AMD Radeon 820M vs Intel Arc B390 Comparison

Where Each One Wins

The recorded data splits these two integrated graphics processors into very different competitive positions. The AMD Radeon 820M holds a percentile ranking of 50 against all GPUs in the database, placing it at the midpoint of the entire distribution. The Intel Arc B390, by contrast, sits at the 9th percentile, a much lower overall standing. That percentile gap alone suggests the AMD part is the stronger general-purpose performer, but the benchmark record is not symmetrical between the two.

The AMD Radeon 820M has no individual benchmark scores recorded in the database, so its competitive position is established entirely through its percentile rank. The Intel Arc B390 has one recorded result, a 3DMark Steel Nomad DX12 score of 1482 points. That single score anchors the Intel part to a specific performance tier, one that sits close to several older NVIDIA discrete parts. The nearest rivals to the Arc B390 are the NVIDIA GeForce GT 520MX with an average score of 1463, the NVIDIA GeForce 800M with 1460, the NVIDIA GeForce GT 625 OEM with 1446, and the NVIDIA GeForce GT 710 with 1443. The Arc B390 leads each of these by a small margin, between 1.3% and 2.7%. Those are narrow advantages, not commanding ones.

The AMD Radeon 820M, lacking any recorded benchmark entries, cannot be placed in a head-to-head win count against the Arc B390. The database shows zero wins for each side in direct comparison, which reflects the absence of paired benchmark data rather than a tied result. The use-case split therefore rests on architectural characteristics and the one available Intel score. For workloads that resemble the 3DMark Steel Nomad DX12 test, the Arc B390 demonstrates a measurable, if modest, edge over its closest recorded competitors. For broader workload coverage, the AMD part's 50th percentile standing implies a more balanced position across the full GPU landscape.

Architecture Differences

The two processors come from different manufacturers and use different underlying designs. AMD builds the Radeon 820M on the Krackan Point 2 chip with RDNA 3.5 architecture, belonging to the Navi III IGP generation for Strix Point Mobile. Intel builds the Arc B390 on the Panther Lake chip with Xe3-LPG architecture, belonging to the Arc Graphics-M generation for Panther Lake. The process nodes differ as well: AMD uses a 4 nm process from TSMC, while Intel uses a 3 nm process from its own foundry. Both chips report unknown transistor counts and unknown die sizes, so no comparison can be made on those figures.

The execution resources diverge sharply. The Radeon 820M has 128 shading units, 8 texture mapping units, and 4 raster operation pipelines. The Arc B390 has 1536 shading units, 48 texture mapping units, and 24 raster operation pipelines. Those are large differences in raw resource counts. Ray tracing hardware also differs: AMD includes 2 ray tracing cores, while Intel includes 12. Neither part lists tensor cores.

Clock behavior differs in both direction and magnitude. The AMD part runs at a 400 MHz base clock and boosts to 2800 MHz. The Intel part runs at a 300 MHz base clock and boosts to 2500 MHz. So the AMD part has a higher base and a higher boost clock, but the Intel part compensates with a much larger shading unit count. The memory interfaces are identical in kind: both use system shared memory with system dependent bandwidth and no dedicated VRAM. Both are integrated graphics processors with no power connectors and portable device dependent display outputs.

The API support is effectively identical. Both list DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Neither part is differentiated on software feature support in the database.

Head-to-Head Benchmarks

Direct comparison data between the Radeon 820M and the Arc B390 is not present in the database. The head-to-head benchmark list is empty, and the win counts are zero for both parts. What can be analyzed instead is the Arc B390's performance against its nearest recorded rivals, which provides context for where the Intel part sits.

The Arc B390's 3DMark Steel Nomad DX12 score is 1482. The nearest rival, the NVIDIA GeForce GT 520MX, averages 1463, placing the Arc B390 1.3% ahead. The NVIDIA GeForce 800M averages 1460, a 1.5% gap. The NVIDIA GeForce GT 625 OEM averages 1446, a 2.5% gap. The NVIDIA GeForce GT 710 averages 1443, a 2.7% gap. These deltas are consistent and positive, but they are small. The Arc B390 does not blow past its closest competitors; it edges them out by a narrow margin in every case.

The Radeon 820M has no equivalent recorded scores, so no exact numeric comparison to the Arc B390 or to any rival can be made. The database assigns it a 50th percentile rank against all GPUs, which is substantially higher than the Arc B390's 9th percentile. That percentile field is the only quantitative bridge between the two parts, and it favors AMD by a wide margin. The absence of a measured Intel score at the same percentile scale for AMD means the exact point gap cannot be stated, but the ordinal ranking is clear.

The Arc B390's texture rate of 120.0 GTexel/s and pixel rate of 60.00 GPixel/s far exceed the Radeon 820M's 22.40 GTexel/s and 11.20 GPixel/s. The FP32 throughput shows a similar pattern: 7.680 TFLOPS for Intel versus 716.8 GFLOPS for AMD. The FP16 figures are 15.36 TFLOPS for Intel versus 716.8 GFLOPS for AMD, with the Intel part operating at a 2:1 ratio and the AMD part at 1:1. These specification-level numbers indicate that the Arc B390 has a much higher theoretical peak throughput, even though its only recorded benchmark places it near older NVIDIA parts.

Specification Differences

The power envelope separates the two clearly. The Radeon 820M carries a 15 W TDP, while the Arc B390 carries an 80 W TDP. That is a fivefold difference in rated power draw. Both are integrated parts with no power connectors, and both are listed as active production parts. The Radeon 820M uses a PCIe 4.0 x8 bus interface, while the Arc B390 lists its bus interface simply as IGP.

Clock speeds differ as noted: the AMD part has a 400 MHz base and 2800 MHz boost, the Intel part a 300 MHz base and 2500 MHz boost. The AMD part is faster in both clock figures. The Intel part is far larger in resource counts: 1536 versus 128 shading units, 48 versus 8 TMUs, 24 versus 4 ROPs, and 12 versus 2 ray tracing cores. The pixel rate of the Arc B390 is 60.00 GPixel/s versus 11.20 GPixel/s for the Radeon 820M. The texture rate is 120.0 GTexel/s versus 22.40 GTexel/s. The FP32 throughput is 7.680 TFLOPS versus 716.8 GFLOPS, and the FP16 throughput is 15.36 TFLOPS versus 716.8 GFLOPS, with the Intel part doubling its FP32 rate and the AMD part matching it.

Release timing also differs. The AMD part has a release date of 2025-02-28, while the Intel part is dated 2026-01-26. The AMD part lists a predecessor, Navi II IGP, while the Intel part lists none. Neither part has a successor, a launch MSRP, or dimensions recorded in the database.

FAQ

Q: Which GPU has the higher boost clock?

A: The AMD Radeon 820M boosts to 2800 MHz, while the Intel Arc B390 boosts to 2500 MHz.

Q: How does the Intel Arc B390 compare to its nearest rival in the database?

A: The Arc B390 scores 1482 in 3DMark Steel Nomad DX12. Its nearest rival, the NVIDIA GeForce GT 520MX, averages 1463, putting the Arc B390 1.3% ahead.

Q: What is the TDP difference between the two parts?

A: The AMD Radeon 820M has a 15 W TDP, and the Intel Arc B390 has an 80 W TDP.

Q: Do the two GPUs support the same graphics APIs?

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

Q: Which part has more shading units?

A: The Intel Arc B390 has 1536 shading units, while the AMD Radeon 820M has 128.

Q: What percentile rank does each GPU hold in the database?

A: The AMD Radeon 820M is at the 50th percentile against all GPUs, and the Intel Arc B390 is at the 9th percentile.

The Verdict

The database presents a split verdict. The Intel Arc B390 has a measured benchmark score and a large theoretical throughput advantage. Its 3DMark Steel Nomad DX12 result of 1482, its 1536 shading units, its 7.680 TFLOPS of FP32 throughput, and its 60.00 GPixel/s pixel rate all point to a part with substantial raw capability. Its 80 W TDP reflects that ambition. The Radeon 820M, with only 128 shading units, 716.8 GFLOPS of FP32, and an 11.20 GPixel/s pixel rate, is far smaller on paper.

The percentile standings tell a different story. The Radeon 820M sits at the 50th percentile against all GPUs, while the Arc B390 sits at the 9th. That places the AMD part in the middle of the full distribution and the Intel part near the bottom, despite the Intel part's much larger specification sheet. The Arc B390's nearest rivals are all older NVIDIA discrete parts from the GT 500, 800M, and GT 600 families, and it beats them by only 1.3% to 2.7%. The recorded data does not include a direct benchmark between the two, so the percentile gap is the only direct ranking evidence.

For users focused on measured results, the Arc B390 provides a concrete score and a small but consistent lead over its nearest recorded competitors. For users focused on overall standing across all GPUs, the Radeon 820M holds a far higher percentile position. The data does not support a single winner; it supports two different profiles. The AMD part is the higher-ranked part in the database, while the Intel part is the higher-throughput part on paper with the only recorded benchmark score.