Intel Arc B390 vs NVIDIA RTX 6000D Comparison

Head-to-Head Benchmarks

The recorded data contains a single shared benchmark between the Intel Arc B390 and the NVIDIA RTX 6000D: 3DMark Steel Nomad (DX12). The NVIDIA RTX 6000D wins decisively, scoring 3,522 points against 1,482 points for the Intel Arc B390. This represents a 57.9% deficit for the Intel part, meaning the RTX 6000D delivers more than double the performance in this test.

The RTX 6000D's result places it in the 98th percentile of all GPUs in the database, while the Arc B390 sits at the 9th percentile. The gap between these two parts is not merely incremental; it spans nearly the entire performance distribution. The RTX 6000D's nearest rivals in the database include the NVIDIA Tesla V100S PCIe 32 GB (0.8% slower), the NVIDIA A100 PCIe 80 GB (5.4% faster), the NVIDIA A100 SXM4 40 GB (4.7% slower), and the NVIDIA RTX 5000 Ada Generation (6.1% slower). The Arc B390, by contrast, competes with entry-level mobile parts: the NVIDIA GeForce GT 520MX (1.3% slower), the NVIDIA GeForce 800M (1.5% slower), the NVIDIA GeForce GT 625 OEM (2.5% slower), and the NVIDIA GeForce GT 710 (2.7% slower).

The RTX 6000D also posts a Geekbench OpenCL score of 388,405, a test not run on the Arc B390. When combining both benchmark results, the RTX 6000D's average benchmark score reaches 195,964, compared to the Arc B390's average of just 1,482. The RTX 6000D's average score is roughly 132 times higher, though this comparison mixes different test types and should be interpreted with caution. The head-to-head table records one win for the RTX 6000D and zero wins for the Arc B390.

FAQ

Q: How much faster is the NVIDIA RTX 6000D in 3DMark Steel Nomad?

A: The RTX 6000D scores 3,522 points, which is 2,040 points higher than the Arc B390's 1,482 points. The delta is 57.9% in favor of the RTX 6000D.

Q: What percentile rankings do these GPUs hold in the database?

A: The RTX 6000D sits at the 98th percentile of all GPUs, while the Arc B390 sits at the 9th percentile. This places the RTX 6000D among the top tier of recorded GPUs and the Arc B390 near the bottom.

Q: Which GPUs are closest to the Intel Arc B390 in performance?

A: The nearest rivals are the NVIDIA GeForce GT 520MX (1.3% faster), the NVIDIA GeForce 800M (1.5% faster), the NVIDIA GeForce GT 625 OEM (2.5% faster), and the NVIDIA GeForce GT 710 (2.7% faster). All of these are within a few percentage points of the Arc B390.

Q: Which GPUs are closest to the NVIDIA RTX 6000D in performance?

A: The nearest rivals are the NVIDIA Tesla V100S PCIe 32 GB (0.8% slower), the NVIDIA A100 SXM4 40 GB (4.7% slower), the NVIDIA A100 PCIe 80 GB (5.4% faster), and the NVIDIA RTX 5000 Ada Generation (6.1% slower).

Q: Does the Intel Arc B390 have any benchmark win over the RTX 6000D?

A: No. The head-to-head data records zero wins for the Arc B390 and one win for the RTX 6000D. The only shared test, 3DMark Steel Nomad, went to the RTX 6000D.

Q: What is the average benchmark score for each GPU?

A: The RTX 6000D has an average benchmark score of 195,964, driven largely by its Geekbench OpenCL result of 388,405. The Arc B390 has an average benchmark score of 1,482, derived solely from its 3DMark Steel Nomad score.

The Verdict

The data leaves no ambiguity. The NVIDIA RTX 6000D outperforms the Intel Arc B390 by a margin that spans the entire GPU performance spectrum. In the only direct comparison available, the RTX 6000D delivers 2.4 times the 3DMark Steel Nomad score of the Arc B390. The percentile rankings reinforce this: 98th versus 9th.

For any workload represented by these benchmarks, the RTX 6000D is the only choice. The Arc B390's performance class, as indicated by its nearest rivals (GeForce GT 710, GT 625 OEM, 800M, GT 520MX), is that of basic display output or very light graphical tasks. The RTX 6000D's nearest rivals (Tesla V100S, A100 variants, RTX 5000 Ada) are professional compute and workstation accelerators.

The RTX 6000D also brings a massive memory advantage with 84 GB of GDDR7 and 1.40 TB/s bandwidth, compared to the Arc B390's system-shared memory. The compute gap is equally stark: 97.04 TFLOPS FP32 versus 7.680 TFLOPS FP32, a 12.6-fold difference. The RTX 6000D's 19968 shading units dwarf the Arc B390's 1536. The RTX 6000D has 156 ray tracing cores and 624 tensor cores; the Arc B390 has 12 ray tracing cores and no tensor core field in the database.

Users selecting between these two parts are choosing between entirely different performance tiers. The Arc B390 suits integrated graphics scenarios where low power draw (80 W) and compact integration matter. The RTX 6000D, with its 600 W TDP, dual-slot design, and 1x 16-pin power connector, requires a workstation-class platform. The RTX 6000D's launch MSRP is 8,565 USD. The data records no launch MSRP for the Arc B390.

Specification Differences

The two GPUs differ in nearly every measurable specification. The Intel Arc B390 uses a 3 nm process node fabricated by Intel, while the NVIDIA RTX 6000D uses a 5 nm node from TSMC. The RTX 6000D contains 92,200 million transistors on a 750 mm² die, with a transistor density of 122.9M per mm². The Arc B390's transistor count and die size are recorded as unknown.

Clock speeds differ significantly. The Arc B390 has a base clock of 300 MHz and a boost clock of 2500 MHz. The RTX 6000D has a base clock of 1992 MHz and a boost clock of 2430 MHz. Memory configurations are completely different: the Arc B390 uses system-shared memory with system-dependent bandwidth, while the RTX 6000D has 84 GB of GDDR7 on a 448-bit bus, delivering 1.40 TB/s at 1560 MHz (25 Gbps effective).

Compute resources are heavily skewed toward the RTX 6000D. It has 19968 shading units, 624 texture mapping units, and 192 render output units. The Arc B390 has 1536 shading units, 48 TMUs, and 24 ROPs. The RTX 6000D's pixel rate is 466.6 GPixel/s versus 60.00 GPixel/s for the Arc B390. Texture rate is 1,516.3 GTexel/s versus 120.0 GTexel/s. FP32 throughput is 97.04 TFLOPS versus 7.680 TFLOPS. FP16 throughput is 97.04 TFLOPS (1:1) versus 15.36 TFLOPS (2:1).

Power and physical specifications diverge sharply. The Arc B390 has an 80 W TDP, is an integrated graphics processor (IGP) with no power connectors, and is not a slot card. The RTX 6000D has a 600 W TDP, occupies a dual-slot form factor, uses a single 16-pin power connector, requires a 1000 W suggested PSU, and connects via PCIe 5.0 x16. The RTX 6000D measures 304 mm by 137 mm by 40 mm. Display outputs are 4x DisplayPort 2.1b on the RTX 6000D, while the Arc B390's outputs are portable device dependent.

Architecture Differences

The Intel Arc B390 is built on the Xe3-LPG architecture, using the Panther Lake chip, and belongs to the Arc Graphics-M (Panther Lake) generation. The NVIDIA RTX 6000D uses the Blackwell 2.0 architecture with the GB202 chip, part of the Blackwell PRO W (x000) generation.

The Arc B390's architecture integrates ray tracing hardware with 12 RT cores but has no tensor core field recorded. The RTX 6000D includes 156 RT cores and 624 tensor cores, indicating substantial acceleration for ray-traced workloads and AI compute. The RTX 6000D's FP16 performance matches its FP32 at 97.04 TFLOPS (1:1 ratio), suggesting full-rate half-precision compute. The Arc B390's FP16 is 15.36 TFLOPS at a 2:1 ratio, meaning half the rate of FP32.

Both GPUs support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The RTX 6000D's successor field is empty, while its predecessor is listed as Workstation Ada. The Arc B390 lists no predecessor or successor. Production status for both is Active. The RTX 6000D was released on 2025-07-13, while the Arc B390's release date is 2026-01-26.

The RTX 6000D's memory architecture uses dedicated GDDR7 with a 448-bit bus, whereas the Arc B390 relies on system-shared memory. This architectural choice fundamentally affects bandwidth: the RTX 6000D delivers 1.40 TB/s, while the Arc B390's bandwidth is system dependent. The RTX 6000D's 84 GB capacity also far exceeds any shared-memory configuration typically available.

Where Each One Wins

The RTX 6000D wins in every benchmark category recorded in the database. In 3DMark Steel Nomad, it is 57.9% ahead. Its Geekbench OpenCL score of 388,405 represents a compute capability that the Arc B390 does not approach, given the latter's 7.680 TFLOPS FP32 ceiling.

The RTX 6000D is the clear choice for high-end 3D rendering, AI inference, and professional compute workloads. Its 97.04 TFLOPS FP32, 97.04 TFLOPS FP16, 156 RT cores, and 624 tensor cores position it for demanding tasks. The 84 GB memory capacity with 1.40 TB/s bandwidth supports large datasets and high-resolution textures. The 98th percentile ranking confirms its top-tier status.

The Arc B390 wins in power efficiency and integration, though the database does not provide efficiency benchmarks. Its 80 W TDP versus 600 W represents a 7.5-fold difference. The IGP form factor means no separate card, no power connectors, and no additional PSU requirement. It suits ultra-portable devices where space and thermal limits are tight. Its 9th percentile ranking places it alongside basic office GPUs like the GeForce GT 710, which indicates suitability for standard desktop graphics, video playback, and light 2D workloads.

For ray tracing, the RTX 6000D's 156 RT cores versus 12 on the Arc B390 means the NVIDIA part handles complex light transport far more effectively. For AI tasks, the RTX 6000D's 624 tensor cores (with no equivalent on the Arc B390) provide hardware acceleration that the Intel part lacks entirely.

The Arc B390's only recorded win is in power consumption and physical footprint. At 80 W with no external power connector and IGP slot width, it can be deployed in systems where a 600 W dual-slot card cannot fit. The RTX 6000D requires a 1000 W PSU and a PCIe 5.0 x16 slot, plus space for its 304 mm length.

In summary, the RTX 6000D wins on raw performance, memory, compute features, and professional capability. The Arc B390 wins on power draw, integration simplicity, and portability. No benchmark in the database favors the Intel part.