Intel Arc Pro B60 vs NVIDIA GeForce RTX 4010 Comparison

Intel Arc Pro B60 and NVIDIA GeForce RTX 4010 are both active, dual-slot or slimmer professional/consumer hybrid cards, yet the data reveals they are engineered for entirely different workloads. The Arc Pro B60 is a 200 W, 24 GB workstation-oriented card built on Intel's Xe2-HPG architecture, while the RTX 4010 is a 50 W, 4 GB entry-level Ampere part. The single head-to-head benchmark available shows the RTX 4010 taking a narrow victory, but the surrounding specification and benchmark data paint a far more nuanced picture of where each card actually excels.

Head-to-Head Benchmarks

The only direct comparison in the data is the 3DMark Steel Nomad DX12 test, and it is a tight race. The NVIDIA GeForce RTX 4010 scores 2893, while the Intel Arc Pro B60 scores 2646. That translates to a delta of -8.5% for Intel, meaning NVIDIA wins by roughly 9%. This is a surprising result given the massive differences in memory capacity, bandwidth, and raw compute throughput between the two cards. The RTX 4010’s win suggests that its Ampere architecture, with 768 shading units and 24 tensor cores, is highly efficient in this particular DX12 workload, even with a 64-bit memory bus and just 96.00 GB/s of bandwidth.

However, the broader benchmark picture complicates this narrative. The Arc Pro B60 has a PassMark G3D score of 14580, which is a substantial number, while the RTX 4010 has no corresponding PassMark results listed. The Arc Pro B60 also posts a PassMark GPU Compute score of 7029, a PassMark DirectX 12 score of 76, and a DirectX 11 score of 122. The RTX 4010’s only listed benchmark aside from Steel Nomad is its average score of 2893, which is exactly its Steel Nomad result. This means the RTX 4010’s victory is isolated to a single test, whereas the Arc Pro B60 demonstrates broader capability across legacy and compute-oriented workloads.

Looking at the nearest rivals further contextualizes the two cards. The Arc Pro B60’s average benchmark score is 3182, placing it just 0.6% ahead of the NVIDIA Quadro P1000 (3163) and 3.5% ahead of the GeForce RTX 5080 SUPER (3075). It is 0.9% behind the GeForce GT 640 (3210) and 3.2% behind the GeForce 920M (3287). These are tight margins, indicating the Arc Pro B60 sits in a highly competitive mid-range cluster. The RTX 4010, with an average score of 2893, is 0.5% behind the RTX 4060 Ti 16 GB (2907) and 1% behind the Quadro P600 (2923), but it is only 0.6% and 0.7% behind the RTX PRO 4000 Blackwell SFF and RTX 4060 Ti 8 GB respectively. The RTX 4010’s percentile rank is 18, while the Arc Pro B60 sits at 20, so both are in the lower quartile of all GPUs, but Intel edges out NVIDIA in overall standing.

FAQ

Q: Which card wins the only direct benchmark comparison?

A: The NVIDIA GeForce RTX 4010 wins the 3DMark Steel Nomad DX12 test with a score of 2893, beating the Intel Arc Pro B60’s 2646 by a margin of -8.5% (i.e., Intel is 8.5% slower).

Q: How much memory does each card have, and does that affect their benchmark scores?

A: The Intel Arc Pro B60 has 24 GB of GDDR6 memory on a 192-bit bus, delivering 456.0 GB/s of bandwidth. The NVIDIA GeForce RTX 4010 has 4 GB of GDDR6 on a 64-bit bus, with only 96.00 GB/s. Despite this 4.75x bandwidth disadvantage, the RTX 4010 still wins the Steel Nomad test, suggesting that memory bandwidth is not the limiting factor in that specific workload.

Q: What are the compute performance figures for each card?

A: The Arc Pro B60 offers 12.29 TFLOPS of FP32 performance and 24.58 TFLOPS of FP16 (2:1 ratio). The RTX 4010 provides 2.706 TFLOPS of FP32 and 2.706 TFLOPS of FP16 (1:1 ratio). This means the Intel card has roughly 4.5x the FP32 throughput and 9x the FP16 throughput of the NVIDIA card.

Q: How do the two cards compare in terms of power consumption?

A: The Intel Arc Pro B60 has a TDP of 200 W and requires a 550 W power supply, while the NVIDIA GeForce RTX 4010 has a TDP of 50 W and only needs a 250 W PSU. The RTX 4010 also draws power solely from the PCIe slot with no additional power connectors, whereas the Arc Pro B60 requires a single 8-pin connector.

Q: What is the performance percentile of each card relative to all GPUs?

A: The Intel Arc Pro B60 sits in the 20th percentile of all GPUs, while the NVIDIA GeForce RTX 4010 is in the 18th percentile. This indicates that both cards are in the lower performance tier, but the Intel card is slightly higher-ranked overall.

Q: Which card has more shading units and texture mapping units?

A: The Intel Arc Pro B60 has 2560 shading units, 160 TMUs, and 80 ROPs. The NVIDIA GeForce RTX 4010 has 768 shading units, 24 TMUs, and 16 ROPs. The Intel card has over 3x the shading units and nearly 7x the TMUs.

Where Each One Wins

The NVIDIA GeForce RTX 4010 wins the only direct head-to-head test, which is a modern DX12 benchmark (3DMark Steel Nomad). This suggests it may be better suited for current-generation gaming or workloads that rely heavily on DX12 features, despite its modest 4 GB memory capacity. Its low 50 W TDP and single-slot design also make it a winner for low-power, space-constrained systems where cooling and power delivery are limited. The RTX 4010’s 24 tensor cores also provide AI acceleration capability that the Arc Pro B60 lacks entirely (the Intel card has no tensor core field listed).

The Intel Arc Pro B60 wins decisively in raw compute and memory capacity. With 24 GB of VRAM and 456.0 GB/s of bandwidth, it is clearly positioned for large datasets, high-resolution textures, or compute workloads that exceed 4 GB. Its PassMark GPU Compute score of 7029 indicates strong general-purpose compute performance, and its DirectX 9, 10, 11, and 12 scores (179, 61, 122, and 76 respectively) show broad API support. The Arc Pro B60 also wins on pixel rate (192.0 GPixel/s vs 28.19 GPixel/s) and texture rate (384.0 GTexel/s vs 42.29 GTexel/s), which are critical for fill-rate-bound tasks.

Specification Differences

The two cards differ on nearly every major specification. The Intel Arc Pro B60 uses a 5 nm process (TSMC) with 19,600 million transistors on a 272 mm² die, while the NVIDIA RTX 4010 uses an 8 nm process (Samsung) with 8,700 million transistors on a 200 mm² die. This gives Intel a transistor density of 72.1M/mm² versus NVIDIA’s 43.5M/mm². Clock speeds favor Intel: base 2000 MHz vs 1417 MHz, and boost 2400 MHz vs 1762 MHz. Memory is a major gap: 24 GB vs 4 GB, 192-bit vs 64-bit bus, and 456.0 GB/s vs 96.00 GB/s bandwidth. The Intel card also has significantly more shading units (2560 vs 768), TMUs (160 vs 24), ROPs (80 vs 16), and RT cores (20 vs 6). The RTX 4010 has 24 tensor cores, which the Intel card lacks. Power and physical specs differ: 200 W vs 50 W TDP, dual-slot vs single-slot, 1x 8-pin vs no connector, and 550 W vs 250 W suggested PSU. Bus interfaces also differ: PCIe 5.0 x8 vs PCIe 4.0 x8. Display outputs are both mini-DisplayPort, but Intel uses 2.1 and NVIDIA uses 1.4a.

Architecture Differences

The architectures represent two different design philosophies. Intel’s Arc Pro B60 is based on Xe2-HPG (Battlemage Pro Series), built on a 5 nm TSMC process with the BMG-G21 chip. It uses a 2:1 FP16 ratio, meaning FP16 is twice as fast as FP32, and it has dedicated RT cores but no tensor cores. NVIDIA’s RTX 4010 uses the Ampere architecture with the GA107 chip, built on Samsung’s 8 nm process. It has a 1:1 FP16 ratio, meaning FP16 and FP32 are the same speed, and it includes both RT cores and tensor cores. The Intel card’s larger die (272 mm² vs 200 mm²) and higher transistor count (19,600 million vs 8,700 million) reflect its more complex design, but the NVIDIA card achieves similar benchmark results with far fewer resources. The Arc Pro B60’s memory clock is 2375 MHz (19 Gbps effective), while the RTX 4010’s is 1500 MHz (12 Gbps effective). Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, so API support is identical.

The Verdict

The data suggests that the NVIDIA GeForce RTX 4010 is the better choice for users who prioritize a single modern DX12 benchmark and need extreme power efficiency. Its 50 W TDP, lack of power connectors, and single-slot design make it ideal for small form factor or low-power builds. It wins the Steel Nomad test by 8.5%, which may translate to better performance in similar DX12 gaming or professional workloads. The RTX 4010’s 24 tensor cores also give it an edge in AI-accelerated tasks, a feature the Intel card does not offer.

The Intel Arc Pro B60 is the clear winner for memory-intensive and compute-heavy workloads. Its 24 GB VRAM is six times larger than the RTX 4010’s 4 GB, and its 456.0 GB/s bandwidth is nearly five times higher. The Arc Pro B60 also has 4.5x the FP32 throughput and 9x the FP16 throughput, making it far more capable for rendering, simulation, or data processing that exceeds the RTX 4010’s memory limit. Its higher PassMark G3D score (14580) and GPU Compute score (7029) indicate strong overall performance, even if it loses the specific Steel Nomad test. For any workload that demands large textures, big datasets, or heavy compute, the Arc Pro B60 is the superior card. For low-power, DX12-focused tasks, the RTX 4010 wins.