Intel Arc A380E vs Intel Arc Graphics 128EU Mobile Comparison

FAQ

Q: What are the core specifications of the Intel Arc A380E?

A: The Intel Arc A380E uses the DG2-128 chip on the Xe-HPG architecture, built on a 6 nm process at TSMC. It has 1024 shading units, 64 TMUs, 32 ROPs, 8 ray tracing cores, and a base and boost clock of 2000 MHz. It comes with 6 GB of GDDR6 memory on a 96-bit bus, delivering 186.0 GB/s of bandwidth. The card is rated for 75 W TDP, uses a PCIe 4.0 x8 interface, and has 4x DisplayPort 2.0 outputs.

Q: What are the core specifications of the Intel Arc Graphics 128EU Mobile?

A: The Intel Arc Graphics 128EU Mobile is based on the Meteor Lake chip with Xe-LPG architecture, built on a 10 nm process at Intel. It has the same 1024 shading units, 64 TMUs, and 32 ROPs as the A380E, but no dedicated ray tracing cores are listed. Its base clock is 300 MHz with a boost of 2250 MHz. Memory is system shared, so size, type, bus width, and bandwidth are system dependent. The TDP is 28 W, and it connects via a ring bus.

Q: How do the two compare in raw compute performance?

A: The Intel Arc Graphics 128EU Mobile has a higher peak FP32 throughput at 4.608 TFLOPS versus 4.096 TFLOPS for the Intel Arc A380E. The mobile part also leads in texture rate (144.0 GTexel/s versus 128.0 GTexel/s) and pixel rate (72.00 GPixel/s versus 64.00 GPixel/s), driven by its higher boost clock of 2250 MHz compared to the A380E's fixed 2000 MHz.

Q: What are the key differences in memory configuration?

A: The Intel Arc A380E has dedicated 6 GB of GDDR6 memory on a 96-bit bus with 186.0 GB/s bandwidth. The Intel Arc Graphics 128EU Mobile uses system shared memory, meaning its size, type, bus width, and bandwidth are all dependent on the host system. This gives the A380E a fixed, predictable memory subsystem while the mobile part relies on shared system resources.

Q: How do their API and feature support differ?

A: The Intel Arc A380E supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The Intel Arc Graphics 128EU Mobile supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4. The A380E has the higher DirectX feature level (12_2 versus 12_1), which reflects its dedicated ray tracing hardware.

Q: What is the production status and release timing for each?

A: The Intel Arc A380E is listed as end-of-life, with a release date of 2024-03-31. The Intel Arc Graphics 128EU Mobile is active, with a release date of 2023-12-13.

The Verdict

The benchmark data presents a clear split. The Intel Arc A380E is a dedicated, end-of-life discrete graphics card aimed at embedded or small-form-factor systems with its own 6 GB GDDR6 memory, a 96-bit bus, and fixed 186.0 GB/s bandwidth. It also brings DirectX 12 Ultimate support with 8 ray tracing cores, making it the pick for applications that require dedicated VRAM and hardware ray tracing capabilities.

The Intel Arc Graphics 128EU Mobile is an active integrated GPU within the Meteor Lake chip, designed for laptops and portable devices. It uses system shared memory, has a lower 28 W TDP, and achieves higher peak clocks (2250 MHz boost versus 2000 MHz). It delivers slightly higher raw FP32, texture, and pixel throughput in the recorded data, but its memory performance is entirely system dependent.

For users with a fixed system where memory bandwidth is critical, the A380E is the only choice due to its dedicated GDDR6. For portable devices where power efficiency and integration matter, the 128EU Mobile is the active, shipping option. The A380E's end-of-life status means it is a legacy part, while the 128EU Mobile remains in production.

Head-to-Head Benchmarks

The recorded data does not include any head-to-head benchmark scores. Neither part has a listed average benchmark score, and the head-to-head benchmark table is empty. Wins for each part are zero.

What the specifications do show is a direct computation of peak rates. The Intel Arc Graphics 128EU Mobile leads in FP32 throughput at 4.608 TFLOPS, which is 12.5% higher than the A380E's 4.096 TFLOPS. Its FP16 output of 9.216 TFLOPS (2:1) similarly exceeds the A380E's 8.192 TFLOPS (2:1) by the same margin. Texture rate on the mobile part is 144.0 GTexel/s versus 128.0 GTexel/s, a 12.5% advantage. Pixel rate is 72.00 GPixel/s versus 64.00 GPixel/s, also a 12.5% lead.

The Intel Arc A380E counters with a fixed memory bandwidth of 186.0 GB/s over a 96-bit GDDR6 bus. The mobile part's bandwidth is listed as system dependent, so no direct comparison number exists. The A380E also has a higher base clock (2000 MHz versus 300 MHz), though the mobile part boosts to 2250 MHz.

The A380E includes 8 ray tracing cores, which the mobile part lacks entirely. In API support, the A380E reaches DirectX 12 Ultimate (12_2), while the mobile part is limited to DirectX 12 (12_1). These are structural advantages that no clock speed or TFLOPS figure can offset.

Specification Differences

The two GPUs differ across nearly every measured field. Process node: the A380E uses 6 nm at TSMC, the mobile part uses 10 nm at Intel. Transistor count and die size are only listed for the A380E: 7,200 million transistors on a 157 mm² die, giving a density of 45.9M per mm². The mobile part has no listed transistor or die size data.

Clock behavior diverges sharply. The A380E runs a fixed 2000 MHz base and boost. The mobile part has a 300 MHz base and 2250 MHz boost, indicating a wide dynamic range for power management. Memory clocks differ: the A380E has a memory clock of 1937 MHz (15.5 Gbps effective), while the mobile part has no dedicated memory clock, using system shared memory instead.

Memory size, type, bus width, and bandwidth all differ. The A380E has 6 GB GDDR6 on a 96-bit bus with 186.0 GB/s. The mobile part lists all four as system shared or system dependent.

TDP: 75 W for the A380E versus 28 W for the mobile part. Slot width: single-slot for the A380E, IGP for the mobile part. Power connectors: none for the A380E, not listed for the mobile part. Suggested PSU: 250 W for the A380E, not listed for the mobile part.

Bus interface: PCIe 4.0 x8 for the A380E, ring bus for the mobile part. Display outputs: 4x DisplayPort 2.0 for the A380E, portable device dependent for the mobile part.

DirectX support: 12 Ultimate (12_2) for the A380E, 12 (12_1) for the mobile part. OpenGL and Vulkan are identical at 4.6 and 1.4 respectively.

Physical dimensions are only listed for the A380E: 254 mm length, 127 mm height, 20 mm width. The mobile part has none, as it is an integrated GPU.

Production status: end-of-life for the A380E, active for the mobile part. Release dates: 2024-03-31 for the A380E, 2023-12-13 for the mobile part. Predecessors differ: Xe Graphics for the A380E, HD Graphics-M for the mobile part. The A380E has a successor listed as Battlemage; the mobile part has no successor.

Architecture Differences

The Intel Arc A380E is built on the Xe-HPG architecture, specifically the DG2-128 chip, belonging to the Alchemist (Arc 3) generation. It uses TSMC's 6 nm process with 7,200 million transistors on a 157 mm² die. This architecture includes 8 dedicated ray tracing cores and supports DirectX 12 Ultimate (12_2), which is the full feature set for hardware ray tracing and mesh shading.

The Intel Arc Graphics 128EU Mobile is built on the Xe-LPG architecture, using the Meteor Lake chip, part of the Arc Graphics-M (Meteor Lake) generation. It uses Intel's 10 nm process. No transistor count, die size, or ray tracing core count is listed. Its DirectX support stops at 12 (12_1), which omits the DirectX 12 Ultimate features that require dedicated ray tracing hardware.

Both share the same shading unit count (1024), TMU count (64), and ROP count (32). The key architectural difference is the presence of ray tracing hardware in the A380E and its absence in the mobile part. The mobile part compensates with a higher boost clock (2250 MHz versus 2000 MHz), which explains its higher peak FP32, texture, and pixel rates.

The memory architecture is fundamentally different. The A380E uses a dedicated GDDR6 interface with a 96-bit bus, giving it deterministic bandwidth. The mobile part integrates with the system's shared memory pool, making its bandwidth dependent on the host platform's memory configuration. The A380E also uses a PCIe 4.0 x8 interface, while the mobile part connects via a ring bus, reflecting its position as an integrated GPU.

The process node difference (6 nm versus 10 nm) and the TDP difference (75 W versus 28 W) show the trade-off between a discrete card with dedicated memory and an integrated part designed for power-constrained mobile systems.

Where Each One Wins

The Intel Arc A380E wins in scenarios that require dedicated, fixed memory resources. Its 6 GB GDDR6 with 186.0 GB/s bandwidth is not dependent on system memory configuration, making it suitable for workloads where memory bandwidth consistency is essential. It also wins in any application that uses DirectX 12 Ultimate features, as it has 8 ray tracing cores and supports the 12_2 feature level. The A380E has a fixed 2000 MHz clock, which means performance is predictable without reliance on boost behavior. It is also the only one of the two with a listed successor (Battlemage), indicating a defined product lineage.

The Intel Arc Graphics 128EU Mobile wins in raw peak compute metrics. Its 4.608 TFLOPS FP32, 9.216 TFLOPS FP16, 144.0 GTexel/s texture rate, and 72.00 GPixel/s pixel rate are all higher than the A380E's corresponding figures. This is a direct result of its 2250 MHz boost clock versus the A380E's 2000 MHz. The mobile part also wins on power efficiency in terms of TDP: 28 W versus 75 W, making it the appropriate choice for battery-powered or thermally constrained devices. Its active production status and newer release date (2023-12-13 versus 2024-03-31) indicate it is the currently supported part, while the A380E is end-of-life.

For a system builder selecting a GPU, the choice depends on the form factor and workload. The A380E is a discrete, single-slot card with its own memory and ray tracing support, suited for embedded systems or small desktops where a fixed memory pool and DirectX 12 Ultimate features matter. The 128EU Mobile is an integrated part for laptops and portable devices, offering higher peak compute rates and much lower power draw, at the cost of shared memory and no hardware ray tracing. The recorded data gives no benchmark scores for either, so real-world performance comparisons beyond the specified peak rates are not available in the database.