AMD Radeon 820M vs Intel Arc Pro B370 Comparison

# AMD Radeon 820M vs Intel Arc Pro B370

The AMD Radeon 820M and Intel Arc Pro B370 are both integrated graphics solutions, but they target different performance strata within the mobile and compact device space. The Radeon 820M is built on AMD’s RDNA 3.5 architecture, while the Arc Pro B370 uses Intel’s Xe3-LPG architecture. The benchmark database records zero head-to-head wins for either part, and both sit at the 50th percentile among all GPUs, indicating that neither is positioned as an extreme outlier in either direction. The data, however, reveals substantial architectural and specification differences that dictate where each integrated GPU is better suited.

Where Each One Wins

The AMD Radeon 820M is designed for efficiency and basic visual output. Its 15 W TDP and 4 nm TSMC process node make it a low-power option for ultrathin laptops and portable devices. The Radeon 820M operates with a base clock of 400 MHz and a boost clock of 2800 MHz. Those clocks are higher than the Intel part’s figures, which suggests that for lightly threaded or latency-sensitive tasks, the AMD solution may respond more quickly. The Radeon 820M also uses a PCIe 4.0 x8 bus interface, which provides a dedicated connection path for system memory access. For everyday desktop tasks, video playback, and light productivity workloads, the Radeon 820M’s lower power envelope and compact design are clear advantages.

The Intel Arc Pro B370, by contrast, is built for noticeably heavier graphics loads. It carries a 25 W TDP, which is 10 W higher than the AMD part, and it uses Intel’s 3 nm process node. The Arc Pro B370 has 1280 shading units, 40 texture mapping units, and 20 ROPs, all of which dwarf the Radeon 820M’s 128 shading units, 8 TMUs, and 4 ROPs. The Intel part also includes 10 ray tracing cores compared to 2 on the AMD side. These specifications indicate that the Arc Pro B370 is meant to handle more demanding 3D rendering, ray-traced workloads, and higher-resolution output. Its pixel rate of 48.00 GPixel/s is more than four times the Radeon 820M’s 11.20 GPixel/s, and its texture rate of 96.00 GTexel/s is similarly more than four times the AMD part’s 22.40 GTexel/s.

The data shows a clean split: the Radeon 820M wins on power efficiency and clock speed headroom, while the Arc Pro B370 wins on raw throughput and feature count. For users who need maximum battery life and only light graphics work, the AMD part is the logical choice. For users who want integrated graphics that can approach discrete-class performance in a thin and light chassis, the Intel part dominates the specification sheet.

Architecture Differences

The two GPUs come from completely different architectural lineages. The AMD Radeon 820M uses RDNA 3.5, a refinement of AMD’s RDNA family, and it is part of the Navi III IGP generation for Strix Point Mobile. The chip is named Krackan Point 2, and it is fabricated on a 4 nm process at TSMC. The Radeon 820M’s FP32 throughput is 716.8 GFLOPS, and its FP16 throughput is the same 716.8 GFLOPS with a 1:1 ratio. This means the AMD part does not gain any extra throughput by switching to half-precision math.

The Intel Arc Pro B370 uses Xe3-LPG architecture, which is part of the Arc Graphics-WM generation for Panther Lake. It is fabricated on Intel’s own 3 nm process. The FP32 throughput is 6.144 TFLOPS, which is roughly 8.6 times higher than the AMD part. The FP16 throughput jumps to 12.29 TFLOPS with a 2:1 ratio, meaning the Intel GPU doubles its throughput when operating on half-precision data. That is a significant advantage for compute workloads that can use FP16, such as certain AI inference tasks or shader-heavy effects.

Both GPUs support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, so API-level feature parity is present. The difference lies in execution resources. The Arc Pro B370 has 10 times the shading units, 5 times the TMUs, and 5 times the ROPs compared to the Radeon 820M. The ray tracing core count is 10 versus 2, a 5:1 difference. The Intel part also has a lower boost clock at 2400 MHz versus 2800 MHz, but the massive execution resource advantage more than compensates in throughput-bound scenarios.

Memory is system-shared on both parts, with bus width and bandwidth marked as system dependent. That means actual memory performance will vary with the host platform’s memory configuration, but the compute and rasterization differences are fixed by the silicon.

The Verdict

The benchmark database shows no direct head-to-head wins for either GPU, and both parts share the same 50th percentile ranking among all GPUs. That suggests that in the aggregate, they are not statistically separable in overall performance, despite the huge specification gap. The explanation likely lies in the use cases: the Radeon 820M is a low-power IGP for basic tasks, while the Arc Pro B370 is a higher-power IGP for more intensive work. Neither is a top-tier part, and both sit at the median of the database.

From the recorded data, the Radeon 820M should be chosen for devices where power consumption is the primary constraint. Its 15 W TDP and 4 nm process make it suitable for fanless or ultraportable designs. The higher boost clock of 2800 MHz also helps with bursty, short-duration workloads. The Arc Pro B370 is the choice for integrated graphics users who need substantial compute throughput, higher pixel and texture rates, and ray tracing capability, all within a 25 W envelope. The 3 nm process and 6.144 TFLOPS FP32 throughput put it in a different performance class.

There is no price data in the database, so no cost-based comparison is possible. The production status for both is Active, and both are current products. The Radeon 820M was released on February 28, 2025, while the Arc Pro B370 has a release date of January 26, 2026, making the Intel part roughly 11 months newer. The AMD part’s predecessor is Navi II IGP, and the Intel part’s predecessor is HD Graphics-WM, so both are successors to earlier integrated graphics families.

FAQ

Q: Which GPU has higher FP32 compute throughput?

A: The Intel Arc Pro B370 delivers 6.144 TFLOPS, while the AMD Radeon 820M delivers 716.8 GFLOPS. The Intel part has roughly 8.6 times the FP32 throughput.

Q: Do both GPUs support ray tracing?

A: Yes, both support DirectX 12 Ultimate (12_2), which includes ray tracing. The Intel Arc Pro B370 has 10 ray tracing cores, while the AMD Radeon 820M has 2.

Q: Which GPU uses more power?

A: The Intel Arc Pro B370 has a 25 W TDP, which is 10 W higher than the AMD Radeon 820M’s 15 W TDP.

Q: What are the process nodes for each GPU?

A: The AMD Radeon 820M is fabricated on a 4 nm process at TSMC, while the Intel Arc Pro B370 is fabricated on a 3 nm process at Intel.

Q: Which GPU has a higher boost clock?

A: The AMD Radeon 820M boosts to 2800 MHz, while the Intel Arc Pro B370 boosts to 2400 MHz. The AMD part has a 400 MHz higher boost clock.

Q: Are both GPUs integrated or discrete?

A: Both are IGPs with no power connectors and a slot width of IGP. Both use system-shared memory, with bandwidth dependent on the host system.

Head-to-Head Benchmarks

The database contains no recorded head-to-head benchmark entries for the AMD Radeon 820M and Intel Arc Pro B370. The winsA and winsB fields are both zero, and the headToHeadBenchmarks array is empty. As a result, direct performance comparisons must be inferred from the specification data rather than measured results.

The most decisive specification gap is in pixel throughput. The Intel Arc Pro B370 achieves 48.00 GPixel/s, which is 4.29 times the Radeon 820M’s 11.20 GPixel/s. Texture rate follows the same pattern: 96.00 GTexel/s versus 22.40 GTexel/s, a 4.29x difference. These ratios are directly proportional to the ROP and TMU counts, which are 20 versus 4 and 40 versus 8, respectively.

FP32 throughput shows an even larger divide. The Intel part’s 6.144 TFLOPS versus the AMD part’s 716.8 GFLOPS represents an 8.57x advantage. FP16 throughput is more lopsided still: 12.29 TFLOPS versus 716.8 GFLOPS, a 17.14x difference, because the Intel architecture uses a 2:1 FP16 ratio while the AMD architecture uses a 1:1 ratio.

Clock speeds favor the AMD part. The Radeon 820M has a base clock of 400 MHz and a boost clock of 2800 MHz. The Arc Pro B370 has a base clock of 300 MHz and a boost clock of 2400 MHz. The AMD part’s boost clock is 16.7% higher, which may help in latency-bound or lightly threaded scenarios. However, the Intel part has 10 times the shading units, so any clock advantage on the AMD side is overwhelmed in parallel workloads.

Both GPUs share the same API support: DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Both have system-shared memory, so memory capacity and bandwidth are not fixed specifications. The bus interface differs: the AMD part uses PCIe 4.0 x8, while the Intel part is marked simply as IGP, which may indicate a different integration path.

Specification Differences

The recorded data shows the following differences between the AMD Radeon 820M and Intel Arc Pro B370:

• Architecture: RDNA 3.5 versus Xe3-LPG
• Generation: Navi III IGP (Strix Point Mobile) versus Arc Graphics-WM (Panther Lake)
• Process Node: 4 nm at TSMC versus 3 nm at Intel
• Base Clock: 400 MHz versus 300 MHz
• Boost Clock: 2800 MHz versus 2400 MHz
• Shading Units: 128 versus 1280
• Texture Mapping Units: 8 versus 40
• Render Output Units: 4 versus 20
• Ray Tracing Cores: 2 versus 10
• Pixel Rate: 11.20 GPixel/s versus 48.00 GPixel/s
• Texture Rate: 22.40 GTexel/s versus 96.00 GTexel/s
• FP32 Throughput: 716.8 GFLOPS versus 6.144 TFLOPS
• FP16 Throughput: 716.8 GFLOPS (1:1) versus 12.29 TFLOPS (2:1)
• TDP: 15 W versus 25 W
• Bus Interface: PCIe 4.0 x8 versus IGP
• Release Date: February 28, 2025 versus January 26, 2026
• Predecessor: Navi II IGP versus HD Graphics-WM
• Chip Name: Krackan Point 2 versus Panther Lake

Both parts share identical memory configurations (system shared, system dependent bandwidth), no power connectors, no dimensions, no suggested PSU, and no launch MSRP. Transistor count and die size are unknown for both. The slot width is IGP for both, and display outputs are portable device dependent for both. The production status is Active for both, and both are at the 50th percentile among all GPUs.