AMD Radeon 840M vs Intel Arc Pro A60M Comparison

Head-to-Head Benchmarks

The recorded data places the AMD Radeon 840M and Intel Arc Pro A60M in directly comparable positions, both holding the 50th percentile against all GPUs in the database. Neither part has accumulated a single benchmark score in the current dataset, so the head-to-head comparison must be drawn entirely from their architectural specifications and computed throughput figures. What the data shows is a decisive gap in raw rendering capability, paired with a substantial difference in power envelope.

The Intel Arc Pro A60M delivers 5.325 TFLOPS of FP32 compute, while the AMD Radeon 840M produces 1,484.8 GFLOPS. That places the Intel part approximately 3.6 times ahead in single-precision throughput. The gap widens further in FP16: the Arc Pro A60M reaches 10.65 TFLOPS using a 2:1 ratio, whereas the Radeon 840M manages 1,484.8 GFLOPS at a 1:1 ratio. The Intel accelerator therefore holds a 7.2x advantage in half-precision workloads, which matters for AI inference and compute-heavy shaders.

Pixel throughput tells the same story. The Arc Pro A60M outputs 83.20 GPixel/s, compared to 23.20 GPixel/s for the Radeon 840M. That is a 3.6x difference in fill rate. Texture rate follows with 166.4 GTexel/s against 46.40 GTexel/s, again a 3.6x margin. These ratios are consistent across the board, indicating that the Intel part scales its entire render pipeline proportionally rather than favoring one stage.

The hardware resources behind these numbers explain the gap. Intel fields 2,048 shading units, 128 texture mapping units, and 64 render output units. AMD counters with 256 shading units, 16 TMUs, and 8 ROPs. The Intel part carries eight times the shader count, eight times the TMU count, and eight times the ROP count. The ray tracing hardware shows a similar pattern: 16 RT cores on Intel versus 4 on AMD, a 4x disparity.

Memory bandwidth amplifies the separation. The Arc Pro A60M uses 8 GB of GDDR6 on a 128-bit bus running at 2000 MHz with 16 Gbps effective speed, producing 256.0 GB/s. The Radeon 840M relies on system shared memory with bandwidth described as system dependent. No fixed number exists for the AMD part, which means its memory throughput is entirely contingent on the host platform's RAM configuration. In any realistic laptop setup, dedicated GDDR6 at 256.0 GB/s will exceed shared memory bandwidth, often by a wide margin.

Clock speeds complicate the comparison somewhat. The Radeon 840M boosts to 2900 MHz from a 400 MHz base, while the Arc Pro A60M boosts to 1300 MHz from 900 MHz. The AMD part runs at more than double the boost frequency, yet the Intel part still wins every throughput metric because its wider architecture compensates for the lower clock. This is a classic wide-versus-fast design tradeoff, and the data confirms Intel's approach yields far higher absolute performance.

The power draw difference is stark. The Radeon 840M carries a 15 W TDP, while the Arc Pro A60M is rated at 95 W. That is a 6.3x power consumption increase for the Intel part. The performance per watt calculation, using the recorded FP32 figures, shows the AMD part delivers approximately 99 GFLOPS per watt, while the Intel part delivers approximately 56 GFLOPS per watt. The Radeon 840M is therefore about 1.8 times more energy-efficient in raw compute per watt, despite being absolutely slower.

Process technology favors AMD as well. The Radeon 840M uses a 4 nm TSMC node, while the Arc Pro A60M uses 6 nm TSMC. The smaller node explains part of the AMD part's efficiency advantage. The Intel chip does report its transistor count at 11,500 million on a 269 mm² die, yielding a transistor density of 42.8M per mm². AMD lists its transistor count and die size as unknown, so no direct density comparison is possible from the data.

Bus interface width also differs. The Radeon 840M connects via PCIe 4.0 x8, while the Arc Pro A60M uses PCIe 4.0 x16. The Intel part has twice the host interface bandwidth, which matters for shared memory reads and PCIe-attached workloads. Both parts support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, so API feature parity is complete.

The Verdict

The benchmark data, sparse as it is, points to a clear stratification. The Intel Arc Pro A60M is the faster GPU by every measured throughput metric. It leads in FP32, FP16, pixel rate, texture rate, and memory bandwidth. It carries more shading units, more TMUs, more ROPs, and more RT cores. For any workload that stresses raw rendering performance, the data shows no contest: the Arc Pro A60M is roughly 3.6x faster in rasterization and 7.2x faster in FP16 compute.

The AMD Radeon 840M wins only in efficiency and integration. Its 15 W TDP versus 95 W means it can operate in thin-and-light chassis where the Intel part cannot fit thermally. Its 4 nm node and 2900 MHz boost clock demonstrate a modern, high-frequency design that extracts respectable performance from a minimal power budget. The data indicates it delivers about 1.8x better compute per watt, making it the rational choice for battery-constrained portable devices.

Who should pick which, strictly from the data? Anyone needing maximum frames per second, high-resolution textures, or compute acceleration should choose the Arc Pro A60M. The 8 GB dedicated GDDR6 pool and 256.0 GB/s bandwidth provide headroom for large assets and memory-intensive effects. The 16 RT cores give it a fourfold advantage in ray-traced scenes. The 5.325 TFLOPS FP32 throughput supports demanding professional applications without compromise.

Anyone prioritizing battery life, thermal headroom, or compact design should choose the Radeon 840M. Its 15 W TDP is six times lower, enabling fanless or low-noise implementations. Its 4 nm process and 1:1 FP16 ratio suit light compute tasks. The shared memory architecture eliminates the need for dedicated VRAM, reducing cost and complexity in the host system. The 2900 MHz boost clock shows the architecture can ramp quickly when needed.

The 50th percentile ranking for both parts suggests they sit in the middle of the overall GPU distribution. Neither is a flagship, neither is entry-level. The Arc Pro A60M's absolute performance places it in a higher tier despite the equal percentile, because the percentile reflects the full database including desktop parts. The Radeon 840M achieves the same percentile while consuming far less power, which underscores its efficiency profile.

Architecture Differences

The two GPUs come from fundamentally different design philosophies. AMD's Radeon 840M uses the Krackan Point chip based on RDNA 3.5, part of the Navi III IGP generation for Strix Point mobile platforms. Intel's Arc Pro A60M uses the DG2-256 chip based on Xe-HPG, part of the Alchemist generation in the Pro-Series mobile lineup. The process nodes differ: 4 nm TSMC for AMD, 6 nm TSMC for Intel.

The Radeon 840M is an integrated graphics processor. Its memory is system shared, its bus width is system shared, its bandwidth is system dependent. It has no dedicated VRAM and no power connectors. The slot width is listed as IGP, meaning it is soldered to the motherboard and cannot be upgraded or replaced. Its display outputs are portable device dependent, so the actual ports depend on the laptop design.

The Arc Pro A60M is also listed as an IGP in slot width, which is unusual for a chip with dedicated GDDR6. It carries 8 GB of GDDR6 on a 128-bit bus with 256.0 GB/s bandwidth. It uses PCIe 4.0 x16, doubling the lane count of the AMD part. Its display outputs are similarly portable device dependent. The power connector field is null, suggesting it draws power through the motherboard rather than a discrete connector.

The compute architectures reflect different priorities. AMD's RDNA 3.5 uses 256 shading units in a compact configuration optimized for power efficiency. The FP16 ratio is 1:1, meaning each FP32 unit handles half-precision at the same rate. Intel's Xe-HPG uses 2,048 shading units in a wide arrangement optimized for throughput. Its FP16 ratio is 2:1, meaning it doubles the rate for half-precision operations. This makes the Intel part substantially stronger for FP16-heavy workloads like machine learning.

Ray tracing hardware also differs. The Radeon 840M has 4 RT cores, while the Arc Pro A60M has 16. Both support DirectX 12 Ultimate, so ray-traced effects are available on either, but the Intel part has four times the dedicated hardware. Texture mapping units and ROPs follow the same pattern: 16 TMUs and 8 ROPs for AMD, 128 TMUs and 64 ROPs for Intel.

The transistor counts and die sizes are known for Intel (11,500 million transistors, 269 mm²) but unknown for AMD. The Intel chip achieves 42.8M transistors per square millimeter on 6 nm. The AMD part uses a smaller 4 nm node but its density cannot be calculated from the available data. Clock behavior differs sharply: the Radeon 840M has a 400 MHz base and 2900 MHz boost, while the Arc Pro A60M has a 900 MHz base and 1300 MHz boost. The AMD part has a much wider clock range, suggesting aggressive dynamic frequency scaling.

Release dates show a generation gap. The Radeon 840M launched on 2025-02-28, while the Arc Pro A60M launched on 2023-06-05. The AMD part is nearly two years newer. Its predecessor is listed as Navi II IGP, while Intel's predecessor is not specified. Both parts remain active in production.

FAQ

Q: Which GPU has higher raw compute performance?

A: The Intel Arc Pro A60M delivers 5.325 TFLOPS FP32 and 10.65 TFLOPS FP16, while the AMD Radeon 840M delivers 1,484.8 GFLOPS in both FP32 and FP16. The Intel part is roughly 3.6x faster in FP32 and 7.2x faster in FP16.

Q: How do their memory systems compare?

A: The Arc Pro A60M uses 8 GB of dedicated GDDR6 on a 128-bit bus with 256.0 GB/s bandwidth. The Radeon 840M uses system shared memory with bandwidth described as system dependent, meaning it varies by host platform.

Q: Which GPU is more power efficient?

A: The Radeon 840M has a 15 W TDP and delivers approximately 99 GFLOPS per watt. The Arc Pro A60M has a 95 W TDP and delivers approximately 56 GFLOPS per watt. The AMD part is about 1.8x more efficient in compute per watt.

Q: Do both support the same graphics APIs?

A: Yes. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

Q: Which GPU has more ray tracing hardware?

A: The Arc Pro A60M has 16 RT cores, while the Radeon 840M has 4. This gives the Intel part a 4x advantage in dedicated ray tracing resources.

Q: What are their process nodes?

A: The Radeon 840M uses a 4 nm TSMC process, while the Arc Pro A60M uses a 6 nm TSMC process. Both are fabricated by TSMC but on different node sizes.

Where Each One Wins

The AMD Radeon 840M wins in integrated scenarios. Its 15 W TDP fits within the thermal budget of ultraportable laptops. The 4 nm process and 2900 MHz boost clock show a design optimized for rapid frequency response. Its 1:1 FP16 ratio provides balanced compute for mixed workloads. The shared memory architecture eliminates dedicated VRAM costs. The PCIe 4.0 x8 interface is sufficient for system memory access. For battery-powered devices where every watt matters, the Radeon 840M is the clear choice based on the recorded efficiency advantage.

The Intel Arc Pro A60M wins in performance scenarios. Its 5.325 TFLOPS FP32 throughput handles demanding 3D rendering. The 10.65 TFLOPS FP16 accelerates AI inference and compute shaders. The 256.0 GB/s memory bandwidth feeds large textures and geometry without stalling. The 16 RT cores enable hardware-accelerated ray tracing at usable frame rates. The 128 TMUs and 64 ROPs sustain high resolutions and complex shading. The PCIe 4.0 x16 interface maximizes host data transfer. For workstation-class mobile tasks where absolute speed is the priority, the Arc Pro A60M dominates every measurable metric.

The pixel rate of 83.20 GPixel/s versus 23.20 GPixel/s means the Intel part can drive higher refresh rates and more displays. The texture rate of 166.4 GTexel/s versus 46.40 GTexel/s supports detailed surfaces and anisotropic filtering. The 8 GB GDDR6 pool versus system shared memory means the Intel part avoids bandwidth contention with the CPU. The 900 MHz base clock versus 400 MHz means the Intel part maintains higher minimum performance under sustained load.

The Radeon 840M wins in latency-sensitive scenarios where power draw is the constraint. Its 400 MHz base clock conserves energy during idle, and the 2900 MHz boost provides burst performance. The 1,484.8 GFLOPS throughput is sufficient for light gaming and productivity. The 4 RT cores provide entry-level ray tracing capability. The system shared memory allows fluid data exchange with the CPU without copying to dedicated VRAM.

The Arc Pro A60M wins in throughput-sensitive scenarios where sustained performance is required. Its 95 W TDP indicates it can maintain high clocks under load. The 16 RT cores and 2,048 shading units process complex scenes in parallel. The 256.0 GB/s bandwidth ensures the GPU never starves for data. The 11,500 million transistors on 269 mm² provide substantial hardware resources. For professional mobile workstations, the data clearly favors Intel's dedicated memory and wider architecture.