AMD Radeon 860M vs Intel Arc A380E Comparison

Where Each One Wins

The AMD Radeon 860M and Intel Arc A380E occupy different corners of the hardware landscape, and the recorded data reflects that split clearly. The Radeon 860M is an integrated graphics processor built into the Krackan Point chip, designed for portable devices where space and power draw are tightly constrained. Its benchmark results place it in the 72nd percentile among all GPUs in the database, with an average score of 26,401 across its two recorded tests. The Arc A380E, by contrast, is a discrete single-slot card with its own 6 GB of GDDR6 memory, a 75 W TDP, and a 50th percentile ranking. The database contains no benchmark scores for the Arc A380E, so its wins cannot be quantified numerically; the comparison relies entirely on the architectural and specification data recorded for both parts.

Where the Radeon 860M wins is in efficiency and integration. It uses system-shared memory, requires no power connectors, and carries a 15 W TDP, making it suitable for thin-and-light portable designs. Its boost clock reaches 3000 MHz, and it delivers 3.072 TFLOPS of FP32 compute. The Arc A380E, with its 2000 MHz fixed clock and 4.096 TFLOPS FP32, is a larger, hungrier part. The Radeon also wins on process technology: it uses a 4 nm node from TSMC, while the Arc A380E uses a 6 nm node from the same foundry. The Radeon's production status is Active, whereas the Arc A380E is marked End-of-life with a successor named Battlemage.

The Arc A380E wins on raw specification counts. It has 1024 shading units, 64 texture mapping units, and 32 render output units, double the Radeon's 512, 32, and 16 respectively. Its pixel rate is 64.00 GPixel/s versus 48.00 GPixel/s, and its texture rate is 128.0 GTexel/s versus 96.00 GTexel/s. Its memory subsystem is dedicated and fast: 6 GB of GDDR6 on a 96-bit bus with 186.0 GB/s of bandwidth, compared to the Radeon's system-dependent shared memory. The Arc A380E also has a fixed 1937 MHz memory clock at 15.5 Gbps effective, and its FP16 throughput doubles to 8.192 TFLOPS, whereas the Radeon maintains a 1:1 ratio at 3.072 TFLOPS. The Arc A380E is also a physical card with four DisplayPort 2.0 outputs, while the Radeon's display outputs are portable-device dependent.

Architecture Differences

The two GPUs come from different architectural lineages. The Radeon 860M is built on RDNA 3.5, part of the Navi III IGP generation for Strix Point Mobile. Its chip is Krackan Point, fabricated on a 4 nm TSMC process. The Arc A380E uses Intel's Xe-HPG architecture, specifically the DG2-128 chip from the Alchemist generation, also known as Arc 3. It uses a 6 nm TSMC process. The Radeon's transistor count and die size are recorded as unknown in the database; the Arc A380E is fully specified with 7,200 million transistors on a 157 mm² die, giving a transistor density of 45.9 million per square millimeter.

The Radeon 860M has no dedicated tensor cores listed, and neither does the Arc A380E. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. Both have 8 ray tracing cores. The Radeon uses PCIe 4.0 x8, and the Arc A380E also uses PCIe 4.0 x8. The Radeon's base clock is 600 MHz with a boost of 3000 MHz, a 5x boost ratio that reflects its power-scaled integrated design. The Arc A380E runs at a flat 2000 MHz for both base and boost, typical of a discrete card with a fixed power envelope.

Memory architecture is the largest divergence. The Radeon shares system memory with the host CPU, and its bandwidth is listed as system dependent, meaning performance scales with the platform's RAM configuration. The Arc A380E has its own 6 GB of GDDR6, a 96-bit bus, and 186.0 GB/s of fixed bandwidth. The Radeon has no memory clock listed beyond system shared; the Arc has a specific 1937 MHz memory clock. The Radeon's FP16 and FP32 both sit at 3.072 TFLOPS, while the Arc's FP16 doubles to 8.192 TFLOPS, indicating a 2:1 FP16 rate versus the Radeon's 1:1.

Physical differences are stark. The Radeon is an IGP with no slot width, no power connectors, and no dimensions recorded. The Arc A380E is a single-slot card measuring 254 mm by 127 mm by 20 mm, with no power connectors but a suggested PSU of 250 W. The Radeon's release date is February 28, 2025, with a predecessor of Navi II IGP. The Arc A380E was released March 31, 2024, with a predecessor of Xe Graphics and a successor of Battlemage. The Radeon remains active; the Arc is end-of-life.

Head-to-Head Benchmarks

The database records two benchmark scores for the AMD Radeon 860M. In Geekbench OpenCL, it scores 22,759. In Geekbench Vulkan, it scores 30,043. Its average benchmark score is 26,401. The Intel Arc A380E has no benchmark entries in the database; its average score is recorded as zero, and it has no nearest rivals listed. The head-to-head benchmark table is empty, with zero wins recorded for either side.

Because the Arc A380E lacks benchmark data, the comparison must rely on the Radeon's nearest rivals to contextualize its performance. The Radeon's average score of 26,401 sits within 1.1% of several discrete GPUs. The NVIDIA GeForce MX550 averages 26,421, which is 0.1% ahead of the Radeon. The NVIDIA GeForce RTX 5060 averages 26,331, which is 0.3% behind. The AMD Radeon RX 5700 XT 50th Anniversary averages 26,553, 0.6% ahead. The NVIDIA RTX A4000 averages 26,683, 1.1% ahead. These deltas are small, indicating that the integrated Radeon 860M lands in the same performance band as these discrete cards in the database's average scoring.

The Radeon's two individual scores show a notable split. Its Vulkan score of 30,043 is roughly 32% higher than its OpenCL score of 22,759. This suggests the RDNA 3.5 architecture responds differently to the two APIs, with Vulkan extracting more of the available hardware throughput. The average of the two, 26,401, masks that divergence. The Arc A380E's raw specification advantages, 4.096 TFLOPS FP32 versus 3.072 TFLOPS, would suggest it should outperform the Radeon in compute-heavy workloads, but without recorded benchmark results, that inference cannot be confirmed from the database.

The Arc A380E's memory bandwidth of 186.0 GB/s is a fixed, dedicated figure, while the Radeon's is system dependent. In a platform with high-speed shared memory, the Radeon could approach or exceed that bandwidth, but the database records no such configuration. The Arc's 6 GB GDDR6 capacity is also fixed, versus the Radeon's shared pool that varies by system. These are structural differences that would surface in texture-heavy or bandwidth-limited workloads, but the absence of Arc benchmark data prevents a definitive numerical comparison.

The Verdict

The recorded data supports a clear split by use case. For portable devices where power draw is critical, the AMD Radeon 860M is the only viable option. Its 15 W TDP, IGP form factor, and lack of power connectors align with thin-and-light designs. It is an active product with a 2025 release date. Its benchmark scores place it in the 72nd percentile, and it trades nearly evenly with discrete GPUs like the GeForce MX550 and RTX 5060, within 0.3% or less. That is a strong result for an integrated part.

For systems that need a dedicated card with its own memory, the Intel Arc A380E offers the full specification sheet. It doubles the Radeon's shading units, TMUs, and ROPs. It has 6 GB of dedicated GDDR6 with 186.0 GB/s bandwidth, a fixed 2000 MHz clock, and 4.096 TFLOPS FP32. It supports four DisplayPort 2.0 outputs and fits in a single slot. However, it is end-of-life, its successor is Battlemage, and it carries a 75 W TDP with a suggested 250 W PSU.

The database records no benchmark scores for the Arc A380E, so its actual performance cannot be verified against the Radeon's measured results. The Radeon's nearest rivals, all discrete GPUs with average scores within 1.1%, demonstrate that integrated graphics can now match older or lower-tier discrete cards. The Arc A380E's specifications suggest it should outcompute the Radeon, but the absence of recorded data means that conclusion remains unproven. The verdict from the data: choose the Radeon 860M for integrated efficiency and verified benchmark performance; choose the Arc A380E only if the discrete form factor, fixed memory, and full display output configuration are mandatory requirements.

FAQ

Q: What is the AMD Radeon 860M's average benchmark score?

A: The Radeon 860M has an average benchmark score of 26,401, based on a Geekbench OpenCL score of 22,759 and a Geekbench Vulkan score of 30,043.

Q: How does the Radeon 860M compare to its nearest rival, the NVIDIA GeForce MX550?

A: The GeForce MX550 has an average score of 26,421, which is 0.1% higher than the Radeon 860M's 26,401, making the two effectively tied in the database's average scoring.

Q: What memory configuration does the Intel Arc A380E use?

A: The Arc A380E uses 6 GB of GDDR6 memory on a 96-bit bus with 186.0 GB/s of bandwidth, at a memory clock of 1937 MHz (15.5 Gbps effective).

Q: What is the process node difference between the two GPUs?

A: The AMD Radeon 860M is fabricated on a 4 nm TSMC process, while the Intel Arc A380E uses a 6 nm TSMC process.

Q: Does the Intel Arc A380E have any recorded benchmark scores?

A: No, the database lists no benchmark entries for the Arc A380E, and its average benchmark score is recorded as zero.

Q: What are the TDP and slot requirements for each GPU?

A: The Radeon 860M has a 15 W TDP and is an IGP with no slot width. The Arc A380E has a 75 W TDP, is single-slot, and has a suggested PSU of 250 W.