AMD Radeon RX 6550S vs Intel Arc B390 Comparison

Head-to-Head Benchmarks

The database contains no shared benchmark runs for the AMD Radeon RX 6550S and Intel Arc B390, so a direct score-for-score comparison is not possible from the recorded data. The AMD part has no benchmark entries in the database, while the Intel part has a single recorded result in the 3DMark Steel Nomad DX12 test.

The Intel Arc B390 scores 1482 points in 3DMark Steel Nomad DX12. That result places it in the 9th percentile of all GPUs in the database, meaning the vast majority of recorded graphics processors outperform it in this particular workload. The AMD Radeon RX 6550S has no recorded benchmark score, so its percentile ranking of 50 is based on the database's classification of its specifications rather than any measured performance data.

Because there are no head-to-head benchmark entries, the only quantitative performance comparison available is the Intel part's position relative to its nearest rivals. The Arc B390 leads the NVIDIA GeForce GT 520MX by 1.3 percent, with the NVIDIA GeForce 800M trailing by 1.5 percent. The gap widens slightly against the NVIDIA GeForce GT 625 OEM, which sits 2.5 percent behind, and the NVIDIA GeForce GT 710 trails by 2.7 percent. These are narrow margins, indicating that the Arc B390 performs in the same performance class as those older NVIDIA parts in the Steel Nomad test.

The AMD Radeon RX 6550S cannot be placed relative to those rivals because it has no measured results. Its theoretical compute figures are substantially higher than the Intel part's, with 4.915 TFLOPS FP32 against 7.680 TFLOPS for the Arc B390, but the absence of benchmark data means the database cannot confirm how that theoretical advantage translates into actual application performance. The Intel part's 7.680 TFLOPS is roughly 56 percent higher than the AMD part's 4.915 TFLOPS on paper, yet the Arc B390's measured Steel Nomad score sits in the 9th percentile, which suggests that raw FP32 throughput alone does not determine real-world results in this benchmark.

The data shows that the Intel Arc B390 has a confirmed, if modest, performance baseline in one modern DX12 test. The AMD Radeon RX 6550S has no such baseline in the database, and its 50th percentile classification should be read as a specification-based estimate rather than a tested result.

FAQ

Q: Which GPU has a higher FP32 compute rating?

A: The Intel Arc B390 is rated at 7.680 TFLOPS FP32, while the AMD Radeon RX 6550S is rated at 4.915 TFLOPS. The Intel part also has a higher FP16 rating at 15.36 TFLOPS (2:1) compared to 9.830 TFLOPS (2:1) for the AMD part.

Q: How does the Intel Arc B390 perform in 3DMark Steel Nomad DX12?

A: The Arc B390 scores 1482 points, which places it in the 9th percentile of all GPUs in the database. It leads the NVIDIA GeForce GT 520MX by 1.3 percent, the GeForce 800M by 1.5 percent, the GeForce GT 625 OEM by 2.5 percent, and the GeForce GT 710 by 2.7 percent.

Q: Does the AMD Radeon RX 6550S have any recorded benchmark scores?

A: No. The database lists no benchmark entries for the RX 6550S, and its average benchmark score is recorded as 0. Its 50th percentile ranking is derived from its specifications rather than measured performance.

Q: What are the memory configurations of the two GPUs?

A: The AMD Radeon RX 6550S uses 4 GB of GDDR6 memory on a 64-bit bus with 128.0 GB/s of bandwidth. The Intel Arc B390 uses system shared memory, with system-dependent bandwidth and a system shared bus width.

Q: Which GPU has more shading units?

A: The Intel Arc B390 has 1536 shading units, compared to 1024 for the AMD Radeon RX 6550S. However, the AMD part has more texture mapping units (64 versus 48) and more render output units (32 versus 24).

Q: What are the power ratings of the two GPUs?

A: The AMD Radeon RX 6550S has a TDP of 50 W, while the Intel Arc B390 has a TDP of 80 W. Both are integrated-class parts with no power connectors and portable-device-dependent display outputs.

The Verdict

The recorded data supports a straightforward choice for users who need a confirmed performance baseline: the Intel Arc B390 has a measured 3DMark Steel Nomad DX12 score of 1482, placing it in the 9th percentile, and it outperforms its nearest recorded rivals by margins between 1.3 and 2.7 percent. The AMD Radeon RX 6550S has no benchmark entries, so the database cannot verify its actual performance.

For compute-oriented tasks, the Intel part holds the advantage on paper. Its 7.680 TFLOPS FP32 and 15.36 TFLOPS FP16 ratings are both higher than the AMD part's 4.915 TFLOPS and 9.830 TFLOPS respectively. The Intel part also has 1536 shading units against 1024 for the AMD part, and 12 ray tracing cores against 16 for the AMD part, though the AMD part counters with more TMUs (64 versus 48) and ROPs (32 versus 24).

The AMD Radeon RX 6550S offers a lower TDP of 50 W against 80 W for the Intel part, which may matter in thermally constrained portable designs. It also uses dedicated 4 GB GDDR6 memory with 128.0 GB/s bandwidth, whereas the Intel part relies on system shared memory with system-dependent bandwidth. The AMD part's 6 nm process node from TSMC contrasts with the Intel part's 3 nm node from Intel, and the AMD die is measured at 107 mm² with 5,400 million transistors, while the Intel die size and transistor count are listed as unknown.

The database indicates the AMD part is positioned at the 50th percentile of all GPUs based on its specifications, while the Intel part's measured result sits at the 9th percentile. That percentile gap suggests the RX 6550S is categorized as a mid-tier part by specification, while the Arc B390's only recorded benchmark places it low in the overall distribution. Without direct benchmark data for the AMD part, the verification of its theoretical position remains unavailable.

Users who require a tested result in DX12 workloads should consider the Intel Arc B390, as its 1482 Steel Nomad score provides a concrete reference point. Users who prioritize lower power draw and dedicated memory may prefer the AMD Radeon RX 6550S, but they should be aware that the database contains no measured performance data to confirm its real-world behavior.

Specification Differences

The two GPUs differ across nearly every recorded specification field. The AMD Radeon RX 6550S uses the Navi 24 chip with RDNA 2.0 architecture, fabricated on a 6 nm TSMC process with 5,400 million transistors on a 107 mm² die. The Intel Arc B390 uses the Panther Lake chip with Xe3-LPG architecture, fabricated on a 3 nm Intel process, with transistor count and die size listed as unknown.

Clock behavior differs significantly. The AMD part has a base clock of 2000 MHz, a boost clock of 2400 MHz, and a game clock of 2170 MHz. The Intel part has a base clock of 300 MHz and a boost clock of 2500 MHz, with no game clock listed.

Memory configurations are entirely different. The AMD Radeon RX 6550S has 4 GB of GDDR6 on a 64-bit bus with 128.0 GB/s bandwidth and a memory clock of 2000 MHz (16 Gbps effective). The Intel Arc B390 uses system shared memory with a system shared type, bus width, and system-dependent bandwidth.

Compute unit counts differ. The AMD part has 1024 shading units, 64 TMUs, 32 ROPs, and 16 ray tracing cores. The Intel part has 1536 shading units, 48 TMUs, 24 ROPs, and 12 ray tracing cores. Neither part has tensor cores listed.

Rates and throughput differ as well. The AMD part has a pixel rate of 76.80 GPixel/s and a texture rate of 153.6 GTexel/s. The Intel part has a pixel rate of 60.00 GPixel/s and a texture rate of 120.0 GTexel/s.

Power and interface specifications differ. The AMD Radeon RX 6550S has a TDP of 50 W and a PCIe 4.0 x4 bus interface. The Intel Arc B390 has a TDP of 80 W and an IGP bus interface. Both are IGP slot width with no power connectors, and both have portable-device-dependent display outputs.

API support matches: both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Release dates differ, with the AMD part released on 2023-01-03 and the Intel part released on 2026-01-26. The AMD part is listed as the successor to Polaris Mobile, while the Intel part has no predecessor listed.

Architecture Differences

The architectural split is fundamental. The AMD Radeon RX 6550S belongs to the Radeon RX 6000 series and uses RDNA 2.0 architecture, built around the Navi 24 chip. Its generation is recorded as Navi Mobile (RX 6000M), placing it in AMD's mobile-oriented lineup. The Intel Arc B390 uses Xe3-LPG architecture built around the Panther Lake chip, and its generation is recorded as Arc Graphics-M (Panther Lake), indicating an integrated graphics implementation tied to Intel's Panther Lake platform.

The process nodes reflect different manufacturing approaches. The AMD part uses a 6 nm process from TSMC, with a measured die of 107 mm² and a transistor count of 5,400 million, giving a transistor density of 50.5 million transistors per square millimeter. The Intel part uses a 3 nm process from Intel, with transistor count, die size, and density all listed as unknown in the database.

The memory architecture differs fundamentally. The AMD Radeon RX 6550S uses dedicated GDDR6 memory with a fixed 128.0 GB/s bandwidth and a 64-bit bus. The Intel Arc B390 uses system shared memory, meaning its bandwidth is system dependent and its memory type and bus width are not fixed specifications.

The compute architecture also differs in resource allocation. The Intel part allocates more shading units (1536 versus 1024) but fewer TMUs (48 versus 64) and ROPs (24 versus 32). Ray tracing cores also differ, with 16 on the AMD part and 12 on the Intel part.

The Intel part's higher FP32 throughput of 7.680 TFLOPS against 4.915 TFLOPS for the AMD part, combined with its lower TMU and ROP counts, indicates a design that emphasizes raw shader compute over texture and pixel throughput. The AMD part's higher TMU and ROP counts relative to its shading units suggest a more balanced allocation toward texture sampling and raster output work.

The AMD part's 6 nm TSMC process with a 107 mm² die and 5,400 million transistors stands in contrast to the Intel part's 3 nm Intel process with unknown die measurements. The AMD part also carries a lower TDP of 50 W against 80 W for the Intel part, which aligns with its smaller shading unit count and lower clock base. The Intel part compensates with a higher boost clock of 2500 MHz versus 2400 MHz for the AMD part.

The bus interfaces reflect different integration strategies. The AMD part uses PCIe 4.0 x4, indicating a discrete-style connection even within a mobile context. The Intel part uses an IGP bus interface, confirming its integrated position within the Panther Lake processor package. Both parts share identical API support profiles, including DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.