AMD Radeon 840M vs Intel Arc Pro B390 Comparison

AMD Radeon 840M and Intel Arc Pro B390 are both integrated graphics processors designed for mobile platforms, but they represent fundamentally different design philosophies and performance tiers. The database records show that the AMD Radeon 840M, part of the Krackan Point chip, uses the RDNA 3.5 architecture on a 4 nm TSMC process, while the Intel Arc Pro B390, built on the Panther Lake chip, employs the Xe3-LPG architecture on a 3 nm Intel process. Both are classified as active production parts with a percentile ranking of 50 against all GPUs, indicating they sit in the middle of the performance distribution, though the raw specifications suggest a significant gap in compute capabilities.

The Verdict

The data indicates two distinct roles for these integrated GPUs. The AMD Radeon 840M, with a 15 W TDP, is positioned for power-constrained, thin-and-light portable devices where battery life and thermal limits take priority over raw graphics throughput. Its boost clock of 2900 MHz is higher than the Intel part’s 2500 MHz, but the sheer difference in execution units tells a different story. The Intel Arc Pro B390, carrying an 80 W TDP, is a much larger integrated GPU designed for workloads that demand sustained compute performance, likely in professional mobile workstations or high-performance laptops where the increased power budget is acceptable.

For users whose primary concern is maximum graphics performance in a compact form factor, the Intel Arc Pro B390 is the clear choice from the recorded data. Its shading unit count of 1536 is six times that of the AMD Radeon 840M’s 256, and its FP32 throughput of 7.680 TFLOPS dwarfs the 1,484.8 GFLOPS of the AMD part. Conversely, for users who prioritize efficiency and lower power consumption, the AMD Radeon 840M at 15 W offers a much lighter thermal footprint, making it suitable for devices where sustained heavy loads are rare and where the lower power draw extends battery operation. The Intel part’s 80 W TDP is more than five times higher, which in a portable device would require substantial cooling and a larger battery.

The release dates also inform the verdict. The AMD Radeon 840M launched on February 28, 2025, while the Intel Arc Pro B390 launched on January 26, 2026. The later arrival of the Intel part suggests it benefits from a newer generation of process technology, and indeed the 3 nm node versus the 4 nm node for AMD aligns with this timeline. There are no benchmark scores recorded for either GPU in the database, so the verdict rests entirely on architectural specifications and power envelopes. The data shows that the Intel Arc Pro B390 is the performance-oriented option, while the AMD Radeon 840M is the efficiency-oriented option.

Architecture Differences

The two GPUs diverge at the fundamental level of their underlying architectures. The AMD Radeon 840M is built on RDNA 3.5, a refinement of AMD’s RDNA line, and is fabricated on a 4 nm process at TSMC. The Intel Arc Pro B390 uses Xe3-LPG, the third generation of Intel’s Xe architecture, manufactured on a 3 nm process at Intel’s own foundry. The process node difference, 4 nm versus 3 nm, gives Intel a potential density and power efficiency advantage per transistor, though the actual transistor counts and die sizes are listed as unknown in the database.

The compute resources are starkly different. The AMD part has 256 shading units, 16 texture mapping units, and 8 render output units. The Intel part has 1536 shading units, 48 texture mapping units, and 24 render output units. This means Intel’s GPU has six times the shading units, three times the TMUs, and three times the ROPs. The ray tracing cores also differ, with AMD offering 4 and Intel offering 12, a threefold advantage for Intel. Neither GPU has tensor cores listed, so any AI acceleration features are not recorded in the database.

Clock speeds show a different pattern. The AMD Radeon 840M has a base clock of 400 MHz and a boost clock of 2900 MHz. The Intel Arc Pro B390 has a base clock of 300 MHz and a boost clock of 2500 MHz. AMD’s part runs at higher frequencies, which partially compensates for its lower execution unit count, but not enough to close the gap. The FP16 performance also reveals an architectural difference: the AMD part lists FP16 at 1,484.8 GFLOPS with a 1:1 ratio to FP32, meaning it processes FP16 at the same rate as FP32. The Intel part lists FP16 at 15.36 TFLOPS with a 2:1 ratio, meaning it processes FP16 at twice the rate of FP32. This indicates Intel’s architecture has dedicated or more efficient FP16 paths, which can benefit certain compute workloads.

Memory interfaces are identical in description: both use system shared memory with a system shared bus width and system dependent bandwidth. Neither has dedicated VRAM, so performance will depend heavily on the host system’s memory configuration, but the database does not record specific bandwidth numbers. The pixel rate and texture rate further highlight the gap: the AMD part achieves 23.20 GPixel/s and 46.40 GTexel/s, while the Intel part achieves 60.00 GPixel/s and 120.0 GTexel/s. The Intel GPU’s pixel rate is roughly 2.6 times higher, and its texture rate is roughly 2.6 times higher as well.

Both GPUs support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, so API compatibility is identical. The bus interface differs: the AMD part uses PCIe 4.0 x8, while the Intel part is listed simply as IGP, which likely means it is integrated into the processor with no separate PCIe link. Display outputs are portable device dependent for both, meaning the number and type of ports depend on the laptop or handheld design.

FAQ

Q: Which GPU has a higher boost clock?

A: The AMD Radeon 840M has a boost clock of 2900 MHz, while the Intel Arc Pro B390 has a boost clock of 2500 MHz. The AMD part runs 400 MHz higher at maximum boost.

Q: How do the two GPUs compare in terms of power consumption?

A: The AMD Radeon 840M has a TDP of 15 W, while the Intel Arc Pro B390 has a TDP of 80 W. The Intel part consumes more than five times the power of the AMD part.

Q: Which GPU has more shading units?

A: The Intel Arc Pro B390 has 1536 shading units, compared to 256 shading units on the AMD Radeon 840M. This is a sixfold difference in favor of Intel.

Q: Do both GPUs support the same DirectX version?

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

Q: What is the FP32 performance difference?

A: The Intel Arc Pro B390 delivers 7.680 TFLOPS of FP32 performance, while the AMD Radeon 840M delivers 1,484.8 GFLOPS. Intel’s part is approximately 5.2 times faster in FP32 compute.

Q: When was each GPU released?

A: The AMD Radeon 840M was released on February 28, 2025, and the Intel Arc Pro B390 was released on January 26, 2026.

Specification Differences

The following fields differ between the two GPUs according to the database. The process node is 4 nm for AMD and 3 nm for Intel. The foundry is TSMC for AMD and Intel for Intel. The base clock is 400 MHz for AMD and 300 MHz for Intel. The boost clock is 2900 MHz for AMD and 2500 MHz for Intel. The shading units are 256 for AMD and 1536 for Intel. The texture mapping units are 16 for AMD and 48 for Intel. The render output units are 8 for AMD and 24 for Intel. The ray tracing cores are 4 for AMD and 12 for Intel. The pixel rate is 23.20 GPixel/s for AMD and 60.00 GPixel/s for Intel. The texture rate is 46.40 GTexel/s for AMD and 120.0 GTexel/s for Intel. The FP32 performance is 1,484.8 GFLOPS for AMD and 7.680 TFLOPS for Intel. The FP16 performance is 1,484.8 GFLOPS with a 1:1 ratio for AMD and 15.36 TFLOPS with a 2:1 ratio for Intel. The TDP is 15 W for AMD and 80 W for Intel. The bus interface is PCIe 4.0 x8 for AMD and IGP for Intel. The production status is active for both, but the release date is February 28, 2025 for AMD and January 26, 2026 for Intel. The predecessor is listed as Navi II IGP for AMD and HD Graphics-WM for Intel. The chip is Krackan Point for AMD and Panther Lake for Intel. The architecture is RDNA 3.5 for AMD and Xe3-LPG for Intel. The generation is Navi III IGP (Strix Point Mobile) for AMD and Arc Graphics-WM (Panther Lake) for Intel.

Fields that are the same include memory size, type, bus width, and bandwidth, all listed as system shared or system dependent. Both have no power connectors and a slot width of IGP. Both have portable device dependent display outputs. Both have the same API support for DirectX 12 Ultimate, OpenGL 4.6, and Vulkan 1.4. Neither has a launch MSRP recorded in the database.

Head-to-Head Benchmarks

The database contains no recorded head-to-head benchmark results for these two GPUs, so the comparison must be derived from the specification-level performance indicators. The most direct measure of compute throughput is FP32, where the Intel Arc Pro B390 achieves 7.680 TFLOPS against the AMD Radeon 840M’s 1,484.8 GFLOPS. This gives Intel a 5.2x advantage in raw single-precision floating-point performance. In FP16, the Intel part reaches 15.36 TFLOPS, which is 10.3 times the AMD part’s 1,484.8 GFLOPS, and the 2:1 ratio on Intel versus the 1:1 ratio on AMD means Intel can double its FP16 rate relative to FP32, while AMD cannot.

The pixel rate shows Intel at 60.00 GPixel/s versus AMD at 23.20 GPixel/s, a 2.6x difference. The texture rate shows Intel at 120.0 GTexel/s versus AMD at 46.40 GTexel/s, also a 2.6x difference. These rates correspond to the fill rate capabilities of the GPUs, which affect how fast they can draw pixels and apply textures. The ray tracing core count is 12 for Intel and 4 for AMD, a 3x difference, though without benchmark data the actual ray tracing performance cannot be quantified.

The clock speed advantage goes to AMD, with a boost clock of 2900 MHz versus 2500 MHz for Intel. This means AMD’s smaller GPU runs at a 16% higher frequency at peak boost. However, the execution unit count disparity overwhelms this clock advantage. The AMD part has 256 shading units, and even at a higher clock, the total shading capacity is far lower than Intel’s 1536 units. The power efficiency also differs: AMD delivers 1,484.8 GFLOPS of FP32 at 15 W, which is approximately 99 GFLOPS per watt, while Intel delivers 7.680 TFLOPS at 80 W, which is 96 GFLOPS per watt. These are nearly identical efficiency levels, but the Intel part’s absolute performance is much higher.

The bus interface difference is notable. The AMD Radeon 840M uses PCIe 4.0 x8, which provides a dedicated link to the host processor. The Intel Arc Pro B390 is listed simply as IGP, suggesting it does not have a separate bus interface but is directly integrated. This could affect how quickly each GPU accesses system memory, though the database does not provide bandwidth numbers to confirm this.

Where Each One Wins

The AMD Radeon 840M wins in power efficiency and clock speed. Its 15 W TDP is a fraction of the Intel part’s 80 W, making it suitable for devices with small batteries and passive cooling or minimal fan profiles. The higher boost clock of 2900 MHz means that for lightly threaded or short-burst workloads, the AMD part can respond quickly, though its limited execution units constrain sustained performance. The PCIe 4.0 x8 interface provides a standardized connection that may offer more predictable memory access patterns in some systems, though the database does not include benchmarks to confirm this.

The Intel Arc Pro B390 wins in every raw performance category recorded. It has more shading units, TMUs, ROPs, and ray tracing cores. It delivers higher pixel and texture rates, and its FP32 and FP16 throughput are multiples of the AMD part. The 3 nm process node and the 2:1 FP16 ratio indicate an architecture optimized for compute-heavy tasks, potentially including professional graphics workloads, video encoding, or scientific calculations. The 80 W TDP, while high for an integrated GPU, is still low compared to discrete graphics cards, and the absence of power connectors means it draws power through the motherboard, simplifying system integration.

For gaming, the higher pixel rate and texture rate on the Intel part suggest it can handle higher resolutions and more complex scenes, though the specific game performance is not recorded. For portable productivity, the AMD part’s lower power draw allows for thinner chassis designs and longer battery life. For professional applications that leverage FP16 or ray tracing, the Intel part’s 2:1 FP16 ratio and 12 ray tracing cores provide a substantial advantage. The data does not include any benchmark scores, so these conclusions are drawn solely from the architectural specifications and power envelopes recorded in the database.