AMD Radeon R9 M390X vs Intel Arc A350M Comparison

Head-to-Head Benchmarks

The recorded data contains one direct benchmark comparison between the Intel Arc A350M and the AMD Radeon R9 M390X. This test, Geekbench Vulkan, reveals a substantial performance gap. The Intel Arc A350M scores 24,747 points, while the AMD Radeon R9 M390X scores 16,951 points. This represents a 46% advantage for the Intel part, a decisive margin in any workload that leverages the Vulkan API.

Looking beyond the direct head-to-head, the average benchmark scores from the database place these two products in different performance tiers. The Intel Arc A350M achieves an average benchmark score of 24,647, which places it at the 70th percentile among all GPUs. The AMD Radeon R9 M390X, by contrast, has an average benchmark score of 20,662, sitting at the 66th percentile. The difference in average scores is roughly 19%, a substantial gap that appears consistently across the recorded results.

The Intel Arc A350M's nearest rivals in the database further contextualize its performance. It sits within 1.5% of the AMD Radeon RX 590, the NVIDIA RTX A5000 Mobile, the AMD Radeon RX 6600 XT, and the NVIDIA GeForce GTX 1630. Specifically, the Arc A350M trails the RX 590 by 0.4%, trails the RTX A5000 Mobile by 0.5%, leads the RX 6600 XT by 0.8%, and leads the GTX 1630 by 1.5%. This clustering indicates that the Arc A350M performs at a level comparable to a range of desktop and mobile discrete GPUs from previous generations.

The AMD Radeon R9 M390X, on the other hand, has nearest rivals that include the AMD Radeon RX 5600 XT, the Intel Arc A750, the Intel Arc B570, and the NVIDIA GeForce RTX 3070 Mobile. The R9 M390X trails the RX 5600 XT by 0.2%, leads the Arc A750 by 0.4%, leads the Arc B570 by 0.5%, and leads the RTX 3070 Mobile by 0.6%. These delta values are all extremely tight, suggesting the R9 M390X occupies a performance band that includes several well-known modern GPUs.

A point-by-point breakdown of wins across all recorded benchmark categories shows the Intel Arc A350M secures 1 win, while the AMD Radeon R9 M390X secures 0 wins. The Vulkan result is the only head-to-head test in the database, and it is the clearest signal for comparing the two directly.

Architecture Differences

The architectural divide between the Intel Arc A350M and the AMD Radeon R9 M390X is stark, starting with their underlying chip designs. The Intel part uses the DG2-128 chip, built on Intel's Xe-HPG architecture, which belongs to the Alchemist (Arc 3 Mobile) generation. AMD's chip, named Amethyst, is based on the older GCN 3.0 architecture and belongs to the Gem System (R9 M300) generation. Manufacturing is handled by TSMC for both chips, but the process nodes differ significantly: Intel uses a 6 nm process, while AMD uses a 28 nm process.

The transistor counts and die sizes reflect these generational differences. Intel's DG2-128 packs 7,200 million transistors on a die size of 157 mm², resulting in a transistor density of 45.9 million transistors per square millimeter. AMD's Amethyst die contains 5,000 million transistors on a much larger 366 mm² die, yielding a density of just 13.7 million transistors per square millimeter. The newer process node allows Intel to pack roughly 44% more transistors into less than half the die area.

Memory subsystems also diverge sharply. The Intel Arc A350M uses 4 GB of GDDR6 memory on a 64-bit bus, offering a bandwidth of 112.0 GB/s. The AMD Radeon R9 M390X also has 4 GB of memory, but it is GDDR5 on a 256-bit bus, delivering 160.0 GB/s. Despite the older memory type, the wider bus gives AMD a bandwidth advantage of roughly 43%. The memory clock differs as well: Intel's memory runs at 1750 MHz with 14 Gbps effective, while AMD's runs at 1250 MHz with 5 Gbps effective. The effective data rate is much higher on the Intel part, but the narrow bus still leaves it behind in total bandwidth.

Compute resources present a mixed picture. The AMD Radeon R9 M390X has 2,048 shading units, 128 texture mapping units, and 32 render output units. The Intel Arc A350M has 768 shading units, 48 TMUs, and 24 ROPs. AMD holds a 2.7x advantage in shading units, a 2.7x advantage in TMUs, and a 1.3x advantage in ROPs. However, Intel introduces hardware that AMD lacks entirely: 6 ray tracing cores. The R9 M390X has no ray tracing cores listed. This feature gap is significant for games that support ray tracing.

Clock speeds and resulting throughput metrics tell another side of the story. Intel lists a base clock of 1150 MHz and a boost clock of 2200 MHz. AMD lists no base or boost clock in the records, only the memory clock. In terms of pixel fill rate, Intel achieves 52.80 GPixel/s versus AMD's 23.14 GPixel/s, a 2.3x advantage. Texture fill rate is closer: Intel achieves 105.6 GTexel/s, AMD achieves 92.54 GTexel/s, a 14% advantage for Intel. FP32 compute favors Intel at 3.379 TFLOPS versus AMD's 2.961 TFLOPS, a 14% edge. FP16 performance shows the biggest architectural difference: Intel delivers 6.758 TFLOPS with a 2:1 ratio, while AMD delivers only 2.961 TFLOPS with a 1:1 ratio. Intel's FP16 throughput is 2.3x higher, reflecting the modern design of the Xe-HPG architecture.

Power and interface specifications differ as well. Intel's TDP is 25 W with an IGP slot width and PCIe 4.0 x8 interface. AMD's TDP is 75 W with an MXM Module slot width and PCIe 3.0 x16 interface. Both use portable device dependent display outputs. API support also reflects the generation gap: Intel supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, while AMD supports DirectX 12 (12_0), OpenGL 4.6, and Vulkan 1.2.170.

The Verdict

The data points to the Intel Arc A350M as the stronger performer in the only direct benchmark available. Its 46% lead in Geekbench Vulkan over the AMD Radeon R9 M390X is a large margin, and its average benchmark score, 24,647 versus 20,662, reinforces that conclusion. The Arc A350M also holds the 70th percentile among all GPUs, while the R9 M390X sits at the 66th percentile. For users focused on Vulkan-based workloads or modern feature support, the Intel part is clearly the better choice per the recorded data.

However, the AMD Radeon R9 M390X is not without its own strengths, strictly speaking from the database records. Its memory bandwidth of 160.0 GB/s exceeds the Intel part by 43%, which could be relevant in bandwidth-sensitive scenarios. Its higher shading unit count, 2,048 versus 768, gives it a theoretical advantage in raw shader throughput. Yet those advantages do not result in a benchmark win in the recorded tests. The Vulkan result shows a decisive loss, and the average benchmark score is 19% lower.

The production status of both is end-of-life, so neither is a future-proof option. The Intel Arc A350M was released on 2022-03-29, while the AMD Radeon R9 M390X was released on 2015-05-04. The seven-year gap in release timing is visible in the architecture and feature set.

Users should pick the Intel Arc A350M if they need modern API support, ray tracing hardware, higher compute throughput, or better Vulkan performance. Users should consider the AMD Radeon R9 M390X only if their specific workloads rely heavily on memory bandwidth or raw shader count, and even then, the benchmark data does not show those advantages translating into a win in the recorded tests.

FAQ

Q: Which GPU wins in Geekbench Vulkan?

A: The Intel Arc A350M wins with a score of 24,747 versus the AMD Radeon R9 M390X's 16,951, a 46% advantage for Intel. The R9 M390X secures 0 wins in the database, while the Arc A350M secures 1 win.

Q: Do both GPUs have 4 GB of VRAM and the same memory bus width?

A: Yes, both GPUs have 4 GB of memory, but they differ in memory type and bus width. The Intel Arc A350M uses GDDR6 on a 64-bit bus, while the AMD Radeon R9 M390X uses GDDR5 on a 256-bit bus. AMD has higher bandwidth at 160.0 GB/s, versus Intel's 112.0 GB/s, and higher bandwidth by 43%.

Q: Does the AMD Radeon R9 M390X have ray tracing support, and what about the Arc A350M?

A: The Intel Arc A350M has 6 ray tracing cores, while the AMD Radeon R9 M390X has no ray tracing cores listed. This makes ray tracing hardware available to Intel Arc A350M users.

Q: Which GPU supports the higher PCIe interface, and what about API differences?

A: The Intel Arc A350M supports PCIe 360 and is built on PCIe 4.0 x8, while the AMD Radeon R9 M390X supports PCIe 3.0 x16. The Intel Arc A350M supports DirectX 12 Ultimate (12_2), Vulkan 1.4, and OpenGL 4.6, and the AMD Radeon R9 M390X supports DirectX 12 (12_0), Vulkan 1.2.170, and OpenGL 4.6, and Vulkan 1.2.170. The Intel Arc A350M has DirectX 12 Ultimate (12_2), Vulkan 1.4, and OpenGL 4.6, while the AMD Radeon R9 M390X has DirectX 12 (12_0), Vulkan 1.2.170, and OpenGL 4.6.

Q: How do the shading unit and ROP counts compare between the two GPUs?

A: The AMD Radeon R9 M390X has 2,048 shading units and 32 ROPs, while the Intel Arc A350M has 768 shading units and 24 ROPs. AMD has 32 ROPs, and Intel has 24 ROPs, while Intel leads in pixel rate at 52.80 GPixel/s versus AMD's 23.14 GPixel/s, and Intel leads in texture rate at 105.6 GTexel/s versus AMD's 92.54 GTexel/s.

Q: Which GPU has the higher FP32 and FP16 performance?

A: Intel Arc A350M has a 14% higher FP32 at 3.379 TFLOPS versus AMD's 2.961 TFLOPS, and a 2.3x higher FP16 at 6.758 TFLOPS versus AMD's 2.961 TFLOPS. AMD Radeon R9 M390X has FP32 2.961 TFLOPS, FP16 2.961 TFLOPS, and FP16 1:1. Intel Arc A350M has FP32 3.379 TFLOPS, FP16 6.758 TFLOPS, FP16 (2:1).

Where Each One Wins

The database records show the Intel Arc A350M wins in the only head-to-head benchmark. The AMD Radeon R9 M390X does not win in any of the recorded benchmark tests, having 0 wins. The Intel Arc A350M has 1 win.

Focusing on compute and throughput metrics, the Intel Arc A350M outperforms the AMD Radeon R9 M390X in FP32 compute at 3.379 TFLOPS to 2.961 TFLOPS, pixel fill rate at 52.80 GPixel/s to 23.23 GPixel/s, texture fill rate at 105.6 GTexel/s to 92.54 GTexel/s, and FP16 compute at 6.638 TFLOPS to 2.961 TFLOPS. The Intel Arc A350M also has a higher average benchmark score of 24,647, and a 70th percentile ranking, versus AMD's 20,662 and 66th percentile, with a 19% difference between averages. The AMD Radeon R9 M390X holds the AMD Radeon R9 M390X has higher memory bandwidth at 160.0 GB/s versus 112.0 GB/s, a 256-bit bus versus the Intel 64-bit bus, and higher memory of GDDR5 versus GDDR6, and a 5 Gbps effective memory rate versus 14 Gbps effective, but the Intel Arc A350M has a 1750 MHz memory and 14 Gbps effective, and is 6 nm versus 28 nm. The AMD Radeon R9 M390X also has more shading units (2,048 to 768), more TMUs (128 to 48), and more ROPs (32 to 24), but Intel has ray tracing cores (6) and a higher boost clock (2200 MHz) and higher base clock (1150 MHz), plus a lower TDP of 25 W versus 75 W. The Intel Arc A350M supports PCIe 4.0 x8, and the AMD Radeon R9 M390X supports PCIe 3.0 x16. The Intel Arc A350M uses an IGP slot width, and AMD uses an MXM Module slot width.

Specification Differences

The following specifications differ between the Intel Arc A350M and the AMD Radeon R9 M390X:

• Chip: Intel uses DG2-128, AMD uses Amethyst.
• Architecture: Intel uses Xe-HPG, AMD uses GCN 3.0.
• Generation: Intel is Alchemist (Arc 3 Mobile), AMD is Gem System (R9 M300).
• Process node: Intel uses 6 nm, AMD uses 28 nm.
• Transistors: Intel has 7,200 million, AMD has 5,000 million.
• Die size: Intel is 157 mm², AMD is 366 mm².
• Transistor density: Intel is 45.9M / mm², AMD is 13.7M / mm².
• Base clock: Intel is 1150 MHz, AMD is not listed.
• Boost clock: Intel is 2200 MHz, AMD is not listed.
• Memory clock: Intel is 1750 MHz (14 Gbps effective), AMD is 1250 MHz (5 Gbps effective).
• Memory type: Intel uses GDDR6, AMD uses GDDR5.
• Memory bus width: Intel uses 64 bit, AMD uses 256 bit.
• Memory bandwidth: Intel is 112.0 GB/s, AMD is 160.0 GB/s.
• Shading units: Intel has 768, AMD has 2048.
• TMUs: Intel has 48, AMD has 128.
• ROPs: Intel has 24, AMD has 32.
• Ray tracing cores: Intel has 6, AMD has none.
• Pixel rate: Intel is 52.80 GPixel/s, AMD is 23.14 GPixel/s.
• Texture rate: Intel is 105.6 GTexel/s, AMD is 92.54 GTexel/s.
• FP32: Intel is 3.379 TFLOPS, AMD is 2.961 TFLOPS.
• FP16: Intel is 6.758 TFLOPS (2:1), AMD is 2.961 TFLOPS (1:1).
• TDP: Intel is 25 W, AMD is 75 W.
• Slot width: Intel is IGP, AMD is MXM Module.
• Power connectors: Intel has none listed, AMD is none.
• Bus interface: Intel is PCIe 4.0 x8, AMD is PCIe 3.0 x16.
• DirectX support: Intel is 12 Ultimate (12_2), AMD is 12 (12_0).
• Vulkan support: Intel is 1.4, AMD is 1.2.170.
• Release date: Intel is 2022-03-29, AMD is 2015-05-04.
• Predecessor: Intel has none listed, AMD is Solar System.
• Successor: Intel has none listed, AMD is Polaris Mobile.
• Average benchmark score: Intel is 24,647, AMD is 20,662.
• Percentile vs all GPUs: Intel is 70, AMD is 66.
• Manufacturer: Intel vs AMD.