AMD Ryzen Z1 GPU vs Intel Arc B390 Comparison

Head-to-Head Benchmarks

The recorded data for this comparison is limited to a single benchmark result for the Intel Arc B390, while the AMD Ryzen Z1 GPU has no benchmark scores listed in the database. The Intel Arc B390 achieves a 3DMark Steel Nomad DX12 score of 1482. This places the Arc B390 at the 9th percentile among all GPUs in the database, meaning approximately 91% of recorded graphics processors score higher. The AMD Ryzen Z1 GPU holds a 50th percentile ranking, though this figure is not accompanied by any specific average benchmark score, making direct numerical comparison impossible.

The nearest rivals for the Intel Arc B390 provide context for its performance tier. The NVIDIA GeForce GT 520MX scores 1463, which is 1.3% lower than the Arc B390. The NVIDIA GeForce 800M trails by 1.5% with a score of 1460. The NVIDIA GeForce GT 625 OEM records 1446, a 2.5% deficit. The NVIDIA GeForce GT 710 scores 1443, sitting 2.7% behind the Arc B390. These deltas indicate that the Arc B390's performance margin over its closest competitors is narrow, ranging from just over one percentage point to under three percentage points. The data shows a tightly clustered group of low-end mobile and OEM graphics solutions, with the Arc B390 occupying the top position among them by a slim margin.

The AMD Ryzen Z1 GPU, despite lacking an average benchmark score, carries a much higher percentile placement at 50, which is more than five times the percentile rank of the Arc B390. This discrepancy suggests that the Z1 GPU belongs to a fundamentally different performance class, even though the database does not provide a direct score to quantify the gap. The head-to-head benchmark table contains no entries, so no direct wins are recorded for either product in this dataset.

Architecture Differences

The two processors diverge substantially in their underlying designs. The AMD Ryzen Z1 GPU uses the Phoenix chip built on RDNA 3.0 architecture, fabricated 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's own foundry. The Z1 GPU integrates 25,390 million transistors on a 178 mm² die, yielding a transistor density of 142.6 million transistors per square millimeter. The Arc B390's transistor count and die size are listed as unknown, so no density comparison is possible.

The compute configurations differ dramatically. The AMD Ryzen Z1 GPU contains 256 shading units, 16 texture mapping units, 8 raster operation pipelines, and 4 ray tracing cores. The Intel Arc B390 scales up to 1536 shading units, 48 TMUs, 24 ROPs, and 12 ray tracing cores. This represents a 6x increase in shading units, a 3x increase in TMUs, and a 3x increase in ROPs for the Intel part. The pixel rate for the Arc B390 is 60.00 GPixel/s, exactly three times the 20.00 GPixel/s of the Z1 GPU. The texture rate follows the same pattern: 120.0 GTexel/s for the Arc B390 versus 40.00 GTexel/s for the Z1 GPU.

Floating point throughput scales accordingly. The Arc B390 delivers 7.680 TFLOPS FP32 performance, while the Z1 GPU delivers 2.560 TFLOPS, a ratio of exactly 3 to 1. FP16 performance is 15.36 TFLOPS for the Arc B390 and 5.120 TFLOPS for the Z1 GPU, again a 3x difference, with both parts using a 2:1 ratio for FP16 relative to FP32.

Clock behavior differs in notable ways. The AMD part has a base clock of 1500 MHz and a boost clock of 2500 MHz. The Intel part has a much lower base clock of 300 MHz but matches the 2500 MHz boost clock. This large base clock gap suggests different power management strategies, with the Intel integrated part likely spending more time at low clocks when idle or lightly loaded. The memory subsystems are entirely different in kind. The Z1 GPU uses 16 GB of LPDDR5 memory on a 64-bit bus with 51.20 GB/s of bandwidth and a memory clock of 800 MHz, translating to 6.4 Gbps effective. The Arc B390 relies on system shared memory with system dependent bandwidth, meaning its memory performance is not fixed and depends entirely on the host platform's memory configuration.

Power and physical characteristics reinforce the different positioning. The Z1 GPU has a 30 W TDP, while the Arc B390 has an 80 W TDP, nearly three times higher. The Z1 GPU is a discrete card with physical dimensions of 280 mm length, 111 mm height, and 21 mm width, and it uses no power connectors. The Arc B390 is an integrated graphics processor with no physical dimensions listed and no power connectors. Both support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, so API feature parity is present. The Z1 GPU has no display outputs, while the Arc B390's display outputs are portable device dependent, reflecting its integrated nature in mobile systems.

The release timeline separates the products by more than two years. The AMD Ryzen Z1 GPU was released on 2023-09-17 with a launch MSRP of 599 USD. The Intel Arc B390 was released on 2026-01-26 and has no launch MSRP listed.

The Verdict

The data presents a clear structural hierarchy. The Intel Arc B390's only recorded benchmark, 1482 in 3DMark Steel Nomad DX12, places it at the 9th percentile, with nearest rivals all within 2.7% of its score. The AMD Ryzen Z1 GPU, despite having no benchmark score in the database, holds a 50th percentile ranking, which indicates it belongs to a substantially higher performance tier than the Arc B390 based on the database's overall distribution.

The architecture comparison reinforces this separation. The Arc B390 offers 3x the shading units, 3x the TMUs, 3x the ROPs, and 3x the FP32 throughput of the Z1 GPU. It also draws 80 W versus 30 W. However, the percentile data contradicts what raw compute counts might suggest, since the Z1 GPU's 50th percentile placement is far above the Arc B390's 9th percentile. The benchmark score for the Arc B390, which places it among legacy NVIDIA mobile and OEM parts like the GT 520MX and GT 710, indicates that its raw specifications do not translate into competitive real-world performance in the recorded test.

The Z1 GPU appears in the database as a discrete card with dedicated 16 GB LPDDR5 memory, a 64-bit bus, and 51.20 GB/s bandwidth. The Arc B390 is an integrated solution sharing system memory, which may constrain its performance in memory-sensitive workloads. The database records no benchmark score for the Z1 GPU, so its performance cannot be directly compared to the Arc B390's 1482 score. The percentile ranking is the only quantitative performance indicator available for the Z1 GPU, and it suggests the AMD part sits near the middle of the overall GPU distribution, while the Intel part sits near the bottom.

FAQ

Q: Which GPU has the higher percentile ranking in the database?

A: The AMD Ryzen Z1 GPU ranks at the 50th percentile among all GPUs, while the Intel Arc B390 ranks at the 9th percentile.

Q: What is the Intel Arc B390's benchmark score and how does it compare to its nearest rivals?

A: The Arc B390 scores 1482 in 3DMark Steel Nomad DX12. Its nearest rivals are the NVIDIA GeForce GT 520MX at 1463 (1.3% lower), the NVIDIA GeForce 800M at 1460 (1.5% lower), the NVIDIA GeForce GT 625 OEM at 1446 (2.5% lower), and the NVIDIA GeForce GT 710 at 1443 (2.7% lower).

Q: How do the shading unit counts compare between the two GPUs?

A: The Intel Arc B390 has 1536 shading units, while the AMD Ryzen Z1 GPU has 256 shading units, giving the Intel part six times as many.

Q: What memory configurations do the two GPUs use?

A: The AMD Ryzen Z1 GPU uses 16 GB of LPDDR5 memory on a 64-bit bus with 51.20 GB/s bandwidth. The Intel Arc B390 uses system shared memory with system dependent bandwidth.

Q: What are the TDP ratings for each GPU?

A: The AMD Ryzen Z1 GPU has a 30 W TDP, and the Intel Arc B390 has an 80 W TDP.

Q: Do both GPUs support the same graphics APIs?

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

Where Each One Wins

The AMD Ryzen Z1 GPU wins on efficiency and form factor considerations. Its 30 W TDP is less than half the 80 W TDP of the Intel Arc B390. It also carries dedicated memory in the form of 16 GB of LPDDR5 with a fixed 51.20 GB/s bandwidth, which provides predictable memory performance independent of the host system. The Z1 GPU's 50th percentile placement indicates that, within the database's distribution, it sits at the median of all GPUs, a far stronger position than the Arc B390's 9th percentile. The Z1 GPU is a discrete card with specified dimensions of 280 mm length, 111 mm height, and 21 mm width, making it a standalone component, whereas the Arc B390 is an integrated processor.

The Intel Arc B390 wins on raw compute specifications. It delivers 7.680 TFLOPS FP32, triple the 2.560 TFLOPS of the Z1 GPU. Its pixel rate of 60.00 GPixel/s and texture rate of 120.0 GTexel/s are each three times higher than the Z1 GPU's corresponding figures of 20.00 GPixel/s and 40.00 GTexel/s. The Arc B390 also has more ray tracing cores, 12 versus 4, and more texture mapping units, 48 versus 16, and more raster operation pipelines, 24 versus 8. Its boost clock of 2500 MHz matches the Z1 GPU's boost clock exactly, while its base clock of 300 MHz is significantly lower.

The benchmark data only records a win for the Intel Arc B390 in the sense that it is the only product with a score. That score, 1482, places it in a cluster with legacy NVIDIA parts, all within 2.7% of its performance. The Z1 GPU has no recorded score, so no direct benchmark victory exists for either side. The database records 0 wins for each product in head-to-head benchmarks.

For use cases, the Z1 GPU's profile suits scenarios where power draw and dedicated memory matter: its 30 W TDP and fixed 51.20 GB/s bandwidth make it a self-contained solution. The Arc B390's profile suits scenarios where raw throughput numbers matter more than measured performance: its triple compute rates and integrated nature target portable devices where a discrete card cannot be installed. The Arc B390's portable device dependent display outputs and system shared memory tie its real-world behavior to the host platform, while the Z1 GPU's no outputs designation indicates it is not intended for direct display connection. The data does not record any overlapping benchmark results, so the performance relationship between these two parts rests on percentile placement and architectural specifications rather than direct measurements.