Intel Arc Pro B65 vs NVIDIA GeForce RTX 4090 Mobile Comparison

Head-to-Head Benchmarks

The recorded data shows a stark contrast between these two GPUs, with the NVIDIA GeForce RTX 4090 Mobile dominating the benchmark field. Out of the available measurements, the RTX 4090 Mobile holds a 9-0 advantage in wins over the Intel Arc Pro B65, which has no recorded benchmark scores in the database. The Intel Arc Pro B65 sits at the 50th percentile among all GPUs, while the RTX 4090 Mobile sits at the 84th percentile, a substantial gap that reflects its position well above the median.

Looking at the RTX 4090 Mobile’s actual scores, its Geekbench OpenCL result of 180,831 and Vulkan result of 170,774 indicate strong compute and graphics performance across different API environments. In Passmark testing, the GPU delivers a G3D score of 27,212, a compute score of 12,347, and DirectX 11 score of 262, with DirectX 9 at 310 and DirectX 12 at 107. The G2D score of 984 shows 2D throughput that is comparatively modest against its 3D capabilities.

For the Intel Arc Pro B65, the database contains no benchmark entries, so direct comparison of measured performance is impossible. However, the hardware specifications point to a different class of device. The Arc Pro B65 delivers 12.29 TFLOPS of FP32 compute, while the RTX 4090 Mobile delivers 32.98 TFLOPS, a difference of roughly 2.7 times. In rasterization throughput, the Arc Pro B65 reaches 384.0 GTexel/s, while the RTX 4090 Mobile reaches 515.3 GTexel/s, an advantage of about 34%. Pixel throughput is nearly identical: 192.0 GPixel/s for Intel and 189.8 GPixel/s for NVIDIA, so the two are effectively tied in that metric despite their other differences.

The RTX 4090 Mobile also leads in shading resources with 9,728 shading units versus 2,560, and 304 TMUs versus 160, while the Arc Pro B65 has 80 ROPs against NVIDIA’s 112. Ray tracing hardware differs as well: the Intel GPU has 20 RT cores, while the NVIDIA GPU has 76 RT cores, a 3.8 times advantage. Tensor core counts similarly favor NVIDIA at 304 versus none listed for Intel, which affects AI-accelerated workloads.

Memory capacity favors the Intel card at 32 GB versus 16 GB, and its bandwidth of 608.0 GB/s edges out the RTX 4090 Mobile’s 576.0 GB/s, a 5.5% advantage. Both use GDDR6 memory on a 256-bit bus, but the Intel memory clock runs at 2375 MHz (19 Gbps effective) versus 2250 MHz (18 Gbps effective), which explains the bandwidth gap. The Intel card also uses a newer PCIe interface at 5.0 x16, while the NVIDIA part uses PCIe 4.0 x16.

Nearest rival data for the RTX 4090 Mobile places it in a tight cluster. Its average benchmark score of 43,667 is 0.8% above the NVIDIA Quadro M6000 (43,301) and 0.9% above the GeForce RTX 5050 Mobile (43,268). It also sits 0.9% above the Quadro M6000 24 GB (43,262), while trailing the RTX A6000 (44,075) by 0.9%. This grouping indicates that the RTX 4090 Mobile performs within a narrow band of professional and mobile GPUs, rather than being an outlier at the top of the range.

FAQ

Q: How much faster is the RTX 4090 Mobile in FP32 compute?

A: The RTX 4090 Mobile delivers 32.98 TFLOPS of FP32 performance, while the Intel Arc Pro B65 delivers 12.29 TFLOPS. This represents a 2.68 times advantage for the NVIDIA part based on the recorded specification data.

Q: Which GPU has more memory and bandwidth?

A: The Intel Arc Pro B65 has 32 GB of GDDR6 memory on a 256-bit bus with 608.0 GB/s of bandwidth. The RTX 4090 Mobile has 16 GB of GDDR6 on the same 256-bit bus with 576.0 GB/s. Intel leads in capacity by 16 GB and in bandwidth by 32 GB/s.

Q: What are the nearest rivals to the RTX 4090 Mobile in the database?

A: The nearest rivals by average benchmark score are the NVIDIA Quadro M6000 (43,301, 0.8% lower), the GeForce RTX 5050 Mobile (43,268, 0.9% lower), the Quadro M6000 24 GB (43,262, 0.9% lower), and the RTX A6000 (44,075, 0.9% higher).

Q: How do the two GPUs compare in texture and pixel throughput?

A: The RTX 4090 Mobile reaches 515.3 GTexel/s versus 384.0 GTexel/s for the Arc Pro B65, a 34% difference in texture rate. Pixel rates are nearly equal at 189.8 GPixel/s for NVIDIA and 192.0 GPixel/s for Intel, so the Intel part is slightly ahead by 1.2%.

Q: What ray tracing resources does each GPU have?

A: The Intel Arc Pro B65 includes 20 RT cores, while the RTX 4090 Mobile includes 76 RT cores, a 3.8 times advantage for NVIDIA. The NVIDIA part also has 304 tensor cores, while the Intel card lists none.

Q: How does the RTX 4090 Mobile compare to its predecessor?

A: The RTX 4090 Mobile belongs to the GeForce 40 Mobile generation and lists the GeForce 30 Mobile as its predecessor, with the GeForce 50 Mobile as its successor. No benchmark scores are recorded for either adjacent generation in the database.

The Verdict

The data clearly separates these two products into different performance tiers. The RTX 4090 Mobile, with its 84th percentile ranking and an average benchmark score of 43,667, represents a high-end mobile GPU. Its nearest rivals are all within a 0.9% margin, confirming that it performs at a consistent level among top-tier parts. The Intel Arc Pro B65, ranked at the 50th percentile with no recorded benchmark scores, cannot be positioned relative to the RTX 4090 Mobile from measured results alone.

For raw compute throughput, the RTX 4090 Mobile offers 32.98 TFLOPS of FP32 and 32.98 TFLOPS of FP16, both well ahead of the Arc Pro B65’s 12.29 TFLOPS and 24.58 TFLOPS respectively. The NVIDIA GPU also holds advantages in shading units (9,728 versus 2,560), texture units (304 versus 160), ROPs (112 versus 80), RT cores (76 versus 20), and tensor cores (304 versus none). These specification gaps align with its higher percentile ranking.

The Intel Arc Pro B65 counters with 32 GB of memory, double the RTX 4090 Mobile’s 16 GB, and slightly higher bandwidth at 608.0 GB/s versus 576.0 GB/s. It also uses a PCIe 5.0 x16 interface versus PCIe 4.0 x16, and its pixel rate of 192.0 GPixel/s is marginally ahead of 189.8 GPixel/s. These advantages are real but narrow, and they do not compensate for the large compute deficit.

The RTX 4090 Mobile’s TDP of 120 W is lower than the Arc Pro B65’s 200 W, which is notable for a mobile part. The Intel card requires a dual-slot cooler and a 1x 8-pin power connector, while the NVIDIA part is an integrated GPU (IGP) with no power connectors and a suggested PSU not listed. Users with power or space constraints would find the NVIDIA part easier to accommodate.

Neither GPU has a launch MSRP listed in the database, so no pricing analysis is possible. The production status for both is Active, with the Intel part releasing on 2026-03-31 and the NVIDIA part on 2023-01-02.

The RTX 4090 Mobile is the choice for performance-focused workloads, as its benchmark scores, percentile rank, and compute specifications all point in that direction. The Arc Pro B65 is the choice for memory-heavy tasks that require 32 GB of VRAM, and its 608.0 GB/s bandwidth is competitive. For ray tracing, AI features, or general 3D rendering, the RTX 4090 Mobile’s 76 RT cores and 304 tensor cores provide hardware resources the Intel card cannot match.

Specification Differences

The two GPUs differ across nearly every measurable specification. The RTX 4090 Mobile uses the AD103 chip with Ada Lovelace architecture, while the Arc Pro B65 uses the BMG-G21 chip with Xe2-HPG architecture. Both are built on a 5 nm process at TSMC, but the NVIDIA chip integrates 45,900 million transistors on a 379 mm² die, while the Intel chip integrates 19,600 million on a 272 mm² die. Transistor density is 121.1M per mm² for NVIDIA and 72.1M per mm² for Intel.

Clock speeds differ substantially. The Arc Pro B65 runs at a fixed 2400 MHz base and boost, while the RTX 4090 Mobile runs at 1335 MHz base and 1695 MHz boost. Memory clocks are 2375 MHz (19 Gbps effective) for Intel and 2250 MHz (18 Gbps effective) for NVIDIA.

Memory capacity is 32 GB for Intel and 16 GB for NVIDIA. Bandwidth is 608.0 GB/s versus 576.0 GB/s. Both use GDDR6 on a 256-bit bus.

The Intel card has 2,560 shading units, 160 TMUs, 80 ROPs, and 20 RT cores. The NVIDIA card has 9,728 shading units, 304 TMUs, 112 ROPs, and 76 RT cores. NVIDIA also lists 304 tensor cores, while Intel lists none. FP32 is 12.29 TFLOPS for Intel and 32.98 TFLOPS for NVIDIA. FP16 is 24.58 TFLOPS for Intel (2:1 ratio) and 32.98 TFLOPS for NVIDIA (1:1 ratio).

Power and physical specifications differ as well. The Arc Pro B65 has a TDP of 200 W, is dual-slot, and uses a 1x 8-pin power connector with a 550 W suggested PSU. The RTX 4090 Mobile has a TDP of 120 W, is an IGP with no power connectors, and has no suggested PSU listed. The Intel card uses PCIe 5.0 x16, while the NVIDIA card uses PCIe 4.0 x16. Display outputs are 4x DisplayPort 2.1 for Intel, while NVIDIA lists Portable Device Dependent.

Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Architecture Differences

The Intel Arc Pro B65 is built on the Xe2-HPG architecture from the Battlemage Pro Series generation, using the BMG-G21 chip. The NVIDIA GeForce RTX 4090 Mobile uses the Ada Lovelace architecture from the GeForce 40 Mobile generation, based on the AD103 chip. Both are fabricated by TSMC at 5 nm, but the transistor counts and die sizes indicate different design philosophies: NVIDIA packs 45,900 million transistors into 379 mm², while Intel fits 19,600 million into 272 mm².

The RTX 4090 Mobile’s Ada Lovelace architecture includes dedicated tensor cores, 304 of them, which support AI-accelerated workloads. The Arc Pro B65 lists no tensor cores, meaning it lacks this dedicated hardware. Ray tracing hardware is also asymmetric: 76 RT cores on the NVIDIA part versus 20 on the Intel part. This suggests the NVIDIA architecture is designed for heavier ray-traced scenes and AI inference tasks.

FP16 throughput differs in ratio. The Intel GPU achieves 24.58 TFLOPS at a 2:1 ratio relative to FP32, meaning it uses a packed execution path. The NVIDIA GPU achieves 32.98 TFLOPS at a 1:1 ratio, meaning FP16 runs at the same rate as FP32. This indicates a different approach to mixed-precision execution, with NVIDIA maintaining full throughput across both formats.

The Intel architecture uses PCIe 5.0 x16 connectivity, doubling the interface bandwidth available to the NVIDIA PCIe 4.0 x16 link. Display output also differs architecturally: Intel provides 4x DisplayPort 2.1, while the NVIDIA part’s outputs are dependent on the portable device it is integrated into.

Power delivery architecture is another point of divergence. The Intel card is a discrete dual-slot solution with a 1x 8-pin connector and a 550 W suggested PSU, while the NVIDIA part is an integrated GPU without discrete power connectors, drawing at most 120 W. This reflects the intended use cases: the Intel card for desktop workstations, the NVIDIA part for laptops and portable systems.