Intel Arc A380E vs Intel Arc Graphics 64EU Mobile Comparison

Intel Arc A380E and Intel Arc Graphics 64EU Mobile represent two distinct approaches to Intel’s graphics architecture, one as a discrete add-in card and the other as an integrated solution within a mobile processor. The recorded data shows a clear separation in raw compute capability, memory configuration, and intended deployment, yet both share a common Xe-based foundation. The following analysis draws exclusively from database measurements and specification records.

Head-to-Head Benchmarks

The benchmark result set contains no direct head-to-head scores for these two parts. The database records zero wins for either product in direct comparison, and the average benchmark score for both entries is zero. The percentile ranking against all GPUs is identical at 50 for each, which places both in the middle of the historical distribution, but this percentile does not reflect a direct comparison between them. Without recorded performance numbers, the comparison must rely on the measured specifications that determine compute throughput.

The FP32 floating-point rate shows a substantial gap. The Arc A380E delivers 4.096 TFLOPS, while the Arc Graphics 64EU Mobile produces 1.792 TFLOPS. That is a 2.29x difference in single-precision compute. The FP16 rates follow the same 2:1 ratio pattern, with the A380E at 8.192 TFLOPS and the mobile part at 3.584 TFLOPS. The A380E has 1024 shading units versus 512 on the mobile part, exactly double. Texture units are 64 versus 32, and ROPs are 32 versus 16. Every execution resource on the discrete card is twice the count of the integrated part.

The pixel fill rate reflects the ROP difference. The A380E reaches 64.00 GPixel/s, while the mobile part manages 28.00 GPixel/s. Texture fill rate is 128.0 GTexel/s versus 56.00 GTexel/s. These are direct arithmetic consequences of the doubled unit counts and the clock speeds. The A380E runs at a fixed 2000 MHz base and boost, while the mobile part has a 300 MHz base and 1750 MHz boost. The sustained clock on the discrete card is 14.3% higher than the mobile boost clock, which further widens the throughput gap beyond the pure unit-count advantage.

Memory is another major divider. The A380E uses 6 GB of GDDR6 on a 96-bit bus, delivering 186.0 GB/s of bandwidth. The mobile part uses system shared memory with no dedicated VRAM, and the database records its bandwidth as system dependent. Shared memory bandwidth is not specified numerically, but the architecture implies dependence on the host memory subsystem. The discrete card’s memory clock is 1937 MHz, or 15.5 Gbps effective, which provides a fixed and dedicated resource. The integrated part has no such dedicated allocation.

Ray tracing hardware exists only on the A380E, with 8 RT cores recorded. The mobile part lists no RT core count. This creates a functional difference in workloads that use hardware-accelerated ray tracing. The A380E also supports DirectX 12 Ultimate (12_2), while the mobile part is limited to DirectX 12 (12_1). That feature-level difference matters for certain rendering paths.

The Verdict

The data indicates that the Arc A380E is the stronger performer on every measured compute and memory specification. It has double the shading units, double the texture units, double the ROPs, more than double the FP32 throughput, dedicated GDDR6 memory with 186.0 GB/s bandwidth, and hardware ray tracing. The mobile part’s only advantages are a lower 65 W TDP versus 75 W, a smaller physical footprint as an IGP, and an active production status versus the A380E’s end-of-life status.

The database shows no benchmark scores, so the verdict rests on specification-derived capability. The A380E is designed for sustained operation at 2000 MHz with dedicated memory, which suits workloads that need consistent throughput. The mobile part operates at a 300 MHz base clock and relies on shared system memory, which makes its performance dependent on the host platform. For any task that stresses GPU compute, the A380E is the clear choice from the recorded data.

The mobile part’s active production status and newer release date, December 2023 versus March 2024, indicate a longer expected availability. The A380E is end-of-life, so procurement for new systems may favor the mobile part despite its lower performance. The mobile part also consumes less power, 65 W versus 75 W, which matters in thermally constrained environments.

Architecture Differences

The two products use different Xe architectures. The A380E is built on Xe-HPG, the high-performance gaming architecture, while the mobile part uses Xe-LPG, a low-power variant. The A380E’s chip is DG2-128, part of the Alchemist generation under the Arc 3 branding. The mobile part uses the Meteor Lake chip and belongs to the Arc Graphics-M generation.

The manufacturing process differs. The A380E uses a 6 nm process at TSMC, with 7,200 million transistors on a 157 mm² die. That yields a transistor density of 45.9 million per square millimeter. The mobile part uses a 10 nm process at Intel, but the database records no transistor count, die size, or density. The process node difference suggests the A380E has a denser and more modern fabrication, though the lack of mobile die data prevents a full comparison.

Memory architecture is fundamentally different. The A380E has dedicated GDDR6 memory with a 96-bit bus and a fixed 186.0 GB/s bandwidth. The mobile part uses system shared memory, with the type, bus width, and bandwidth all recorded as system dependent. This means the mobile part’s memory performance varies with the host laptop’s RAM configuration, while the A380E’s memory performance is fixed.

The bus interface also differs. The A380E connects via PCIe 4.0 x8, a standard discrete card interface. The mobile part uses a Ring Bus, which ties it to the processor’s internal fabric. The display outputs reflect this split: the A380E provides 4x DisplayPort 2.0, while the mobile part’s outputs are portable device dependent.

The A380E has 8 RT cores and supports DirectX 12 Ultimate (12_2). The mobile part has no recorded RT core count and supports DirectX 12 (12_1). Both support OpenGL 4.6 and Vulkan 1.4. The mobile part’s predecessor is HD Graphics-M, while the A380E’s predecessor is Xe Graphics and its successor is Battlemage.

FAQ

Q: Which GPU has higher FP32 compute?

A: The Intel Arc A380E delivers 4.096 TFLOPS, which is 2.29x the 1.792 TFLOPS of the Intel Arc Graphics 64EU Mobile.

Q: Does the mobile part have dedicated video memory?

A: No. The Arc Graphics 64EU Mobile uses system shared memory, and the database records its memory type, bus width, and bandwidth as system dependent.

Q: Which product supports hardware ray tracing?

A: The A380E has 8 RT cores. The Arc Graphics 64EU Mobile lists no RT core count in the database, so no hardware ray tracing capability is recorded for it.

Q: What are the power requirements for each?

A: The A380E has a 75 W TDP and a suggested PSU of 250 W. The mobile part has a 65 W TDP and no suggested PSU recorded.

Q: Which product is still in production?

A: The Arc Graphics 64EU Mobile is marked as active. The Arc A380E is marked as end-of-life.

Q: What is the DirectX support difference?

A: The A380E supports DirectX 12 Ultimate (12_2), while the mobile part supports DirectX 12 (12_1).

Where Each One Wins

The A380E wins in every computational category. Its FP32 rate of 4.096 TFLOPS versus 1.792 TFLOPS makes it the choice for general compute workloads. The 186.0 GB/s dedicated bandwidth versus system dependent shared memory gives it a fixed memory performance advantage. The 8 RT cores and DirectX 12 Ultimate support make it suitable for ray-traced rendering, which the mobile part cannot match. The 64.00 GPixel/s pixel rate versus 28.00 GPixel/s indicates faster rasterization output. The doubled texture units, 64 versus 32, give it higher texturing throughput at 128.0 GTexel/s versus 56.00 GTexel/s. The 2000 MHz sustained clock versus a 300 MHz base clock on the mobile part ensures consistent performance under load.

The mobile part wins in areas outside raw performance. It consumes less power at 65 W versus 75 W. It is an IGP with no slot width, so it fits into laptop designs without a separate card. It uses a Ring Bus interface, which integrates with the processor. Its production status is active, meaning it remains available for new designs. Its release date of December 2023 predates the A380E’s March 2024 release, so it has a longer market presence. The mobile part’s display outputs are portable device dependent, which gives laptop manufacturers flexibility in configuring video outputs. The A380E’s display outputs are fixed at 4x DisplayPort 2.0.

For mobile computing, the Arc Graphics 64EU Mobile is the only viable option between the two, since the A380E is a discrete card with a 254 mm length, 127 mm height, and 20 mm width. The A380E requires PCIe 4.0 x8 and a 250 W suggested PSU, which are not available in typical laptops. The mobile part has no length, height, or width recorded, consistent with an integrated design.

Specification Differences

The process node differs: 6 nm TSMC for the A380E versus 10 nm Intel for the mobile part. The A380E has 7,200 million transistors on a 157 mm² die with a density of 45.9 million per square millimeter. The mobile part has no transistor, die size, or density data recorded.

Clock speeds differ significantly. The A380E runs at 2000 MHz base and 2000 MHz boost. The mobile part runs at 300 MHz base and 1750 MHz boost. The memory clock on the A380E is 1937 MHz, or 15.5 Gbps effective. The mobile part has no memory clock recorded.

Memory configuration differs. The A380E has 6 GB GDDR6 with a 96-bit bus and 186.0 GB/s bandwidth. The mobile part has system shared memory with system dependent type, bus width, and bandwidth.

Execution resources differ. The A380E has 1024 shading units, 64 TMUs, 32 ROPs, and 8 RT cores. The mobile part has 512 shading units, 32 TMUs, 16 ROPs, and no RT core count recorded.

Compute rates differ. The A380E produces 64.00 GPixel/s and 128.0 GTexel/s. The mobile part produces 28.00 GPixel/s and 56.00 GTexel/s. FP32 is 4.096 TFLOPS versus 1.792 TFLOPS, and FP16 is 8.192 TFLOPS versus 3.584 TFLOPS.

Power and physical attributes differ. The A380E has a 75 W TDP, single-slot width, no power connectors, and a 250 W suggested PSU. The mobile part has a 65 W TDP, IGP slot width, no power connector data, and no suggested PSU. The A380E measures 254 mm by 127 mm by 20 mm. The mobile part has no dimensions recorded.

Bus interface differs: PCIe 4.0 x8 for the A380E versus Ring Bus for the mobile part. Display outputs are 4x DisplayPort 2.0 for the A380E versus portable device dependent for the mobile part. DirectX support is 12 Ultimate (12_2) for the A380E versus 12 (12_1) for the mobile part. Both share OpenGL 4.6 and Vulkan 1.4.

Production status differs: end-of-life for the A380E versus active for the mobile part. Release dates are March 2024 for the A380E and December 2023 for the mobile part. The A380E’s predecessor is Xe Graphics and its successor is Battlemage. The mobile part’s predecessor is HD Graphics-M and it has no successor recorded. Neither part has a launch MSRP listed in the database.