 is built on the Xe2-HPG architecture, belonging to the Battlemage (Pro Series) generation, and uses the BMG-G21 chip. In contrast, the NVIDIA RTX PRO 4000 Blackwell SFF employs the Blackwell 2.0 architecture from the Blackwell PRO W (x000) generation, built on the GB203 chip. Both are manufactured by TSMC on a 5 nm process, but that is where the similarities end.

The transistor counts reveal a massive difference in complexity. NVIDIA's GB203 packs 45,600 million transistors into a 378 mm² die, achieving a transistor density of 120.6M per square millimeter. Intel's BMG-G21 contains just 19,600 million transistors on a 272 mm² die, resulting in a lower density of 72.1M per square millimeter. This disparity translates directly into compute resources: the NVIDIA card features 8960 shading units, 280 TMUs, and 96 ROPs, while the Intel card has only 2560 shading units, 160 TMUs, and 80 ROPs.

The ray tracing and AI acceleration hardware also differs significantly. NVIDIA includes 70 RT cores and 280 tensor cores, whereas Intel offers 20 RT cores and no tensor cores listed. This suggests the NVIDIA card is far more capable in neural rendering and ray-traced workloads. Clock speeds present another stark contrast: the Intel card operates at a 2000 MHz base and 2400 MHz boost, while the NVIDIA card runs at a much lower 405 MHz base and 1342 MHz boost. The NVIDIA card compensates with its wider compute architecture to achieve higher raw throughput, reaching 24.05 TFLOPS FP32 versus Intel's 12.29 TFLOPS.

Memory subsystems show a similar divergence. The Intel card uses 24 GB of GDDR6 at 19 Gbps effective, delivering 456.0 GB/s of bandwidth. The NVIDIA card uses 24 GB of GDDR7 at 18 Gbps effective, providing 432.0 GB/s. Despite the newer memory type, the NVIDIA card has slightly lower bandwidth due to its lower memory clock speed. Both use a 192-bit bus width. The Intel card's higher memory clock does give it a small bandwidth advantage, though the NVIDIA card's overall compute lead is much more pronounced.

Head-to-Head Benchmarks

The only direct comparison available between these two cards is the 3DMark Steel Nomad DX12 test, a demanding modern gaming and rendering benchmark. In this test, the NVIDIA RTX PRO 4000 Blackwell SFF scores 2910, while the Intel Arc Pro B60 scores 2646. The NVIDIA card wins this head-to-head matchup with a 9.1% margin. This result is consistent with the broader benchmark data, which shows the NVIDIA card occupying the 19th percentile among all GPUs, while the Intel card sits at the 20th percentile. The average benchmark scores are 2910 for NVIDIA and 3182 for Intel, though this comparison is complicated by the fact that the Intel card has eight benchmark entries while the NVIDIA card only has one.

The NVIDIA card's nearest rival in the database is the GeForce RTX 4060 Ti 16 GB, which scores an average of 2907, putting the RTX PRO 4000 just 0.1% ahead. It also sits within 0.4% of the Quadro P600 and 0.6% of the GeForce RTX 4010. The Intel card's nearest rival is the NVIDIA Quadro P1000, which scores 3163, placing the Arc Pro B60 just 0.6% ahead. The Intel card also sits 0.9% behind the GeForce GT 640 and 3.2% behind the GeForce 920M.

The 9.1% delta in the head-to-head test is meaningful, but the overall benchmark landscape suggests these cards are far closer than their architectural differences might imply. The Intel card's higher clock speeds and memory bandwidth partially offset its smaller compute footprint in the Steel Nomad test. Still, the NVIDIA card's victory in this single metric indicates it holds an edge in DirectX 12 rendering workloads. For professional users relying on this specific benchmark, the RTX PRO 4000 is the superior choice.

Specification Differences

The specification sheets for these two cards diverge on nearly every technical field. The most striking difference is power consumption: the Intel Arc Pro B60 has a TDP of 200 W and requires a 550 W power supply, while the NVIDIA RTX PRO 4000 Blackwell SFF has a TDP of just 70 W and a suggested power supply of 250 W. The NVIDIA card draws power entirely from the PCIe slot, with no external power connectors listed, whereas the Intel card requires a single 8-pin connector.

Compute resources differ dramatically. The NVIDIA card has 8960 shading units compared to Intel's 2560, 280 TMUs versus 160, 96 ROPs versus 80, 70 RT cores versus 20, and 280 tensor cores versus none. This gives the NVIDIA card nearly double the FP32 throughput at 24.05 TFLOPS versus 12.29 TFLOPS. The FP16 performance also differs: the Intel card achieves 24.58 TFLOPS with a 2:1 ratio, while the NVIDIA card achieves 24.05 TFLOPS with a 1:1 ratio.

Clock speeds are significantly higher on the Intel card: 2000 MHz base and 2400 MHz boost compared to NVIDIA's 405 MHz base and 1342 MHz boost. Memory clocks also favor Intel at 2375 MHz (19 Gbps effective) versus NVIDIA's 1125 MHz (18 Gbps effective). However, the NVIDIA card uses GDDR7 memory while the Intel card uses GDDR6. Bandwidth is marginally higher on the Intel card at 456.0 GB/s versus 432.0 GB/s.

The display outputs differ slightly: both feature four mini-DisplayPort connections, but the Intel card uses DisplayPort 2.1 while the NVIDIA card uses DisplayPort 2.1b. The release dates are close but not identical, with Intel launching on September 4, 2025 and NVIDIA on August 10, 2025. The NVIDIA card has a predecessor in Workstation Ada, while the Intel card lists no predecessor. The Intel card has a launch MSRP of 499 USD, while the NVIDIA card's launch MSRP is not listed.

The Verdict

The data presents a clear trade-off between these two professional GPUs. The NVIDIA RTX PRO 4000 Blackwell SFF wins the sole head-to-head benchmark, delivering 2910 points versus 2646 for the Intel Arc Pro B60, a 9.1% advantage in 3DMark Steel Nomad DX12. For users prioritizing maximum rendering performance in this DirectX 12 test, the NVIDIA card is the better choice.

However, the efficiency story is overwhelmingly in NVIDIA's favor. The RTX PRO 4000 achieves its higher score while consuming just 70 W of power, compared to the Intel card's 200 W. This threefold difference in power draw means the NVIDIA card can operate in systems with a 250 W power supply, while the Intel card demands a 550 W unit. The NVIDIA card's slot-only power design also simplifies installation in compact systems, as it requires no external power connectors.

The architectural data suggests NVIDIA's card is more future-proof. With 8960 shading units, 70 RT cores, and 280 tensor cores, the RTX PRO 4000 has vastly more compute resources than the Intel card's 2560 shading units, 20 RT cores, and no tensor cores. This gives NVIDIA a significant advantage in ray-traced and AI-accelerated workloads, even if the current benchmark suite does not fully capture those differences. The Intel card's higher clock speeds and slightly higher memory bandwidth help it remain competitive in the tested scenario, but the underlying hardware is less capable.

For users who need a compact, dual-slot professional card with 24 GB of memory and the lowest possible power requirements, the NVIDIA RTX PRO 4000 Blackwell SFF is the clear winner based on the data. It offers better performance in the available benchmark, superior efficiency, and a more advanced feature set. The Intel Arc Pro B60, with its 499 USD launch MSRP and higher power envelope, is positioned for users who prioritize raw clock speeds and memory bandwidth over compute density and power efficiency. The benchmark results indicate that NVIDIA's architecture delivers more performance per watt, and the 9.1% margin in the head-to-head test confirms its dominance in this comparison.