AMD Instinct MI350X vs Intel Arc Graphics 128EU Mobile Comparison
AMD Instinct MI350X
Arc Graphics 128EU Mobile
Analysis: AMD Instinct MI350X vs Intel Arc Graphics 128EU Mobile
Head-to-Head Benchmarks
The recorded database contains no direct head-to-head benchmark results for the AMD Instinct MI350X and the Intel Arc Graphics 128EU Mobile. The head-to-head benchmark array is empty, and both products show zero wins in the comparative analysis. This absence of measured data reflects the fundamental positioning of these two devices: the MI350X is a compute-optimized accelerator module, while the Arc Graphics 128EU Mobile is an integrated graphics processor for portable systems.
Without direct comparative scores, the available specification data provides the only basis for performance differentiation. The MI350X delivers 72.09 TFLOPS of FP32 compute, compared to 4.608 TFLOPS for the Intel part. That represents a 15.6x advantage in single-precision floating-point throughput. In FP16, the MI350X maintains 72.09 TFLOPS with a 1:1 ratio, while the Arc part reaches 9.216 TFLOPS with a 2:1 ratio, leaving the MI350X with an 7.8x lead in half-precision compute.
Texture throughput follows a similar pattern. The MI350X achieves 2,252.8 GTexel/s against 144.0 GTexel/s for the Intel Arc, a 15.6x difference. The pixel rate is an outlier: the MI350X reports 0 MPixel/s due to the absence of ROPs, while the Arc Graphics 128EU Mobile delivers 72.00 GPixel/s. This reflects the MI350X's design as a pure compute accelerator with no rasterization pipeline, rather than a performance deficit in any traditional graphics sense.
The two products also differ dramatically in memory resources. The MI350X carries 288 GB of HBM3e memory across an 8192-bit bus, yielding 8.19 TB/s of bandwidth. The Arc Graphics 128EU Mobile uses system shared memory with bandwidth described as "System Dependent." The dedicated memory capacity and bandwidth of the MI350X are orders of magnitude beyond what an integrated GPU can access.
Clock behavior further separates the two. The MI350X has a base clock of 1000 MHz and a boost clock of 2200 MHz. The Arc part starts at 300 MHz base and boosts to 2250 MHz. Despite similar boost ceilings, the MI350X maintains a much higher floor, indicating sustained compute capability under load.
Where Each One Wins
The MI350X wins in every compute-intensive category where the data provides a measurable figure. FP32 performance, FP16 performance, texture rate, memory capacity, memory bandwidth, and memory bus width all favor the AMD accelerator by wide margins. The MI350X also wins on process technology: it uses a 3 nm TSMC node with 185,000 million transistors on a 2380 mm² die, achieving a transistor density of 77.7M per mm². The Intel part uses a 10 nm Intel process with no recorded transistor count or die size.
The Arc Graphics 128EU Mobile wins in the categories that matter for integrated graphics in portable devices. It has 32 ROPs, enabling actual pixel output, while the MI350X has zero. It reports a pixel rate of 72.00 GPixel/s, a functional display pipeline with DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4 support, and display outputs described as "Portable Device Dependent." The MI350X has no display outputs and lists its API support as N/A across the board.
The Intel part also wins on power efficiency in absolute terms. Its TDP is 28 W, while the MI350X draws 1000 W. The suggested PSU for the MI350X is 1400 W, a figure that underscores the scale of its power delivery requirements. The Arc Graphics 128EU Mobile has no suggested PSU listed, consistent with its integration into a mobile processor package.
The Arc part also holds an advantage in production status: it is marked "Active," while the MI350X has no production status recorded. The Arc part is newer in terms of release chronology only in the sense that it launched on 2023-12-13, while the MI350X has a release date of 2025-06-11, making the AMD part the more recent introduction despite the Intel product being listed as active.
Architecture Differences
The MI350X uses CDNA 4.0 architecture, built around the MI350 256CU chip. The Arc Graphics 128EU Mobile uses Xe-LPG architecture based on the Meteor Lake chip. These are fundamentally different design philosophies: CDNA 4.0 is a compute-focused architecture with no graphics pipeline, while Xe-LPG is a graphics architecture integrated into a mobile SoC.
The MI350X packs 16,384 shading units, 1,024 TMUs, and zero ROPs. The Arc part has 1,024 shading units, 64 TMUs, and 32 ROPs. The AMD design allocates all silicon to shading and texturing throughput, while the Intel design retains a full rasterization path. Neither product lists dedicated ray tracing cores or tensor cores in the database.
The MI350X uses a 3 nm process from TSMC, with 185,000 million transistors on a 2380 mm² die. The Arc Graphics 128EU Mobile uses a 10 nm Intel process, with no transistor or die size data recorded. The process difference explains the MI350X's density of 77.7M transistors per mm², a figure not calculable for the Intel part due to missing data.
Memory architecture is entirely different. The MI350X uses HBM3e with a 8192-bit bus and 288 GB capacity, achieving 8.19 TB/s. The Arc Graphics 128EU Mobile uses system shared memory, with bus width, type, and bandwidth all marked "System Shared" or "System Dependent." The MI350X's memory clock is 2000 MHz with 8 Gbps effective, while the Intel part has no dedicated memory clock.
The MI350X connects via PCIe 5.0 x16 and is formatted as an OAM Module with no power connectors. The Arc Graphics 128EU Mobile uses a Ring Bus interface and is an IGP. The AMD module measures 102 mm in length and 165 mm in width, while the Intel part has no recorded dimensions.
Specification Differences
| Specification | AMD Instinct MI350X | Intel Arc Graphics 128EU Mobile |
|---|---|---|
| Architecture | CDNA 4.0 | Xe-LPG |
| Process Node | 3 nm | 10 nm |
| Foundry | TSMC | Intel |
| Transistors | 185,000 million | Not recorded |
| Die Size | 2380 mm² | Not recorded |
| Base Clock | 1000 MHz | 300 MHz |
| Boost Clock | 2200 MHz | 2250 MHz |
| Memory Size | 288 GB | System Shared |
| Memory Type | HBM3e | System Shared |
| Memory Bus | 8192 bit | System Shared |
| Memory Bandwidth | 8.19 TB/s | System Dependent |
| Shading Units | 16,384 | 1,024 |
| TMUs | 1,024 | 64 |
| ROPs | 0 | 32 |
| Pixel Rate | 0 MPixel/s | 72.00 GPixel/s |
| Texture Rate | 2,252.8 GTexel/s | 144.0 GTexel/s |
| FP32 | 72.09 TFLOPS | 4.608 TFLOPS |
| FP16 | 72.09 TFLOPS (1:1) | 9.216 TFLOPS (2:1) |
| TDP | 1000 W | 28 W |
| Slot Width | OAM Module | IGP |
| Power Connectors | None | Not recorded |
| Suggested PSU | 1400 W | Not recorded |
| Bus Interface | PCIe 5.0 x16 | Ring Bus |
| Display Outputs | No outputs | Portable Device Dependent |
| DirectX | N/A | 12 (12_1) |
| OpenGL | N/A | 4.6 |
| Vulkan | N/A | 1.4 |
| Release Date | 2025-06-11 | 2023-12-13 |
| Production Status | Not recorded | Active |
| Predecessor | Radeon Instinct | HD Graphics-M |
FAQ
Q: Which product has higher FP32 compute performance?
A: The AMD Instinct MI350X delivers 72.09 TFLOPS of FP32, compared to 4.608 TFLOPS for the Intel Arc Graphics 128EU Mobile, a 15.6x advantage.
Q: How do the memory subsystems compare?
A: The MI350X uses 288 GB of HBM3e with an 8192-bit bus and 8.19 TB/s bandwidth. The Arc Graphics 128EU Mobile uses system shared memory with bandwidth described as "System Dependent."
Q: Can the MI350X output video to a display?
A: No. The MI350X has no display outputs and lists its APIs as N/A. The Arc Graphics 128EU Mobile has display outputs described as "Portable Device Dependent" and supports DirectX 12 (12_1), OpenGL 4.6, and Vulkan 1.4.
Q: What is the power consumption difference?
A: The MI350X has a TDP of 1000 W with a suggested PSU of 1400 W. The Arc Graphics 128EU Mobile has a TDP of 28 W and no suggested PSU recorded.
Q: Which product has a smaller manufacturing process?
A: The MI350X uses a 3 nm TSMC process with 185,000 million transistors on a 2380 mm² die. The Arc Graphics 128EU Mobile uses a 10 nm Intel process with no transistor or die size data recorded.
Q: Do both products support ray tracing?
A: Neither product lists dedicated ray tracing cores in the database. The MI350X has no RT cores recorded, and the Arc Graphics 128EU Mobile also has no RT core count listed.