AMD Radeon 890M vs Intel Arc A380 Comparison

The AMD Radeon 890M and Intel Arc A380 are closely matched competitors in the low-end GPU space, but the data reveals a clear split: the 890M wins 7 of 10 head-to-head benchmarks, including dominant victories in DirectX 11 and compute workloads, while the Arc A380 takes the newer 3DMark Steel Nomad test and holds a slim edge in OpenCL. The 890M’s average benchmark score of 9210 places it 7.6% ahead of the A380’s 8558, and it sits at the 45th percentile versus the A380’s 44th. However, the A380 is a dedicated dual-slot card with 6 GB of dedicated GDDR6 memory, whereas the 890M is an integrated processor with system-shared memory—a fundamental difference that dictates which users should choose each.

The Verdict

Choose the AMD Radeon 890M for integrated efficiency and legacy API strength. The data shows it wins the majority of tested workloads, particularly in older DirectX titles and general compute. Its PassMark G3D score of 8076 is 29.2% higher than the A380’s 6252, and its PassMark GPU Compute score of 4157 is 50.5% higher. The 890M also wins PassMark DirectX 11 by a massive 92.1% (73 vs 38) and DirectX 9 by 32.9% (97 vs 73). With a 15 W TDP versus the A380’s 75 W, it is the clear pick for thin-and-light laptops or systems where power draw and heat are primary constraints.

Choose the Intel Arc A380 for dedicated memory and newer API performance. Despite losing the overall benchmark war, the A380 wins the most modern test—3DMark Steel Nomad DX12—scoring 808 against the 890M’s 590, a 27% advantage. It also edges out the 890M in Geekbench OpenCL (38224 vs 37254, a 2.5% lead) and PassMark DirectX 10 (37 vs 33, a 10.8% lead). The A380’s 6 GB of GDDR6 memory with 186.0 GB/s bandwidth is system-independent, making it the better choice for desktops where the 890M’s system-shared memory could bottleneck performance. Its launch MSRP is 149 USD.

Architecture Differences

The two GPUs come from opposite design philosophies. The 890M is built on TSMC’s 4 nm process and packs 34,000 million transistors into a 233 mm² die, yielding a transistor density of 145.9M per mm². The A380 uses TSMC’s 6 nm process with 7,200 million transistors on a 157 mm² die, for a density of 45.9M per mm²—meaning the 890M carries nearly five times more transistors per area. The 890M uses AMD’s RDNA 3.5 architecture from the Strix Point chip, designated as a Navi III IGP, while the A380 uses Intel’s Xe-HPG architecture from the DG2-128 chip, part of the Alchemist (Arc 3) generation.

Both GPUs share identical counts of 1024 shading units, 64 TMUs, and 32 ROPs, but they diverge in ray tracing and compute resources. The 890M has 16 RT cores, double the A380’s 8. The A380, however, has no tensor cores listed, and both lack dedicated tensor hardware. Clock speeds differ dramatically: the 890M boosts to 2900 MHz from a 400 MHz base, while the A380 runs at a 2000 MHz base and 2050 MHz boost. The 890M’s higher boost clock and newer node help it achieve 5.939 TFLOPS FP32 performance versus the A380’s 4.198 TFLOPS. In FP16, the 890M delivers 5.939 TFLOPS (1:1 ratio), while the A380 outputs 8.397 TFLOPS (2:1 ratio), reflecting Intel’s different compute path.

Head-to-Head Benchmarks

The biggest win for the 890M is in PassMark DirectX 11, where it scores 73 versus the A380’s 38—a 92.1% lead. This is the single largest margin in the entire comparison and suggests the 890M is far better optimized for legacy DirectX 11 titles. The 890M also dominates in PassMark G2D (979 vs 610, a 60.5% advantage), indicating faster 2D rendering and desktop composition tasks. Its PassMark GPU Compute score of 4157 crushes the A380’s 2762 by 50.5%, making it the stronger choice for OpenCL-style compute workloads. In PassMark DirectX 9, the 890M leads 97 to 73 (32.9% higher), and it wins PassMark DirectX 12 narrowly at 37 vs 35 (5.7% higher). The 890M also takes Geekbench Vulkan with 40808 points against the A380’s 36736, an 11.1% margin.

The A380’s wins are fewer but concentrated in modern or vendor-specific tests. Its 3DMark Steel Nomad DX12 score of 808 beats the 890M’s 590 by 27%, showing a clear edge in the latest DirectX 12 rendering workloads. In Geekbench OpenCL, the A380 scores 38224 versus the 890M’s 37254, a 2.5% lead. It also wins PassMark DirectX 10 with 37 points to the 890M’s 33, a 10.8% margin. These wins suggest the A380 handles newer API paths better, even if its overall throughput is lower.

Specification Differences

The most significant spec difference is memory. The 890M uses system-shared memory with system-dependent bandwidth, while the A380 has 6 GB of GDDR6 on a 96-bit bus delivering a fixed 186.0 GB/s. The 890M’s memory clock is listed as "System Shared," versus the A380’s 1937 MHz (15.5 Gbps effective). Power requirements also differ fundamentally: the 890M is an IGP with a 15 W TDP and no power connectors, while the A380 is a dual-slot card with a 75 W TDP, one 8-pin connector, and a suggested 250 W PSU. The A380’s dimensions are 222 mm long, 114 mm high, and 42 mm wide; the 890M has no listed dimensions as it is integrated. Display outputs differ: the 890M is portable-device dependent, while the A380 offers 1x HDMI 2.1 and 3x DisplayPort 2.0. The 890M uses a 4 nm process versus the A380’s 6 nm, and its release date is 2024-07-14 versus the A380’s 2022-06-13. The 890M is still active in production, while the A380 is end-of-life with its successor listed as Battlemage.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The AMD Radeon 890M has an average benchmark score of 9210, which is 7.6% higher than the Intel Arc A380’s 8558.

Q: Does the Intel Arc A380 win any benchmarks?

A: Yes, the A380 wins 3 of 10 head-to-head tests: 3DMark Steel Nomad DX12 (808 vs 590, a 27% lead), Geekbench OpenCL (38224 vs 37254, a 2.5% lead), and PassMark DirectX 10 (37 vs 33, a 10.8% lead).

Q: How do the two compare in ray tracing hardware?

A: The AMD Radeon 890M has 16 RT cores, which is exactly double the Intel Arc A380’s 8 RT cores.

Q: What is the memory configuration difference?

A: The 890M uses system-shared memory with system-dependent bandwidth, while the A380 has 6 GB of dedicated GDDR6 memory on a 96-bit bus with 186.0 GB/s bandwidth.

Q: Which GPU is more power-efficient?

A: The 890M has a 15 W TDP and is an integrated processor with no power connectors, while the A380 is a dual-slot card with a 75 W TDP and requires one 8-pin power connector.

Q: What is the DirectX 11 performance gap?

A: The 890M leads by 92.1% in PassMark DirectX 11, scoring 73 versus the A380’s 38.

Where Each One Wins

The AMD Radeon 890M wins in seven benchmark categories, making it the better overall performer for most workloads. Its 50.5% lead in PassMark GPU Compute and 60.5% lead in PassMark G2D indicate superiority in general-purpose compute and 2D rendering. The 92.1% DirectX 11 win and 32.9% DirectX 9 win make it the definitive choice for older game libraries and legacy software. Its 11.1% Vulkan advantage also suggests better performance in Vulkan-based titles. The 890M’s 15 W TDP and IGP form factor mean it fits into ultraportable systems with no discrete power requirements.

The Intel Arc A380 wins where the 890M cannot compete. Its 27% lead in 3DMark Steel Nomad DX12 shows it is better equipped for the latest DirectX 12 rendering techniques. The 2.5% OpenCL edge and 10.8% DirectX 10 win round out its modern API portfolio. The A380’s dedicated 6 GB GDDR6 memory with 186.0 GB/s bandwidth provides consistent performance independent of system RAM, making it more reliable for sustained gaming sessions on desktop systems. Its 75 W TDP and dual-slot design require a PSU rated at 250 W, but it offers fixed display outputs (1x HDMI 2.1, 3x DisplayPort 2.0) rather than the 890M’s portable-device-dependent outputs. The A380 is the pick for users who prioritize the newest DirectX 12 games and need dedicated VRAM, while the 890M is the all-around winner for compute, legacy APIs, and power-constrained mobile designs.