AMD Radeon 680M vs Intel Arc B390 Comparison

Head-to-Head Benchmarks

The two integrated GPUs share only one common benchmark in the database: 3DMark Steel Nomad DX12. The Intel Arc B390 records a score of 1482 points, while the AMD Radeon 680M scores 378 points. The Intel part leads by a delta of 74.5 percent, which is the sole head-to-head comparison available. This is a decisive margin in a single test, but the database does not contain additional overlapping workloads for these two parts, so the comparison rests entirely on this one measurement.

Looking at the broader benchmark context, the AMD Radeon 680M appears in three recorded tests. Its OpenCL score is 23,468 points and its Vulkan score is 21,965 points, alongside the 378-point Steel Nomad result. The average benchmark score across these three tests is 15,270 points. The Intel Arc B390 has only the single Steel Nomad result of 1482 points, giving it an average benchmark score of 1482 points. The disparity in test coverage means the averages are not directly comparable as a like-for-like measure, but the Steel Nomad result stands as the only direct point of contact.

The percentile placement reinforces the gap. The Intel Arc B390 sits at the 9th percentile among all GPUs in the database, while the AMD Radeon 680M sits at the 57th percentile. This percentile difference is significant, though it should be interpreted carefully because the AMD part has three benchmark entries and the Intel part has only one. The Steel Nomad test itself, a DX12 workload, favors the Intel architecture by a wide margin in this specific scenario.

Where Each One Wins

The Intel Arc B390 wins the only direct head-to-head benchmark, 3DMark Steel Nomad DX12, with a 74.5 percent lead over the AMD Radeon 680M. This is the single benchmark where both parts have recorded scores, and the Intel part dominates it. The score of 1482 versus 378 indicates that in this particular DX12 workload, the Intel integrated GPU delivers substantially higher performance.

The AMD Radeon 680M, however, holds advantages in benchmark coverage and overall standing. Its average benchmark score of 15,270 points comes from three tests, including OpenCL and Vulkan workloads where it records 23,468 and 21,965 points respectively. The Intel Arc B390 has no recorded scores in those tests, so the AMD part demonstrates capability across a broader set of APIs in the database. Its 57th percentile placement among all GPUs also shows it ranks higher in the overall distribution than the Intel part's 9th percentile placement.

For use-case analysis, the data suggests the Intel Arc B390 excels in DX12-based 3D rendering workloads of the type represented by Steel Nomad. The AMD Radeon 680M shows strength in compute-oriented benchmarks like OpenCL and Vulkan, where its scores are strong in absolute terms. Users or systems relying on DX12 gaming workloads would see the Intel part perform better based on the recorded data, while the AMD part shows more versatility across the benchmark suite it appears in.

Architecture Differences

The AMD Radeon 680M is built on the Rembrandt+ chip using the RDNA 2.0 architecture, belonging to the Navi II IGP generation for Rembrandt Mobile. It is fabricated on a 6 nm process at TSMC, with a transistor count of 13,100 million and a die size of 208 mm². This yields a transistor density of 63.0 million transistors per square millimeter.

The Intel Arc B390 uses the Panther Lake chip with the Xe3-LPG architecture, part of the Arc Graphics-M generation for Panther Lake. It is fabricated on a 3 nm process at Intel, but the database records its transistor count and die size as unknown, with no transistor density figure available. The process node is notably smaller, at 3 nm versus 6 nm, which typically allows for higher density and efficiency, though the Intel part's transistor data is not recorded to confirm this.

In terms of compute resources, the AMD Radeon 680M has 768 shading units, 48 texture mapping units, and 32 render output units. It also includes 12 ray tracing cores. The Intel Arc B390 has 1536 shading units, 48 texture mapping units, and 24 render output units, also with 12 ray tracing cores. The Intel part has twice the shading units, the same number of texture mapping units, and fewer render output units. This configuration suggests the Intel architecture favors shader-heavy workloads, while the AMD part has more rasterization throughput per shading unit.

Clock behavior also differs. The AMD Radeon 680M has a base clock of 2000 MHz and a boost clock of 2200 MHz. The Intel Arc B390 has a base clock of 300 MHz and a boost clock of 2500 MHz. The Intel part's base clock is much lower but its boost clock is higher, indicating a wider dynamic range. The AMD part operates at a higher sustained clock with a smaller boost delta.

Specification Differences

The two integrated GPUs differ in several recorded specifications. The AMD Radeon 680M uses a 6 nm process node from TSMC, while the Intel Arc B390 uses a 3 nm process node from Intel. The AMD part has 13,100 million transistors and a die size of 208 mm², while the Intel part's transistor count and die size are listed as unknown.

Shader configuration is a major difference: the AMD Radeon 680M has 768 shading units, while the Intel Arc B390 has 1536 shading units, exactly double. Render output units differ as well, with 32 on the AMD part and 24 on the Intel part. Texture mapping units are identical at 48 on both. Ray tracing cores are also identical at 12 on both.

Clock speeds differ notably. The AMD Radeon 680M has a base clock of 2000 MHz and a boost clock of 2200 MHz. The Intel Arc B390 has a base clock of 300 MHz and a boost clock of 2500 MHz. The calculated pixel rates reflect the ROP and clock differences: the AMD part produces 70.40 GPixel/s, while the Intel part produces 60.00 GPixel/s. Texture rates differ in the opposite direction: the AMD part produces 105.6 GTexel/s, while the Intel part produces 120.0 GTexel/s.

Floating-point throughput shows the Intel part with a substantial lead. The AMD Radeon 680M delivers 3.379 TFLOPS of FP32 performance and 6.758 TFLOPS of FP16 performance at a 2:1 ratio. The Intel Arc B390 delivers 7.680 TFLOPS of FP32 and 15.36 TFLOPS of FP16 at the same 2:1 ratio. The Intel part has more than double the FP32 throughput of the AMD part.

Power draw differs, with the AMD Radeon 680M rated at 50 W TDP and the Intel Arc B390 rated at 80 W TDP. Both are classified as IGP slot width with no power connectors. The bus interface also differs: the AMD part uses PCIe 4.0 x8, while the Intel part is listed simply as IGP. Both use system-shared memory with system-dependent bandwidth, and both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The release dates differ, with the AMD part dated January 2023 and the Intel part dated January 2026, though the database records both as Active in production status.

FAQ

Q: Which GPU has the higher average benchmark score?

A: The AMD Radeon 680M has an average benchmark score of 15,270 points across three tests, while the Intel Arc B390 has an average of 1,482 points from a single test. The AMD part's average is higher, but it is based on more benchmark entries.

Q: What is the result of the only direct head-to-head benchmark between these two GPUs?

A: In the 3DMark Steel Nomad DX12 test, the Intel Arc B390 scores 1,482 points versus 378 points for the AMD Radeon 680M, giving the Intel part a 74.5 percent lead.

Q: How do the shading unit counts compare?

A: The Intel Arc B390 has 1,536 shading units, exactly double the 768 shading units of the AMD Radeon 680M. Both have 48 texture mapping units and 12 ray tracing cores.

Q: What are the FP32 performance figures for each GPU?

A: The AMD Radeon 680M delivers 3.379 TFLOPS of FP32 performance, while the Intel Arc B390 delivers 7.680 TFLOPS, which is more than double.

Q: Which GPU has the higher render output unit count and pixel rate?

A: The AMD Radeon 680M has 32 render output units and a pixel rate of 70.40 GPixel/s. The Intel Arc B390 has 24 render output units and a pixel rate of 60.00 GPixel/s.

Q: What process nodes are used by these two GPUs?

A: The AMD Radeon 680M uses a 6 nm process from TSMC, while the Intel Arc B390 uses a 3 nm process from Intel. The AMD part has a recorded transistor count of 13,100 million and a die size of 208 mm², while the Intel part's transistor and die size data are unknown in the database.