Intel Arc Pro B60 Dual vs NVIDIA GeForce RTX 5070 SUPER Comparison

Head-to-Head Benchmarks

The recorded database contains a single benchmark result for the NVIDIA GeForce RTX 5070 SUPER: the 3DMark Steel Nomad DX12 test, where it scores 2690. The Intel Arc Pro B60 Dual has no benchmark entries in the database, so direct score comparisons cannot be established from measured data. However, the RTX 5070 SUPER's percentile placement and nearest rival scores provide context for interpreting its position.

The RTX 5070 SUPER sits at the 18th percentile among all GPUs in the database, meaning roughly 82 percent of recorded graphics cards score higher in this specific test. Its nearest rivals in the database are all older or lower-tier parts. The NVIDIA Quadro K1100M averages 2664, placing the RTX 5070 SUPER just 1 percent ahead. The GeForce GT 1030 averages 2662, a 1.1 percent gap. The Intel Arc Pro B50 averages 2660, also a 1.1 percent difference. The GeForce GT 440 averages 2645, with the RTX 5070 SUPER leading by 1.7 percent.

These deltas are exceptionally narrow. A 1 to 1.7 percent advantage over a mobile workstation GPU from years ago, a low-end desktop card, and a professional Intel part suggests the RTX 5070 SUPER's 3DMark Steel Nomad score is low relative to what its specifications would imply. The card's FP32 throughput is 32.15 TFLOPS, its boost clock is 2512 MHz, and it carries 6400 shading units. Yet in this benchmark, it lands barely above the GT 440, a card from a different era entirely. The database records no head-to-head benchmark entries between the two cards in question, and the wins counter shows zero for both sides.

The Intel Arc Pro B60 Dual, meanwhile, has no measured benchmark score in the database and holds a 50th percentile rank among all GPUs, which is a placeholder position rather than a derived result. Its average benchmark score is recorded as zero, indicating no completed tests. Without a Steel Nomad score for the Intel card, the only quantitative comparison available is through the RTX 5070 SUPER's nearest rivals, none of which is the Arc Pro B60 Dual. The nearest Intel entry, the Arc Pro B50, trails the RTX 5070 SUPER by 1.1 percent in this test.

What the data shows is a narrow clustering at the low end of the score distribution. The RTX 5070 SUPER's 2690 score places it among cards that typically serve entry-level or legacy workloads. The 1 percent delta against the Quadro K1100M, a mobile GPU, and the 1.1 percent delta against the GT 1030, a passive-cooled desktop card, indicate that the RTX 5070 SUPER does not separate itself in this particular DirectX 12 workload. The 1.7 percent lead over the GT 440 is the largest margin in its nearest rival set, yet 1.7 percent is a statistically minor difference.

FAQ

Q: What is the RTX 5070 SUPER's score in the 3DMark Steel Nomad DX12 test?

A: The database records a score of 2690 for the NVIDIA GeForce RTX 5070 SUPER in the 3dmark_3dmark_steel_nomad_dx12 test.

Q: How does the RTX 5070 SUPER compare to its nearest rivals in the database?

A: It leads the NVIDIA Quadro K1100M by 1 percent (2664), the GeForce GT 1030 by 1.1 percent (2662), the Intel Arc Pro B50 by 1.1 percent (2660), and the GeForce GT 440 by 1.7 percent (2645).

Q: Does the Intel Arc Pro B60 Dual have any benchmark scores in the database?

A: No. The Intel card has an empty benchmarks array, an average benchmark score of zero, and a percentile of 50 that reflects no measured data.

Q: What is the RTX 5070 SUPER's percentile ranking among all GPUs?

A: The card sits at the 18th percentile, meaning most recorded GPUs score higher in the Steel Nomad DX12 test.

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

A: The Intel Arc Pro B60 Dual has 24 GB of GDDR6, while the RTX 5070 SUPER has 18 GB of GDDR7.

Q: What are the boost clock speeds for both cards?

A: The Intel card boosts to 2400 MHz, and the RTX 5070 SUPER boosts to 2512 MHz.

Where Each One Wins

The RTX 5070 SUPER wins in raw compute throughput. Its FP32 performance is 32.15 TFLOPS versus 12.29 TFLOPS for the Intel Arc Pro B60 Dual, a gap of roughly 2.6 times. Its FP16 output is 32.15 TFLOPS at a 1:1 ratio, while the Intel card delivers 24.58 TFLOPS at a 2:1 ratio, meaning the NVIDIA part maintains full-rate FP16 without the doubling penalty. Texture rate favors NVIDIA at 502.4 GTexel/s against 384.0 GTexel/s, and pixel rate favors NVIDIA at 201.0 GPixel/s versus 192.0 GPixel/s. The RTX 5070 SUPER also has more shading units (6400 versus 2560), more TMUs (200 versus 160), more RT cores (50 versus 20), and 200 tensor cores where the Intel card lists none.

Memory bandwidth is another NVIDIA win. The RTX 5070 SUPER's GDDR7 memory runs at 28 Gbps effective across a 192-bit bus, yielding 672.0 GB/s. The Intel card's GDDR6 runs at 19 Gbps effective on the same 192-bit bus width, yielding 456.0 GB/s. That is a 47 percent bandwidth advantage for NVIDIA. The RTX 5070 SUPER also has a higher base clock (2325 MHz versus 2000 MHz) and a higher boost clock (2512 MHz versus 2400 MHz).

The Intel Arc Pro B60 Dual wins on memory capacity and power efficiency relative to its score potential. It carries 24 GB versus 18 GB, a 6 GB difference that matters for large datasets or multi-application workflows. Its power draw is 400 W against 275 W for the RTX 5070 SUPER, so NVIDIA consumes less power while delivering higher throughput, but the Intel card's higher TDP suggests it is not constrained in the same way. The Intel card uses a PCIe 5.0 x8 interface, while the RTX 5070 SUPER uses PCIe 5.0 x16, giving NVIDIA double the lane width for host communication. The Intel card is also longer at 300 mm versus 245 mm, and taller at 110 mm versus 115 mm, with the same 40 mm width.

In terms of measured benchmark results, the RTX 5070 SUPER is the only card with a score, so it wins the only recorded comparison by default. The Intel card's zero benchmark average and empty results list mean it cannot claim any measured victory. The database records zero wins for both cards in head-to-head comparisons, reflecting the absence of paired test data.

Specification Differences

The two cards differ in nearly every measurable specification. The Intel Arc Pro B60 Dual uses the BMG-G21 chip with Xe2-HPG architecture, while the RTX 5070 SUPER uses the GB205 chip with Blackwell 2.0 architecture. Both are built on a 5 nm process at TSMC, but transistor counts differ substantially: Intel packs 19,600 million transistors on a 272 mm² die, while NVIDIA packs 31,100 million on a slightly smaller 263 mm² die. Transistor density favors NVIDIA at 118.3M per mm² versus 72.1M per mm².

Clock speeds differ. The Intel card has a 2000 MHz base and 2400 MHz boost. The RTX 5070 SUPER has a 2325 MHz base and 2512 MHz boost. Memory clocks also differ: Intel uses 2375 MHz with 19 Gbps effective, while NVIDIA uses 1750 MHz with 28 Gbps effective, a difference driven by memory type (GDDR6 versus GDDR7).

Memory capacity and bandwidth differ as noted: 24 GB GDDR6 at 456.0 GB/s for Intel, 18 GB GDDR7 at 672.0 GB/s for NVIDIA. Both use a 192-bit bus. Shading units are 2560 for Intel and 6400 for NVIDIA. TMUs are 160 versus 200. ROPs are identical at 80. RT cores are 20 versus 50. Tensor cores exist only on the NVIDIA card at 200. Pixel rate is 192.0 GPixel/s for Intel and 201.0 GPixel/s for NVIDIA. Texture rate is 384.0 GTexel/s versus 502.4 GTexel/s. FP32 is 12.29 TFLOPS versus 32.15 TFLOPS. FP16 is 24.58 TFLOPS (2:1) versus 32.15 TFLOPS (1:1).

Power specifications differ: Intel lists a 400 W TDP and an 800 W suggested PSU, while NVIDIA lists 275 W TDP and no suggested PSU. Both use a single 16-pin connector and dual-slot cooling. Bus interfaces differ: PCIe 5.0 x8 for Intel, PCIe 5.0 x16 for NVIDIA. Display outputs differ: Intel provides 4x mini-DisplayPort 2.1, while NVIDIA provides 1x HDMI 2.1b and 3x DisplayPort 2.1b. Dimensions differ: Intel is 300 mm long and 110 mm tall, NVIDIA is 245 mm long and 115 mm tall, with identical 40 mm width. Release dates differ: Intel launched on September 4, 2025, while NVIDIA's release is recorded as December 31, 2025.

Architecture Differences

The architectural split is clear. Intel uses Xe2-HPG on the Battlemage Pro Series generation, built around the BMG-G21 die. NVIDIA uses Blackwell 2.0 on the GeForce 50-series, built around the GB205 die. Both are TSMC 5 nm parts, but NVIDIA's design achieves a higher transistor density (118.3M per mm²) despite the smaller die (263 mm² versus 272 mm²), indicating a more compact layout.

The shading unit count is the largest architectural differentiator. NVIDIA's 6400 shading units are 2.5 times Intel's 2560, and its 50 RT cores are 2.5 times Intel's 20. NVIDIA also integrates 200 tensor cores, which Intel's card does not list at all. This suggests the RTX 5070 SUPER targets AI-accelerated workloads, while the Intel card relies on its Xe2-HPG vector and ray tracing units alone.

Cache and memory hierarchy are not fully specified in the database, but the memory type difference (GDDR6 versus GDDR7) reflects a generational gap. NVIDIA's 1:1 FP16 ratio means its FP16 throughput matches FP32, whereas Intel's 2:1 ratio halves FP16 output relative to its FP32 peak. This makes the RTX 5070 SUPER more efficient for mixed-precision compute. The PCIe interface also differs: NVIDIA uses x16 lanes, Intel uses x8, which affects host data transfer capacity.

Both cards support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, so API compatibility is identical. The display output configurations differ, with Intel favoring multiple mini-DisplayPort 2.1 connections and NVIDIA mixing one HDMI 2.1b with three DisplayPort 2.1b.

The Verdict

The database presents an unusual comparison. The RTX 5070 SUPER has one measured benchmark score (2690 in 3DMark Steel Nomad DX12) and a percentile rank of 18, placing it below most recorded GPUs. The Intel Arc Pro B60 Dual has no measured scores, a placeholder percentile of 50, and an average benchmark score of zero. Neither card has a recorded head-to-head benchmark against the other.

For users selecting based on measured data, the RTX 5070 SUPER is the only card with a quantifiable result. Its score of 2690 places it within 1.7 percent of the GeForce GT 440, a legacy part. This indicates that, at least in the Steel Nomad DX12 workload, the RTX 5070 SUPER's high specifications (32.15 TFLOPS, 6400 shading units, 672.0 GB/s bandwidth) do not translate into a strong benchmark showing. The 18th percentile rank reinforces this: 82 percent of all GPUs in the database outperform it in this test.

The Intel Arc Pro B60 Dual offers 24 GB of memory, a 192-bit bus, and 12.29 TFLOPS of FP32 throughput, but without a benchmark score its practical performance is unmeasured. Its 400 W TDP and 800 W suggested PSU indicate a high-power design, and its PCIe 5.0 x8 interface may limit host bandwidth compared to the RTX 5070 SUPER's x16 connection.

The data supports different choices depending on the priority. A user requiring measured DirectX 12 performance should select the RTX 5070 SUPER, as it has a recorded score and a defined percentile. A user requiring maximum memory capacity should select the Intel Arc Pro B60 Dual, as its 24 GB exceeds the RTX 5070 SUPER's 18 GB. A user requiring lower power draw should select the RTX 5070 SUPER, as its 275 W TDP is 125 W lower than the Intel card's 400 W. A user requiring tensor cores for AI workloads should select the RTX 5070 SUPER, as the Intel card lists none.

The launch MSRP for the Intel Arc Pro B60 Dual is 1,199 USD. The RTX 5070 SUPER has no recorded launch MSRP. The database records zero wins for either card in head-to-head comparisons, so no direct winner can be declared from paired tests. The verdict rests on specification differences and the single available benchmark score.