Intel Arc B390 vs NVIDIA GeForce RTX 5060 GB205 Comparison

FAQ

Q: How do the two GPUs compare in raw compute performance?

A: The NVIDIA GeForce RTX 5060 GB205 delivers 19.18 TFLOPS of FP32 throughput, which is 2.5 times the 7.680 TFLOPS of the Intel Arc B390. The RTX 5060 also achieves 19.18 TFLOPS in FP16 with a 1:1 ratio, while the Arc B390 reaches 15.36 TFLOPS in FP16 with a 2:1 ratio.

Q: What are the memory configurations of each GPU?

A: The RTX 5060 GB205 uses 8 GB of GDDR7 memory on a 128-bit bus, providing 448.0 GB/s of bandwidth. The Arc B390 uses system shared memory, with bandwidth described as system dependent, meaning its performance relies entirely on the host platform's memory subsystem.

Q: Which GPU has a higher pixel and texture throughput?

A: The RTX 5060 GB205 leads with 119.9 GPixel/s and 299.6 GTexel/s. The Arc B390 records 60.00 GPixel/s and 120.0 GTexel/s, putting NVIDIA at roughly double the pixel rate and 2.5 times the texture rate.

Q: What is the difference in power consumption?

A: The Arc B390 has a TDP of 80 W, while the RTX 5060 GB205 is rated at 145 W. The RTX 5060 requires a 1x 8-pin power connector and a 300 W suggested PSU, whereas the Arc B390 uses no power connectors as an integrated graphics processor (IGP).

Q: How do the architectures differ in terms of manufacturing process?

A: Intel fabricates the Arc B390 on a 3 nm process at its own foundry, while NVIDIA uses TSMC's 5 nm process for the RTX 5060 GB205. The RTX 5060 has 31,100 million transistors on a 263 mm² die, whereas the Arc B390's transistor count and die size are not recorded.

Q: What benchmark data exists for each GPU?

A: The Arc B390 has a recorded 3DMark Steel Nomad DX12 score of 1482, placing it at the 9th percentile among all GPUs. The RTX 5060 GB205 has no recorded benchmark scores in the database, with an average score of 0 and a percentile of 50.

Architecture Differences

The Intel Arc B390 and NVIDIA GeForce RTX 5060 GB205 represent fundamentally different design philosophies. The Arc B390 is built on Intel's Xe3-LPG architecture, part of the Arc Graphics-M (Panther Lake) generation, and functions as an integrated graphics processor on a 3 nm node at Intel's foundry. Its 1536 shading units, 48 texture mapping units, and 24 raster operation units are paired with 12 ray tracing cores. The GPU operates with a base clock of 300 MHz and a boost clock of 2500 MHz, drawing 80 W of power. Memory is system shared, meaning the GPU borrows from the host's main memory rather than having dedicated VRAM.

The RTX 5060 GB205 uses NVIDIA's Blackwell 2.0 architecture, belongs to the GeForce 50-series, and is a discrete dual-slot card fabricated on TSMC's 5 nm process. It contains 31,100 million transistors on a 263 mm² die, yielding a transistor density of 118.3M per mm². The GPU packs 3840 shading units, 120 TMUs, 48 ROPs, 30 RT cores, and 120 tensor cores. Clocks run at 2280 MHz base and 2497 MHz boost, with memory clocked at 1750 MHz (28 Gbps effective). The card connects via PCIe 5.0 x8 and requires one 8-pin power connector.

The architectural split is stark: the Arc B390 is an integrated solution with no dedicated memory bus, while the RTX 5060 GB205 is a fully discrete card with 8 GB of GDDR7 on a 128-bit interface delivering 448.0 GB/s. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The RTX 5060 adds tensor cores for AI workloads, a feature absent from the Arc B390's specifications. The RTX 5060's predecessor is the GeForce 40 series, and its successor is the GeForce 60 series. Release dates differ by roughly four months, with the Arc B390 launching on 2026-01-26 and the RTX 5060 on 2026-05-31, though the launch MSRP for the RTX 5060 is 299 USD while no price is recorded for the Arc B390.

Head-to-Head Benchmarks

The database contains only one benchmark entry for the Arc B390 and none for the RTX 5060 GB205, which limits direct comparison. The Arc B390 scores 1482 in 3DMark Steel Nomad DX12, placing it at the 9th percentile among all GPUs. Its nearest rivals are all low-end NVIDIA parts: the GeForce GT 520MX (1463, 1.3% behind), GeForce 800M (1460, 1.5% behind), GeForce GT 625 OEM (1446, 2.5% behind), and GeForce GT 710 (1443, 2.7% behind). This clustering shows the Arc B390 performs just above these legacy entry-level GPUs, with a maximum delta of 2.7% over the GT 710.

Since the RTX 5060 GB205 has no recorded benchmark scores, the head-to-head must rely on specification-derived metrics. The RTX 5060's FP32 throughput of 19.18 TFLOPS is exactly 2.5 times the Arc B390's 7.680 TFLOPS. Texture rate favors NVIDIA at 299.6 GTexel/s versus 120.0 GTexel/s, a 2.5x advantage. Pixel rate shows a 2.0x gap, with the RTX 5060 at 119.9 GPixel/s against the Arc B390's 60.00 GPixel/s. The RTX 5060's 3840 shading units double the Arc B390's 1536, and its 120 TMUs are 2.5 times the Arc B390's 48. ROPs also double, 48 versus 24.

RT core counts tell a similar story: the RTX 5060 has 30 RT cores, while the Arc B390 has 12, a 2.5x difference. Memory bandwidth is where the gap becomes extreme. The RTX 5060 provides 448.0 GB/s through dedicated GDDR7, while the Arc B390's bandwidth is system dependent, with no fixed figure available. The RTX 5060's 8 GB capacity is fixed, whereas the Arc B390 shares system memory of variable size. The Arc B390's boost clock of 2500 MHz is nearly identical to the RTX 5060's 2497 MHz, but the base clocks differ sharply: 300 MHz versus 2280 MHz.

The RTX 5060's FP16 performance matches its FP32 at 19.18 TFLOPS (1:1), while the Arc B390's FP16 of 15.36 TFLOPS is achieved at a 2:1 ratio, meaning half the FP32 rate. This indicates the RTX 5060 has full-rate FP16 capability, whereas the Arc B390 halves its throughput for half-precision work. The RTX 5060's tensor cores, 120 of them, are not present in the Arc B390's specification sheet.

The Verdict

The data indicates a decisive performance advantage for the NVIDIA GeForce RTX 5060 GB205 across every measurable compute dimension. The RTX 5060 delivers 2.5 times the FP32 throughput, 2.5 times the texture rate, and double the pixel rate of the Intel Arc B390. Its dedicated 8 GB GDDR7 memory with 448.0 GB/s bandwidth removes the dependency on system memory that constrains the Arc B390. The RTX 5060 also includes 120 tensor cores and 30 RT cores, features that are either absent or reduced in the Arc B390.

The Arc B390's recorded benchmark score of 1482 in 3DMark Steel Nomad places it at the 9th percentile, surrounded by NVIDIA's GT 520MX, 800M, GT 625 OEM, and GT 710, all legacy parts. The RTX 5060's 50th percentile ranking, while based on no direct scores, reflects its position as a mid-range discrete GPU. The RTX 5060's launch MSRP is 299 USD, and its 145 W TDP is 65 W higher than the Arc B390's 80 W, a trade-off for the substantial performance gain.

For users constrained to integrated graphics, the Arc B390 fits a low-power, no-expansion-slot scenario with a 300 MHz base clock and system shared memory. For anyone requiring discrete performance, the RTX 5060 GB205 is the clear choice, offering more than double the shading units, TMUs, and ROPs, plus dedicated memory bandwidth. The production status of both is active, but the RTX 5060's specification sheet is complete, while the Arc B390 lacks transistor count, die size, and fixed memory figures.

The RTX 5060's 1:1 FP16 ratio and tensor cores make it suitable for mixed-precision workloads, while the Arc B390's 2:1 FP16 ratio limits half-precision efficiency. The RTX 5060's PCIe 5.0 x8 interface and dual-slot form factor with 1x 8-pin power indicate a conventional discrete GPU installation, whereas the Arc B390's IGP bus interface and no power connectors reflect its integrated nature. The RTX 5060's display outputs include 1x HDMI 2.1b and 3x DisplayPort 2.1b, while the Arc B390's outputs are portable device dependent.

Specification Differences

| Specification | Intel Arc B390 | NVIDIA GeForce RTX 5060 GB205 |

|---|---|---|

| Architecture | Xe3-LPG | Blackwell 2.0 |

| Process Node | 3 nm | 5 nm |

| Foundry | Intel | TSMC |

| Transistors | unknown | 31,100 million |

| Die Size | unknown | 263 mm² |

| Transistor Density | null | 118.3M / mm² |

| Base Clock | 300 MHz | 2280 MHz |

| Boost Clock | 2500 MHz | 2497 MHz |

| Memory Size | System Shared | 8 GB |

| Memory Type | System Shared | GDDR7 |

| Memory Bus Width | System Shared | 128 bit |

| Memory Bandwidth | System Dependent | 448.0 GB/s |

| Shading Units | 1536 | 3840 |

| TMUs | 48 | 120 |

| ROPs | 24 | 48 |

| RT Cores | 12 | 30 |

| Tensor Cores | null | 120 |

| Pixel Rate | 60.00 GPixel/s | 119.9 GPixel/s |

| Texture Rate | 120.0 GTexel/s | 299.6 GTexel/s |

| FP32 | 7.680 TFLOPS | 19.18 TFLOPS |

| FP16 | 15.36 TFLOPS (2:1) | 19.18 TFLOPS (1:1) |

| TDP | 80 W | 145 W |

| Slot Width | IGP | Dual-slot |

| Power Connectors | None | 1x 8-pin |

| Suggested PSU | null | 300 W |

| Bus Interface | IGP | PCIe 5.0 x8 |

| Display Outputs | Portable Device Dependent | 1x HDMI 2.1b, 3x DisplayPort 2.1b |

| Release Date | 2026-01-26 | 2026-05-31 |

| Launch MSRP | null | 299 USD |

Where Each One Wins

The Intel Arc B390 wins in power efficiency and integration. Its 80 W TDP is nearly half the RTX 5060's 145 W, and it requires no power connectors or expansion slot, making it suitable for compact portable devices. The 300 MHz base clock suggests aggressive power saving at idle, and the system shared memory eliminates the cost of dedicated VRAM. The Arc B390's earlier release date of 2026-01-26 gives it a time-to-market advantage. In the recorded 3DMark Steel Nomad test, it outperforms the GT 520MX by 1.3%, the GeForce 800M by 1.5%, the GT 625 OEM by 2.5%, and the GT 710 by 2.7%, showing it can edge out those legacy parts.

The NVIDIA GeForce RTX 5060 GB205 wins in every raw performance category. Its 19.18 TFLOPS FP32 is 2.5 times the Arc B390, and the 448.0 GB/s dedicated bandwidth is a fixed figure versus the Arc B390's system dependent bandwidth. The RTX 5060's 120 tensor cores enable AI acceleration, and its 30 RT cores provide 2.5 times the ray tracing hardware of the Arc B390. The 1:1 FP16 ratio allows full-rate half-precision compute, unlike the Arc B390's 2:1 ratio. The RTX 5060's 3840 shading units, 120 TMUs, and 48 ROPs are all exactly double the Arc B390's counts. The 2497 MHz boost clock is nearly identical to the Arc B390's 2500 MHz, but the 2280 MHz base clock is 7.6 times higher, indicating sustained performance under load. The PCIe 5.0 x8 interface and dual-slot design with 1x 8-pin power connector support a 300 W suggested PSU, enabling the higher power draw for greater throughput. The RTX 5060 also offers fixed display outputs (1x HDMI 2.1b, 3x DisplayPort 2.1b) versus the Arc B390's portable device dependent outputs.

The use-case split is clear: the Arc B390 serves low-power integrated scenarios where space and cooling are limited, while the RTX 5060 GB205 serves discrete desktop builds requiring maximum compute, memory bandwidth, and AI features. The RTX 5060's launch MSRP of 299 USD positions it as a mainstream discrete option, while the Arc B390 has no recorded price. The RTX 5060's 50th percentile ranking versus the Arc B390's 9th percentile confirms the performance tier gap, even without direct head-to-head benchmark scores.