AMD Radeon 890M vs Intel Arc B390 Comparison

The AMD Radeon 890M and Intel Arc B390 represent two distinct approaches to integrated graphics, with the data showing a clear performance split that depends entirely on the benchmark in question. The Intel Arc B390 delivers a significantly higher score in the single shared 3DMark Steel Nomad DX12 test, while the AMD Radeon 890M holds a substantial advantage in overall average benchmark scores due to a much broader set of recorded results. This analysis examines the recorded data, the architectural foundations, and the practical implications of these differences.

Head-to-Head Benchmarks

The only direct comparison available in the database is the 3DMark Steel Nomad DX12 test, and the result is decisive. The Intel Arc B390 scores 1482, while the AMD Radeon 890M scores 590. This represents a 60.2 percent deficit for the AMD part, meaning the Intel solution delivers roughly two and a half times the performance in this specific workload. The magnitude of this difference places the two products in entirely different performance tiers for this test, with the Intel Arc B390 emerging as the clear winner in the single head-to-head benchmark recorded.

However, the broader benchmark picture tells a different story. The AMD Radeon 890M has ten recorded benchmark scores across multiple test suites, while the Intel Arc B390 has only one entry in the database. The average benchmark score for the AMD Radeon 890M is 9210, compared to 1482 for the Intel Arc B390. This disparity in average scores is heavily influenced by the number of tests recorded, but it still reflects the AMD part's ability to perform across a wider range of workloads, including OpenCL, Vulkan, and several PassMark tests.

Looking at the AMD Radeon 890M's individual results, the Geekbench Vulkan score of 40808 stands out as its strongest showing, followed by a Geekbench OpenCL score of 37254. The PassMark G3D score of 8076 and PassMark G2D score of 979 round out the higher-end results, while the PassMark DirectX 9 score of 97 and PassMark DirectX 10 score of 33 represent the lower end of its range. These numbers indicate that the AMD part performs best in compute-oriented and modern API workloads, with older DirectX tests showing more modest results.

The Intel Arc B390's single 3DMark Steel Nomad DX12 score of 1482 places it in a different competitive context when compared to its nearest rivals. The database shows the Intel part is 1.3 percent ahead of the NVIDIA GeForce GT 520MX, 1.5 percent ahead of the NVIDIA GeForce 800M, 2.5 percent ahead of the NVIDIA GeForce GT 625 OEM, and 2.7 percent ahead of the NVIDIA GeForce GT 710. These are narrow margins, suggesting that the Intel Arc B390's score, while impressive relative to the AMD Radeon 890M, places it in a competitive cluster with these older NVIDIA parts rather than establishing a clear lead over a wide field.

The AMD Radeon 890M's nearest rivals tell a similar story of tight competition. The AMD Radeon Vega 8 is only 0.1 percent ahead, the NVIDIA GeForce GTX 960 is 0.7 percent ahead, the NVIDIA GeForce GTX 465 is 0.9 percent ahead, and the NVIDIA GeForce GTX 850M is 1 percent ahead. These margins are extremely small, indicating that the AMD Radeon 890M sits in a dense performance cluster where small variations in workload can shift the ranking. The percentile ranking of 45 for the AMD part versus 9 for the Intel part reflects this broader context, though it must be noted that the Intel part's percentile is based on a single benchmark result.

The Verdict

The data supports a clear conclusion: the Intel Arc B390 wins the only directly comparable benchmark by a wide margin, but the AMD Radeon 890M demonstrates broader capability across a wider range of recorded tests. For users prioritizing the specific workload represented by 3DMark Steel Nomad DX12, the Intel Arc B390 is the stronger choice based on the recorded data. The 60.2 percent deficit for the AMD part in this test is too large to ignore, and the Intel part's score of 1482 represents a level of performance that the AMD part does not approach in any recorded benchmark.

However, the AMD Radeon 890M's advantage in average benchmark score, driven by its ten recorded results, suggests it handles a more diverse set of tasks with greater consistency. The Geekbench Vulkan score of 40808 and Geekbench OpenCL score of 37254 indicate strong performance in compute-heavy applications, while the PassMark G3D score of 8076 shows respectable DirectX performance across multiple versions. The Intel Arc B390 has no recorded results for these tests, so its performance in these areas cannot be verified from the database.

The Intel Arc B390's narrow margins over its nearest rivals, ranging from 1.3 to 2.7 percent, suggest that its strong Steel Nomad score does not translate into a commanding lead across the broader GPU landscape. The AMD Radeon 890M's similar narrow margins against its rivals, ranging from 0.1 to 1 percent, indicate that both parts compete in tightly packed performance tiers, with the AMD part slightly further behind its nearest competitors than the Intel part is ahead of its own.

Architecture Differences

The architectural foundations of these two parts differ substantially. The AMD Radeon 890M uses the Strix Point chip with RDNA 3.5 architecture, manufactured on a 4 nm process at TSMC. The Intel Arc B390 uses the Panther Lake chip with Xe3-LPG architecture, manufactured on a 3 nm process at Intel. The process node difference gives the Intel part a density advantage on paper, though the AMD part's transistor count of 34,000 million and die size of 233 mm² are recorded, while the Intel part's transistor count and die size are listed as unknown.

The compute configurations reveal different design priorities. The AMD Radeon 890M has 1024 shading units, 64 texture mapping units, and 32 raster output units, while the Intel Arc B390 has 1536 shading units, 48 texture mapping units, and 24 raster output units. The Intel part has 50 percent more shading units, but the AMD part has 33 percent more texture mapping units and 33 percent more raster output units. This explains the fill rate differences: the AMD part delivers 92.80 GPixel/s and 185.6 GTexel/s, while the Intel part delivers 60.00 GPixel/s and 120.0 GTexel/s.

Ray tracing hardware also differs, with the AMD Radeon 890M featuring 16 ray tracing cores and the Intel Arc B390 featuring 12. The floating-point output shows the Intel part's advantage: 7.680 TFLOPS FP32 versus 5.939 TFLOPS for the AMD part. The FP16 figures are particularly telling, with the Intel Arc B390 delivering 15.36 TFLOPS at a 2:1 ratio, while the AMD Radeon 890M delivers 5.939 TFLOPS at a 1:1 ratio. This indicates the Intel part has dedicated hardware for half-precision workloads, while the AMD part processes them at the same rate as FP32.

Power consumption differs significantly. The AMD Radeon 890M has a TDP of 15 W, while the Intel Arc B390 has a TDP of 80 W. This is a fivefold difference in power envelope, which has direct implications for the systems in which these parts can be deployed. Both are classified as IGP (integrated graphics processor) with no power connectors, and both have display outputs listed as portable device dependent.

FAQ

Q: Which processor has the higher average benchmark score?

A: The AMD Radeon 890M has an average benchmark score of 9210, compared to 1482 for the Intel Arc B390. This is based on ten recorded benchmarks for the AMD part and one for the Intel part.

Q: How do the two compare in the 3DMark Steel Nomad DX12 test?

A: The Intel Arc B390 scores 1482, while the AMD Radeon 890M scores 590. The Intel part wins this test with a 60.2 percent advantage.

Q: What are the process nodes for each processor?

A: The AMD Radeon 890M uses a 4 nm process at TSMC, while the Intel Arc B390 uses a 3 nm process at Intel.

Q: Which processor has more shading units?

A: The Intel Arc B390 has 1536 shading units, while the AMD Radeon 890M has 1024 shading units.

Q: What is the TDP of each processor?

A: The AMD Radeon 890M has a TDP of 15 W, and the Intel Arc B390 has a TDP of 80 W.

Q: How do the nearest rivals compare for each processor?

A: The AMD Radeon 890M's closest rival is the AMD Radeon Vega 8, which is 0.1 percent ahead. The Intel Arc B390's closest rival is the NVIDIA GeForce GT 520MX, which the Intel part leads by 1.3 percent.

Where Each One Wins

The Intel Arc B390 wins in the single directly comparable benchmark, the 3DMark Steel Nomad DX12 test, with a score of 1482 versus 590. This represents a 60.2 percent advantage and is the only recorded head-to-head victory in the database. The Intel part also has a higher FP32 throughput at 7.680 TFLOPS and a significantly higher FP16 throughput at 15.36 TFLOPS, indicating an advantage in raw compute-heavy workloads that can utilize these capabilities. The 3 nm process node provides a manufacturing advantage, and the higher TDP of 80 W allows for sustained performance in systems that can provide adequate cooling and power delivery.

The AMD Radeon 890M wins in overall benchmark coverage and average score, with 9210 versus 1482. The Geekbench Vulkan score of 40808 and OpenCL score of 37254 demonstrate strong compute performance, and the PassMark G3D score of 8076 shows solid DirectX capabilities. The AMD part has a higher pixel rate at 92.80 GPixel/s and texture rate at 185.6 GTexel/s, which benefits fill-rate-bound workloads. The 15 W TDP makes it suitable for low-power systems, and the 4 nm process at TSMC provides a mature manufacturing base. The AMD part's 16 ray tracing cores exceed the Intel part's 12, suggesting an advantage in ray-traced workloads if the software can utilize them.

Specification Differences

The recorded specifications show clear differences between the two parts. The AMD Radeon 890M uses a 4 nm process at TSMC with 34,000 million transistors and a 233 mm² die size, while the Intel Arc B390 uses a 3 nm process at Intel with unknown transistor count and die size. The AMD part has a base clock of 400 MHz and boost clock of 2900 MHz, while the Intel part has a base clock of 300 MHz and boost clock of 2500 MHz.

The compute units differ in count and ratio: the AMD part has 1024 shading units, 64 TMUs, and 32 ROPs, while the Intel part has 1536 shading units, 48 TMUs, and 24 ROPs. The AMD part has 16 ray tracing cores, the Intel part has 12. The pixel rate is 92.80 GPixel/s for AMD and 60.00 GPixel/s for Intel; the texture rate is 185.6 GTexel/s for AMD and 120.0 GTexel/s for Intel. FP32 output is 5.939 TFLOPS for AMD and 7.680 TFLOPS for Intel; FP16 output is 5.939 TFLOPS at a 1:1 ratio for AMD and 15.36 TFLOPS at a 2:1 ratio for Intel.

The TDP is 15 W for AMD and 80 W for Intel. Both parts use system-shared memory with system-dependent bandwidth, both are classified as IGP with no power connectors, and both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The AMD part uses PCIe 4.0 x8 as its bus interface, while the Intel part's bus interface is listed as IGP. The AMD part has a production status of Active with a release date in July 2024, while the Intel part also has a production status of Active with a release date in January 2026. Both parts have no launch MSRP recorded in the database.