Intel Arc B390 vs Intel Arc B580 Comparison

Head-to-Head Benchmarks

The only direct comparison available in the database is the 3DMark Steel Nomad DX12 test, and the results are decisive. The Intel Arc B580 scores 3068 points, while the Intel Arc B390 manages 1482 points. This gives the B580 a 51.7% advantage, a margin that places the two cards in completely different performance tiers.

In percentile terms, the gap is even starker. The B580 sits in the 68th percentile of all GPUs tracked by the database, while the B390 lands in the 9th percentile. That 59-point percentile spread reflects a fundamental class difference: the B580 is a discrete desktop card competing near the mid-range, while the B390 is an integrated graphics solution that ranks among entry-level parts.

The B580's average benchmark score of 23021 across all recorded tests further illustrates its position. Its nearest rivals in the database include the AMD Radeon RX 580 2048SP, which scores 23061 (a 0.2% difference), the NVIDIA GeForce RTX 2080 at 22895 (0.6% ahead), and the NVIDIA GeForce RTX 3080 at 23172 (0.7% behind). These deltas show the B580 sitting in a dense cluster of established discrete GPUs, within roughly one percent of each. The B390, by contrast, has nearest rivals like the NVIDIA GeForce GT 520MX at 1463 (1.3% behind) and the NVIDIA GeForce 800M at 1460 (1.5% behind), placing it firmly in legacy integrated and low-end mobile territory.

Beyond the shared Steel Nomad result, the B580 has a much larger benchmark footprint in the database. Its OpenCL score of 92821 and Vulkan score of 109672 in Geekbench demonstrate strong compute performance. PassMark results show 15748 in G3D, 7729 in GPU compute, and a range of DirectX scores from 76 in DX10 and DX12 up to 183 in DX9. The B390 has no comparable entries, meaning the database contains no evidence of it reaching similar levels in any compute or legacy API workload.

The Verdict

The recorded data points to a clear split. The Intel Arc B580 is the only option for anyone seeking discrete desktop performance. Its 3068 Steel Nomad score, 68th percentile ranking, and compute results across OpenCL, Vulkan, and PassMark put it in a different league from the B390. The B580's nearest rivals, including the RTX 2080 and RTX 3080, confirm that it trades blows with established mid-range and upper-mid-range cards from previous generations.

The Intel Arc B390, meanwhile, is an integrated processor graphics solution. Its 1482 Steel Nomad score and 9th percentile ranking place it alongside the NVIDIA GeForce GT 520MX and GT 710, parts that are multiple generations old and were entry-level even in their heyday. The B390's performance class makes it suitable only for basic graphical tasks, and the database contains no evidence of it handling modern gaming workloads at competitive frame rates.

Both cards share the same DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4 API support, so feature compatibility is not a differentiator. The decision rests entirely on performance and form factor. The B580 delivers desktop-class results; the B390 is an IGP built for mobile and compact systems where a discrete card is not an option.

Architecture Differences

The two Intel parts come from different architectural generations and manufacturing processes. The B390 is built on the Xe3-LPG architecture, part of the Panther Lake chip, and uses Intel's 3 nm process. The B580 uses the Xe2-HPG architecture on the BMG-G21 chip, manufactured by TSMC on a 5 nm process. The B580's datasheet lists 19,600 million transistors on a 272 mm² die, with a transistor density of 72.1M per mm². The B390's transistor count and die size are recorded as unknown.

The B580 has substantially more execution resources. It carries 2560 shading units, 160 texture mapping units, 80 ROPs, and 20 ray tracing cores. The B390 has 1536 shading units, 48 TMUs, 24 ROPs, and 12 ray tracing cores. In raw throughput, the B580 delivers 13.67 TFLOPS of FP32 performance and 27.34 TFLOPS of FP16 (2:1), while the B390 manages 7.680 TFLOPS FP32 and 15.36 TFLOPS FP16 (2:1). Pixel rates are 213.6 GPixel/s for the B580 versus 60.00 GPixel/s for the B390, and texture rates are 427.2 GTexel/s versus 120.0 GTexel/s.

Memory is another major divide. The B580 has 12 GB of GDDR6 on a 192-bit bus, with 2375 MHz memory clock, 19 Gbps effective data rate, and 456.0 GB/s of bandwidth. The B390 uses system shared memory with a system dependent bandwidth, meaning its memory performance varies with the host platform. Clock behavior also differs: the B580 runs at a fixed 2670 MHz base and boost, while the B390 has a 300 MHz base and 2500 MHz boost.

Power and physical characteristics reflect their different roles. The B580 is a dual-slot card, 272 mm long, 115 mm high, and 45 mm wide, with a 190 W TDP and a single 8-pin power connector. It requires a 450 W suggested PSU and connects via PCIe 4.0 x8. The B390 is an IGP with an 80 W TDP, no power connectors, and no dedicated slot width. Its display outputs are portable device dependent, whereas the B580 offers 1x HDMI 2.1a and 3x DisplayPort 2.1. The B580 was released on 2024-12-12 with a launch MSRP of 249 USD; the B390 followed on 2026-01-26 with no recorded MSRP.

FAQ

Q: Which GPU has the higher 3DMark Steel Nomad score?

A: The Intel Arc B580 scores 3068, which is 51.7% higher than the Intel Arc B390's 1482.

Q: How do the two cards compare in overall GPU percentile ranking?

A: The B580 ranks in the 68th percentile of all GPUs, while the B390 ranks in the 9th percentile.

Q: What memory configuration does each card use?

A: The B580 uses 12 GB of GDDR6 on a 192-bit bus with 456.0 GB/s bandwidth. The B390 uses system shared memory with system dependent bandwidth.

Q: Are both cards based on the same architecture?

A: No. The B390 uses Xe3-LPG on Intel's 3 nm process, while the B580 uses Xe2-HPG on TSMC's 5 nm process.

Q: What is the TDP difference between the two?

A: The B580 has a 190 W TDP and requires a 450 W suggested PSU, while the B390 has an 80 W TDP and uses no power connectors.

Q: Do both support the same graphics APIs?

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

Where Each One Wins

The Intel Arc B580 wins every recorded performance comparison. In the single head-to-head benchmark, it leads by 51.7%. Its compute results, including 92821 in Geekbench OpenCL, 109672 in Geekbench Vulkan, and 7729 in PassMark GPU compute, have no counterpart in the B390's recorded data. The B580 also wins on memory bandwidth at 456.0 GB/s versus system dependent, on pixel throughput at 213.6 GPixel/s versus 60.00 GPixel/s, and on texture rate at 427.2 GTexel/s versus 120.0 GTexel/s. Its 12 GB dedicated GDDR6 frame buffer is a clear advantage over shared system memory for gaming and graphics workloads.

The Intel Arc B390 wins only in power efficiency and physical footprint. Its 80 W TDP is 110 W lower than the B580's 190 W, and as an IGP it requires no slot space and no power connectors. For systems where a discrete card is physically impossible, such as ultraportable devices, the B390 is the only choice of the two. The B390's 3 nm process also suggests a modern manufacturing node, though the database records no transistor density or die size for it.

The B580's launch MSRP of 249 USD and its 68th percentile ranking indicate it was positioned as a mainstream desktop part. Its nearest rivals, including the RTX 2080 and RTX 3080, show that it competes against much older but once-high-end hardware. The B390, by contrast, sits near the GT 710 and GT 520MX, parts that were never considered gaming-capable even at launch. For gaming, rendering, or compute tasks, the B580 is the only viable option. For minimal-power integrated graphics in compact systems, the B390 fills that specific role.