Intel Arc B390 vs NVIDIA GeForce RTX 5070 Mobile 12 GB Comparison

FAQ

Q: How does the Intel Arc B390 compare to the NVIDIA GeForce RTX 5070 Mobile 12 GB in raw graphics performance?

A: The recorded data shows the Intel Arc B390 scores 1482 in the 3DMark Steel Nomad DX12 test, placing it at the 9th percentile of all GPUs. The RTX 5070 Mobile has no recorded benchmark scores in the database, so direct comparison is limited to architectural specifications rather than measured results.

Q: Which GPU has a higher boost clock speed?

A: The Intel Arc B390 has a boost clock of 2500 MHz, while the NVIDIA GeForce RTX 5070 Mobile has a boost clock of 1425 MHz. The Arc B390's boost clock is considerably higher, though the two chips serve different power and design targets.

Q: What memory configurations do these two GPUs use?

A: The Intel Arc B390 uses system shared memory, with memory type, bus width, and bandwidth all listed as system dependent. The NVIDIA GeForce RTX 5070 Mobile uses 12 GB of GDDR7 memory on a 192-bit bus, delivering 576.0 GB/s of bandwidth.

Q: How do the shading unit counts differ between the two?

A: The Intel Arc B390 contains 1536 shading units, 48 texture mapping units, and 24 raster operation units. The NVIDIA GeForce RTX 5070 Mobile contains 4608 shading units, 144 TMUs, and 48 ROPs, making it substantially larger in each of these execution resources.

Q: Which GPU has a higher thermal design power?

A: The Intel Arc B390 has a TDP of 80 W, while the NVIDIA GeForce RTX 5070 Mobile has a TDP of 50 W. Despite the RTX 5070 Mobile offering more shading units and dedicated memory, its declared power envelope is lower.

Q: What process nodes are used by each chip?

A: The Intel Arc B390 uses Intel's 3 nm process, while the NVIDIA GeForce RTX 5070 Mobile uses TSMC's 5 nm process. The NVIDIA chip is built on 21,900 million transistors on a 181 mm² die, whereas the Intel chip's transistor count and die size are listed as unknown.

Architecture Differences

The Intel Arc B390 is built on the Panther Lake chip using the Xe3-LPG architecture, part of the Arc Graphics-M (Panther Lake) generation. It is manufactured on Intel's 3 nm process. The transistor count and die size are unknown in the database. The NVIDIA GeForce RTX 5070 Mobile belongs to the GeForce 50-series and uses the GB206 chip with the Blackwell 2.0 architecture. It is manufactured on TSMC's 5 nm process, with 21,900 million transistors on a 181 mm² die, giving a transistor density of 121.0M per mm².

These are fundamentally different design philosophies. The Arc B390 is an integrated graphics processor, indicated by its IGP slot width, system shared memory, and no power connectors. The RTX 5070 Mobile also uses an IGP slot width and no power connectors, but it carries dedicated 12 GB of GDDR7 memory and connects via PCIe 5.0 x16. The memory architecture alone separates the two into different product categories: one relies on system memory, the other on high-speed dedicated VRAM.

The execution resource counts diverge sharply. The Arc B390 has 1536 shading units, 48 TMUs, 24 ROPs, and 12 RT cores. The RTX 5070 Mobile has 4608 shading units, 144 TMUs, 48 ROPs, and 36 RT cores. The NVIDIA chip also includes 144 tensor cores, while the Intel chip has no tensor core count listed. In terms of raw throughput, the Arc B390 delivers 7.680 TFLOPS FP32 and 15.36 TFLOPS FP16 (2:1 ratio). The RTX 5070 Mobile delivers 13.13 TFLOPS FP32 and 13.13 TFLOPS FP16 (1:1 ratio), meaning the NVIDIA chip holds a roughly 71% advantage in FP32 compute, while the Intel chip's FP16 figure is higher on paper due to its 2:1 rate.

Clock behavior also differs. The Arc B390 idles at a 300 MHz base clock and boosts to 2500 MHz. The RTX 5070 Mobile has a 907 MHz base clock and a 1425 MHz boost. The Intel part runs at much higher clock speeds, but the NVIDIA part compensates with far more execution units. Pixel rate is close: 60.00 GPixel/s for the Arc B390 versus 68.40 GPixel/s for the RTX 5070 Mobile. Texture rate is not close: 120.0 GTexel/s versus 205.2 GTexel/s, a 71% advantage for NVIDIA.

Both GPUs support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Both are marked as active production. The Arc B390 was released on January 26, 2026, while the RTX 5070 Mobile was released on May 31, 2026. The RTX 5070 Mobile's predecessor is listed as the GeForce 40 Mobile.

Head-to-Head Benchmarks

The database contains no head-to-head benchmark entries between the Intel Arc B390 and the NVIDIA GeForce RTX 5070 Mobile. The wins counter shows 0 for each GPU. The only benchmark recorded for either part is the Arc B390's 3DMark Steel Nomad DX12 score of 1482, which places it at the 9th percentile among all GPUs. The RTX 5070 Mobile has an average benchmark score of 0 and no recorded benchmark entries.

The Arc B390's nearest rivals in the database are all older NVIDIA discrete GPUs. It sits 1.3% ahead of the GeForce GT 520MX (average score 1463), 1.5% ahead of the GeForce 800M (1460), 2.5% ahead of the GeForce GT 625 OEM (1446), and 2.7% ahead of the GeForce GT 710 (1443). These margins are small, indicating the Arc B390 lands in a performance tier far below modern discrete graphics. The RTX 5070 Mobile's percentile rank of 50, by contrast, places it at the midpoint of all GPUs, though no measured score backs that percentile in the current data.

Without a common benchmark result, the head-to-head analysis must rely on specification-derived throughput estimates. The RTX 5070 Mobile's FP32 output of 13.13 TFLOPS is roughly 71% higher than the Arc B390's 7.680 TFLOPS. Its texture rate of 205.2 GTexel/s is likewise 71% higher. Its pixel rate advantage is smaller, at 68.40 GPixel/s versus 60.00 GPixel/s, about 14% higher. Memory bandwidth cannot be compared numerically because the Arc B390's bandwidth is listed as system dependent, while the RTX 5070 Mobile's is fixed at 576.0 GB/s.

The Arc B390's strengths in this comparison are clock speed and FP16 throughput. Its 2500 MHz boost is 75% higher than the RTX 5070 Mobile's 1425 MHz boost. Its 15.36 TFLOPS FP16 figure exceeds the RTX 5070 Mobile's 13.13 TFLOPS, though the Intel number is achieved through a 2:1 rate conversion while NVIDIA runs a 1:1 rate. The Arc B390 also consumes more power by specification, with an 80 W TDP versus 50 W, so its higher clocks come at a higher declared power cost.

Specification Differences

The two GPUs differ across nearly every recorded specification field. The Arc B390 uses the Panther Lake chip with Xe3-LPG architecture, while the RTX 5070 Mobile uses the GB206 chip with Blackwell 2.0. Process nodes differ: 3 nm for Intel, 5 nm for TSMC. Transistor and die data exist only for NVIDIA: 21,900 million transistors, 181 mm² die, 121.0M per mm² density. Intel's figures are unknown.

Clock speeds differ substantially. The Arc B390 runs at 300 MHz base and 2500 MHz boost. The RTX 5070 Mobile runs at 907 MHz base and 1425 MHz boost. Memory configurations are entirely different: the Arc B390 uses system shared memory with system dependent bandwidth, while the RTX 5070 Mobile has 12 GB of GDDR7 on a 192-bit bus with 576.0 GB/s bandwidth.

Execution resources differ by a factor of three in most categories. The Arc B390 has 1536 shading units, 48 TMUs, 24 ROPs, and 12 RT cores. The RTX 5070 Mobile has 4608 shading units, 144 TMUs, 48 ROPs, 36 RT cores, and 144 tensor cores. The Intel part lists no tensor cores.

Throughput figures reflect these resource differences. Pixel rate: 60.00 GPixel/s versus 68.40 GPixel/s. Texture rate: 120.0 GTexel/s versus 205.2 GTexel/s. FP32: 7.680 TFLOPS versus 13.13 TFLOPS. FP16: 15.36 TFLOPS (2:1) versus 13.13 TFLOPS (1:1). TDP differs as well: 80 W for Intel, 50 W for NVIDIA.

Bus interfaces differ: the Arc B390 uses IGP, while the RTX 5070 Mobile uses PCIe 5.0 x16. Both have IGP slot width, no power connectors, and portable-device-dependent display outputs. Release dates differ by months: January 26, 2026 for the Arc B390, May 31, 2026 for the RTX 5070 Mobile. The RTX 5070 Mobile has a listed predecessor, the GeForce 40 Mobile, while the Arc B390 has none. Neither part has a launch MSRP in the database.

Where Each One Wins

The Intel Arc B390 wins on clock speed. Its 2500 MHz boost is far higher than the RTX 5070 Mobile's 1425 MHz, and its 300 MHz base is lower. The higher boost suggests the Intel part can ramp aggressively when power allows, but its 80 W TDP and system shared memory limit sustained performance in memory-bound workloads.

The Arc B390 also shows a higher FP16 throughput figure at 15.36 TFLOPS, though this comes from a 2:1 rate conversion. Applications that can exploit packed FP16 math may see an advantage on paper, but the NVIDIA part's 1:1 FP16 rate of 13.13 TFLOPS means it does not lose precision when switching from FP32.

The NVIDIA GeForce RTX 5070 Mobile wins on nearly every other measurable specification. It has three times the shading units, TMUs, and RT cores, plus 144 tensor cores where the Intel part has none. Its FP32 compute is 71% higher. Its texture rate is 71% higher. Its pixel rate is 14% higher. It has dedicated 12 GB of GDDR7 memory with 576.0 GB/s of bandwidth, eliminating the system memory dependency that constrains the Arc B390.

The RTX 5070 Mobile also wins on efficiency, at least by declared TDP. It delivers 13.13 TFLOPS FP32 within a 50 W envelope, while the Arc B390 delivers 7.680 TFLOPS within 80 W. The NVIDIA part achieves more than 70% higher FP32 throughput while drawing 37.5% less power by specification.

The Arc B390's only clear territory is its higher clock and its integrated nature. As an IGP with system shared memory, it occupies a different market position than a discrete-class mobile GPU with dedicated VRAM. The RTX 5070 Mobile's PCIe 5.0 x16 interface and 12 GB VRAM place it in a performance class the Arc B390's 9th percentile benchmark score cannot approach.

The Verdict

The recorded data points to two GPUs in different classes. The Intel Arc B390, with its 1482 Steel Nomad score and 9th percentile ranking, sits near the bottom of the database's GPU distribution. Its nearest rivals are legacy NVIDIA parts like the GT 710 and GT 520MX, all within 2.7% of its score. The RTX 5070 Mobile, at the 50th percentile, is positioned in the middle of the distribution, though it lacks a recorded benchmark score to confirm that placement.

For a system that relies on integrated graphics with shared system memory, the Arc B390 offers high boost clocks and a modern feature set: DirectX 12 Ultimate, Vulkan 1.4, and 12 RT cores. It draws 80 W and has no power connectors, making it a plausible IGP solution for Panther Lake-based portable devices. Its FP16 throughput of 15.36 TFLOPS is not matched by the NVIDIA part's 13.13 TFLOPS figure, but the Intel number relies on a 2:1 rate that NVIDIA does not use.

For workloads that need dedicated memory bandwidth, tensor cores, or high FP32 throughput, the RTX 5070 Mobile is the clear choice from the available data. Its 576.0 GB/s of GDDR7 bandwidth, 144 tensor cores, 36 RT cores, and 13.13 TFLOPS FP32 output are all absent from the Arc B390's specification sheet. Its 50 W TDP also makes it the lower-power option despite the larger execution resource pool.

The user who should pick the Arc B390 is one constrained to integrated graphics with no discrete memory path. The user who should pick the RTX 5070 Mobile is one who needs the higher compute, dedicated VRAM, and tensor core support. The benchmark gap, while not directly measured in a head-to-head test, is implied by the specification differences: the RTX 5070 Mobile delivers roughly 71% more FP32 throughput and texture rate, more than double the pixel rate gap, and a memory subsystem that is not system dependent. The data does not support a scenario where the Arc B390 outperforms the RTX 5070 Mobile in any memory-bound or FP32-heavy workload.