AMD Radeon 680M vs Intel Arc Pro B370 Comparison

Head-to-Head Benchmarks

The AMD Radeon 680M and Intel Arc Pro B370 represent two distinct approaches to integrated graphics, yet the recorded data shows only one side with a complete set of measurements. The database contains three benchmark results for the AMD Radeon 680M: a 3DMark Steel Nomad DX12 score of 378, a Geekbench OpenCL score of 23468, and a Geekbench Vulkan score of 21965. The Intel Arc Pro B370 has no recorded benchmark scores in the database at this time, with its average benchmark score listed as 0 and no entries in the nearest rivals field. This absence of data for the Intel part means direct head-to-head comparisons cannot be calculated, and the wins tally reflects that asymmetry with zero wins for each side.

The AMD Radeon 680M's average benchmark score of 15270 places it in the 57th percentile of all GPUs in the database. Its nearest rivals provide useful context for interpreting this figure. The NVIDIA GeForce RTX 2060 sits just ahead with an average score of 15290, a delta of -0.1% relative to the Radeon 680M, meaning the AMD part trails by a mere 20 points. The NVIDIA GeForce GTX 580 posts an average of 15283, also a -0.1% delta, showing the Radeon 680M is effectively tied with that older discrete card. The NVIDIA GeForce RTX 3050 OEM scores 15199, which places it 0.5% behind the Radeon 680M. The AMD Radeon RX 7600, a desktop discrete GPU, scores 15171, a 0.7% deficit relative to the integrated Radeon 680M. These deltas are all within a narrow band of roughly 1% across four rival GPUs, indicating the Radeon 680M's performance clusters tightly around the level of these mid-range and older high-end discrete cards.

The Geekbench results for the Radeon 680M further illustrate its compute capabilities. The OpenCL score of 23468 exceeds the Vulkan score of 21965 by approximately 6.8%, suggesting the device's OpenCL driver path delivers slightly higher throughput in that benchmark workload. The FP32 rating of 3.379 TFLOPS and FP16 rating of 6.758 TFLOPS (at a 2:1 ratio) provide the theoretical peak compute figures, while the texture rate of 105.6 GTexel/s and pixel rate of 70.40 GPixel/s describe the rasterization throughput limits. For the Intel Arc Pro B370, the FP32 rating of 6.144 TFLOPS is roughly 81.8% higher than the Radeon 680M's 3.379 TFLOPS, and the FP16 figure of 12.29 TFLOPS nearly doubles the AMD part's 6.758 TFLOPS. These raw compute specifications suggest the Intel architecture has a substantial theoretical advantage in shader throughput, though no benchmark data exists to confirm whether that advantage materializes in real workloads.

The texture rates differ notably as well. The Radeon 680M achieves 105.6 GTexel/s from its 48 TMUs at a 2200 MHz boost clock, while the Intel Arc Pro B370 manages 96.00 GTexel/s from 40 TMUs at a 2400 MHz boost clock. Despite having fewer texture units, the Intel part's higher clock rate narrows the gap to a 9.1% deficit. Pixel rates show a larger divergence: the Radeon 680M's 70.40 GPixel/s from 32 ROPs compares to the Intel part's 48.00 GPixel/s from 20 ROPs, a 31.8% advantage for the AMD design. These architectural differences in fixed-function units will influence performance scaling in fill-rate-bound scenarios, though the lack of Intel benchmark results prevents any empirical verification.

The Verdict

The recorded data presents an incomplete picture. The AMD Radeon 680M has full benchmark coverage, including an average score of 15270 and a 57th percentile ranking, while the Intel Arc Pro B370 has no benchmark entries, an average score of 0, and sits at the 50th percentile by default. For the Radeon 680M, the data confirms it performs essentially on par with the NVIDIA GeForce GTX 580, NVIDIA GeForce RTX 2060, NVIDIA GeForce RTX 3050 OEM, and AMD Radeon RX 7600, with deltas ranging from -0.1% to 0.7%. The RTX 2060 and GTX 580 are marginally faster, while the RTX 3050 OEM and RX 7600 are marginally slower. This indicates the Radeon 680M delivers a level of integrated graphics performance that competes with those discrete offerings from several generations.

The Intel Arc Pro B370 cannot be evaluated on the same terms due to missing benchmark data. Its theoretical specifications show higher FP32 and FP16 throughput, a higher boost clock of 2400 MHz versus 2200 MHz, and a smaller process node of 3 nm versus 6 nm. However, the absence of measured scores means the database cannot confirm whether these specifications translate into superior real-world performance. The Intel part also has a lower pixel rate and texture rate, which could offset its compute advantages in certain workloads. Without benchmark entries, any verdict on the Intel Arc Pro B370 must remain provisional, and the data does not support a definitive choice between these two integrated GPUs.

The production status for both parts is listed as Active, and both target portable devices with display outputs described as Portable Device Dependent. The Radeon 680M uses a PCIe 4.0 x8 bus interface, while the Intel Arc Pro B370 lists its bus interface simply as IGP. The AMD part has a TDP of 50 W, while the Intel part has a TDP of 25 W, a factor of two difference that could influence thermal and power constraints in mobile designs. Users of the Radeon 680M can reference its measured performance against known discrete GPUs, which provides a concrete expectation for gaming and compute tasks. Users of the Intel Arc Pro B370 would need to rely on its architectural specifications alone, as no comparable measurement exists in the database. The data favors the AMD Radeon 680M for those who want verified performance numbers, while the Intel Arc Pro B370 remains an unquantified option with strong theoretical compute specifications.

FAQ

Q: What is the average benchmark score for the AMD Radeon 680M?

A: The AMD Radeon 680M has an average benchmark score of 15270, placing it in the 57th percentile of all GPUs in the database.

Q: Does the Intel Arc Pro B370 have any benchmark results in the database?

A: No, the Intel Arc Pro B370 has no recorded benchmark scores. Its average benchmark score is listed as 0, and it has no nearest rivals entries.

Q: How does the AMD Radeon 680M compare to the NVIDIA GeForce RTX 2060?

A: The Radeon 680M trails the RTX 2060 by 0.1%. The RTX 2060 has an average score of 15290, while the Radeon 680M has an average score of 15270.

Q: Which GPU has a higher FP32 compute rating?

A: The Intel Arc Pro B370 has a higher FP32 rating of 6.144 TFLOPS, compared to the AMD Radeon 680M's 3.379 TFLOPS.

Q: What are the TDP values for these two GPUs?

A: The AMD Radeon 680M has a TDP of 50 W, and the Intel Arc Pro B370 has a TDP of 25 W.

Q: Which GPU has a higher pixel rate?

A: The AMD Radeon 680M has a pixel rate of 70.40 GPixel/s, which is higher than the Intel Arc Pro B370's 48.00 GPixel/s.

Specification Differences

The AMD Radeon 680M and Intel Arc Pro B370 differ across several key specification fields. The process node for the Radeon 680M is 6 nm, fabricated by TSMC, while the Intel Arc Pro B370 uses a 3 nm node from Intel's own foundry. Transistor counts also diverge: the Radeon 680M contains 13,100 million transistors on a 208 mm² die, giving a transistor density of 63.0M per mm², while the Intel Arc Pro B370's transistor count and die size are listed as unknown.

Clock speeds show a notable difference in base frequency. The Radeon 680M has a base clock of 2000 MHz and a boost clock of 2200 MHz. The Intel Arc Pro B370 has a base clock of 300 MHz, significantly lower, but a boost clock of 2400 MHz, which is 200 MHz higher than the AMD part. Both use System Shared memory with System Dependent bandwidth.

The shader configurations differ substantially. The Radeon 680M has 768 shading units, 48 TMUs, 32 ROPs, and 12 ray tracing cores. The Intel Arc Pro B370 has 1280 shading units, 40 TMUs, 20 ROPs, and 10 ray tracing cores. Neither part lists tensor cores. The throughput rates reflect these differences: the Radeon 680M achieves 70.40 GPixel/s and 105.6 GTexel/s, while the Intel Arc Pro B370 achieves 48.00 GPixel/s and 96.00 GTexel/s. Compute figures show the Intel part ahead with 6.144 TFLOPS FP32 and 12.29 TFLOPS FP16, against the Radeon 680M's 3.379 TFLOPS FP32 and 6.758 TFLOPS FP16.

Power consumption differs by a factor of two: the Radeon 680M has a TDP of 50 W, and the Intel Arc Pro B370 has a TDP of 25 W. Both are IGP slot width with no power connectors. The bus interface differs: the Radeon 680M uses PCIe 4.0 x8, while the Intel Arc Pro B370 lists IGP. Both have Portable Device Dependent display outputs. The release dates are distinct, with the Radeon 680M released on 2023-01-02 and the Intel Arc Pro B370 released on 2026-01-26. The Radeon 680M's predecessor is the Vega II IGP and its successor is the Navi III IGP, while the Intel Arc Pro B370's predecessor is the HD Graphics-WM with no listed successor.

Architecture Differences

The architectural foundations of these two GPUs are fundamentally different. The AMD Radeon 680M uses the RDNA 2.0 architecture, built on the Rembrandt+ chip, and belongs to the Navi II IGP generation for Rembrandt Mobile. The Intel Arc Pro B370 uses the Xe3-LPG architecture, built on the Panther Lake chip, and belongs to the Arc Graphics-WM generation for Panther Lake. These are separate design philosophies: RDNA 2.0 is AMD's second-generation Radeon DNA architecture optimized for gaming efficiency, while Xe3-LPG is Intel's third-generation Xe architecture tailored for low-power graphics in mobile processors.

The manufacturing processes reflect a generational difference. The Radeon 680M is fabricated on TSMC's 6 nm node with 13,100 million transistors on a 208 mm² die, yielding a density of 63.0M transistors per mm². The Intel Arc Pro B370 is fabricated on Intel's 3 nm node, though its transistor count and die size are unknown in the database. The smaller process node for the Intel part suggests a more advanced manufacturing technology, which could contribute to its lower 25 W TDP.

Core configuration differences are substantial. The Radeon 680M packs 768 shading units, 48 TMUs, and 32 ROPs, while the Intel Arc Pro B370 has 1280 shading units, 40 TMUs, and 20 ROPs. The Intel part has 66.7% more shading units, which drives its higher FP32 and FP16 throughput. However, the AMD part has more texture units and more ROPs, giving it higher texture and pixel fill rates despite lower clock frequencies in the case of texture rate. Ray tracing core counts are close: 12 for the Radeon 680M and 10 for the Intel Arc Pro B370. Both architectures support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, indicating feature parity at the API level.

The clock behavior differs in a way that hints at different power management strategies. The Radeon 680M runs at a 2000 MHz base clock, while the Intel Arc Pro B370 idles at a 300 MHz base clock but boosts to 2400 MHz. This suggests the Intel part relies more aggressively on boost states to conserve power during light loads, consistent with its 25 W TDP. The Radeon 680M's higher base clock and 50 W TDP indicate a more sustained performance profile at the cost of higher power consumption. Both parts use System Shared memory, making their memory performance dependent on the host system's memory configuration rather than dedicated VRAM.