AMD Ryzen Z2 Extreme GPU vs Intel Arc B390 Comparison

Where Each One Wins

The recorded benchmark data splits cleanly between these two parts. The AMD Ryzen Z2 Extreme GPU and the Intel Arc B390 were measured in exactly one shared test, 3DMark Steel Nomad DX12, and the Intel part took the only win. The AMD solution scored 516, while the Intel Arc B390 scored 1482. That is a delta of negative 65.2 percent from the AMD side, meaning the Intel part finishes roughly 2.9 times higher in this specific workload. The AMD Ryzen Z2 Extreme GPU has zero wins in the head-to-head comparison, while the Intel Arc B390 has one.

The use-case split is therefore not about which part wins in different categories, because the database contains only one shared benchmark. Instead, the split is about what each component is designed to do within its own context. The AMD Ryzen Z2 Extreme GPU is a console-class GPU built into the Strix Point chip, rated at 28 W TDP, with a 4 nm process from TSMC. The Intel Arc B390 is an integrated graphics processor within the Panther Lake chip, built on Intel's 3 nm process, rated at 80 W TDP, and classified as an IGP. The AMD part presents itself as a low-power embedded graphics solution for handheld or compact devices, while the Intel part functions as the graphics block inside a mobile processor.

Looking at the nearest rivals in the database provides additional context for where each part sits. The AMD Ryzen Z2 Extreme GPU scores 516, which places it 3.4 percent below the Intel HD Graphics P4000's average score of 534, and 3.7 percent below the AMD Radeon HD 6870's 536. It sits 6.6 percent above the AMD Radeon HD 6750M's 484, and 9.3 percent below the AMD Radeon HD 6770M's 569. These are old discrete and integrated parts from previous generations, and the AMD Ryzen Z2 Extreme GPU lands in the middle of that group. The Intel Arc B390 scores 1482, which is 1.3 percent above the NVIDIA GeForce GT 520MX's 1463, 1.5 percent above the NVIDIA GeForce 800M's 1460, 2.5 percent above the NVIDIA GeForce GT 625 OEM's 1446, and 2.7 percent above the NVIDIA GeForce GT 710's 1443. The Intel part sits at the top of its immediate cluster, slightly ahead of a set of older NVIDIA mobile and OEM parts.

The percentile data reinforces this picture. The AMD Ryzen Z2 Extreme GPU sits at the 1st percentile among all GPUs in the database, meaning nearly every other recorded part scores higher. The Intel Arc B390 sits at the 9th percentile, which is still low but represents a meaningfully higher position in the overall distribution. For users comparing these two parts directly, the Intel Arc B390 is the stronger performer in the single measured test, while the AMD part is the lower-power, more compact option.

Architecture Differences

The two components come from different manufacturers and use different underlying designs. The AMD Ryzen Z2 Extreme GPU is based on the Strix Point chip and uses the RDNA 3.5 architecture. It is manufactured on a 4 nm process by TSMC, with 34,000 million transistors on a 233 mm² die. That works out to a transistor density of 145.9 million transistors per square millimeter. The Intel Arc B390 is based on the Panther Lake chip and uses the Xe3-LPG architecture. It is manufactured on a 3 nm process by Intel, and the database records its transistor count and die size as unknown. The AMD part belongs to the Console GPU generation from AMD, while the Intel part belongs to the Arc Graphics-M generation for Panther Lake.

The execution resources differ substantially. The AMD part has 1024 shading units, 64 texture mapping units, 48 render output units, and 16 ray tracing cores. The Intel part has 1536 shading units, 48 texture mapping units, 24 render output units, and 12 ray tracing cores. The AMD part has more TMUs and ROPs, while the Intel part has more shading units and slightly fewer ray tracing cores. Neither part lists tensor cores in the database.

Clock behavior also differs. The AMD Ryzen Z2 Extreme GPU has a base clock of 800 MHz and a boost clock of 2700 MHz, with memory clocked at 1000 MHz or 8 Gbps effective. The Intel Arc B390 has a base clock of 300 MHz and a boost clock of 2500 MHz, with memory marked as system shared. The AMD part runs a higher base clock and a higher boost clock, but the Intel part compensates with a larger shading unit count.

Memory configuration is another major split. The AMD part uses 16 GB of LPDDR5X on a 128-bit bus, delivering 128.0 GB/s of bandwidth. The Intel part uses system shared memory, meaning its capacity, bus width, and bandwidth are all system dependent. This is a fundamental difference in how the two parts access memory. The AMD part has dedicated memory resources with a fixed bandwidth figure, while the Intel part relies on whatever the host system provides.

The output and interface configuration also differs. The AMD Ryzen Z2 Extreme GPU lists a single USB Type-C display output. The Intel Arc B390 lists its display outputs as portable device dependent, and its bus interface is IGP. The AMD part does not list a bus interface in the database. Both parts use no power connectors, which aligns with their integrated or embedded nature.

The process node difference is notable. Intel uses a 3 nm process for Panther Lake, while AMD uses a 4 nm process for Strix Point. The database does not provide transistor counts for the Intel part, so a direct density comparison is impossible. The AMD part has a known die size of 233 mm² and a known transistor count of 34,000 million. The Intel part's die size and transistor count remain unrecorded.

The Verdict

The data points to a clear performance winner in the shared 3DMark Steel Nomad DX12 test. The Intel Arc B390 delivers a score of 1482, which is 65.2 percent higher than the AMD Ryzen Z2 Extreme GPU's 516. That is a substantial gap in the only direct comparison available. The Intel part also sits at the 9th percentile among all GPUs, while the AMD part sits at the 1st percentile. For any task that relies on the measured 3DMark workload, the Intel Arc B390 is the stronger choice.

However, the verdict is not simply about raw score. The AMD Ryzen Z2 Extreme GPU operates at 28 W TDP, while the Intel Arc B390 operates at 80 W TDP. The AMD part is built on a 4 nm TSMC process, while the Intel part uses a 3 nm Intel process. The AMD part has its own 16 GB memory pool with 128.0 GB/s bandwidth, while the Intel part depends on system shared memory. The AMD part also has more render output units (48 versus 24) and more texture mapping units (64 versus 48), which could matter in fill-rate-bound scenarios, though the database does not include such tests for these parts.

The recorded data shows the AMD Ryzen Z2 Extreme GPU's pixel rate at 129.6 GPixel/s and texture rate at 172.8 GTexel/s. The Intel Arc B390's pixel rate is 60.00 GPixel/s and texture rate is 120.0 GTexel/s. The AMD part has higher fill rates despite its lower overall 3DMark score. This suggests the AMD part is optimized for specific rendering paths that the Steel Nomad test does not heavily exercise, or that the Intel part's larger shading unit count gives it an advantage in compute-heavy workloads.

For users who prioritize the single measured benchmark, the Intel Arc B390 is the clear pick. For users who prioritize power efficiency, memory bandwidth, or fill rate, the AMD Ryzen Z2 Extreme GPU has advantages that the benchmark score does not capture. The database records no wins for the AMD part, but the architectural data indicates it is not uniformly inferior.

FAQ

Q: Which part scores higher in 3DMark Steel Nomad DX12?

A: The Intel Arc B390 scores 1482, while the AMD Ryzen Z2 Extreme GPU scores 516. The Intel part wins by a delta of negative 65.2 percent from the AMD side.

Q: How do these parts compare to their nearest rivals?

A: The AMD Ryzen Z2 Extreme GPU sits 3.4 percent below the Intel HD Graphics P4000 and 3.7 percent below the AMD Radeon HD 6870, while sitting 6.6 percent above the AMD Radeon HD 6750M. The Intel Arc B390 sits 1.3 percent above the NVIDIA GeForce GT 520MX, 1.5 percent above the NVIDIA GeForce 800M, 2.5 percent above the NVIDIA GeForce GT 625 OEM, and 2.7 percent above the NVIDIA GeForce GT 710.

Q: What memory configurations do the two parts use?

A: The AMD Ryzen Z2 Extreme GPU uses 16 GB of LPDDR5X on a 128-bit bus with 128.0 GB/s bandwidth. The Intel Arc B390 uses system shared memory, with capacity, bus width, and bandwidth all marked as system dependent.

Q: What are the power ratings for each part?

A: The AMD Ryzen Z2 Extreme GPU has a TDP of 28 W. The Intel Arc B390 has a TDP of 80 W.

Q: How do the shading unit counts differ?

A: The Intel Arc B390 has 1536 shading units, while the AMD Ryzen Z2 Extreme GPU has 1024 shading units. The Intel part also has 48 texture mapping units and 24 render output units, while the AMD part has 64 texture mapping units and 48 render output units.

Q: What process nodes are used for each part?

A: The AMD Ryzen Z2 Extreme GPU is manufactured on a 4 nm process by TSMC. The Intel Arc B390 is manufactured on a 3 nm process by Intel.

Head-to-Head Benchmarks

The only shared benchmark in the database is 3DMark Steel Nomad DX12. The Intel Arc B390 scored 1482, and the AMD Ryzen Z2 Extreme GPU scored 516. The delta is negative 65.2 percent from the AMD perspective, meaning the AMD part trails by roughly two-thirds of its own score. The Intel part's score is approximately 2.87 times the AMD part's score.

This single result dominates the head-to-head comparison. The Intel Arc B390 takes the win in the one recorded test, and the AMD Ryzen Z2 Extreme GPU takes zero wins. The magnitude of the gap is large enough that no other measured data in the database can offset it. The Intel part's nearest rivals cluster around the 1443 to 1463 range, and the Intel part leads all of them by small margins. The AMD part's nearest rivals cluster around the 484 to 569 range, and the AMD part sits in the middle of that group.

The fill rate data provides a counterpoint. The AMD Ryzen Z2 Extreme GPU has a pixel rate of 129.6 GPixel/s and a texture rate of 172.8 GTexel/s. The Intel Arc B390 has a pixel rate of 60.00 GPixel/s and a texture rate of 120.0 GTexel/s. The AMD part is more than twice as fast in pixel throughput and about 44 percent faster in texture throughput. Yet the Intel part still wins the 3DMark test by a wide margin. This indicates that the Steel Nomad workload is not primarily fill-rate bound, or that the Intel part's additional shading units provide a decisive advantage in the shader-heavy portions of the test.

The FP32 compute figures tell a similar story. The Intel Arc B390 delivers 7.680 TFLOPS of FP32 performance, while the AMD Ryzen Z2 Extreme GPU delivers 5.530 TFLOPS. The Intel part is about 39 percent higher in raw FP32 throughput. The FP16 figures diverge even more: the Intel part delivers 15.36 TFLOPS with a 2:1 ratio, while the AMD part delivers 5.530 TFLOPS with a 1:1 ratio. The Intel part's FP16 throughput is nearly three times the AMD part's, which could matter in workloads that use reduced precision.

The memory bandwidth difference runs the other way. The AMD part has a fixed 128.0 GB/s from its LPDDR5X memory, while the Intel part's bandwidth is system dependent and therefore not directly comparable. The database does not record a bandwidth figure for the Intel part, so no numerical comparison is possible.

Specification Differences

The two parts differ across nearly every recorded specification field. The AMD Ryzen Z2 Extreme GPU uses the Strix Point chip with RDNA 3.5 architecture, while the Intel Arc B390 uses the Panther Lake chip with Xe3-LPG architecture. The AMD part is manufactured by TSMC on a 4 nm process, while the Intel part is manufactured by Intel on a 3 nm process. The AMD part has 34,000 million transistors on a 233 mm² die, while the Intel part's transistor count and die size are unknown.

The clock specifications differ. The AMD part has a base clock of 800 MHz and a boost clock of 2700 MHz, with memory at 1000 MHz or 8 Gbps effective. The Intel part has a base clock of 300 MHz and a boost clock of 2500 MHz, with memory marked as system shared.

The memory configuration is fundamentally different. The AMD part uses 16 GB of LPDDR5X on a 128-bit bus with 128.0 GB/s bandwidth. The Intel part uses system shared memory, with all memory fields marked as system dependent or system shared.

The execution resources differ in both count and distribution. The AMD part has 1024 shading units, 64 TMUs, 48 ROPs, and 16 ray tracing cores. The Intel part has 1536 shading units, 48 TMUs, 24 ROPs, and 12 ray tracing cores. The AMD part has more TMUs, ROPs, and ray tracing cores, while the Intel part has more shading units.

The compute rates differ. The AMD part has a pixel rate of 129.6 GPixel/s, a texture rate of 172.8 GTexel/s, and FP32 of 5.530 TFLOPS with FP16 at 5.530 TFLOPS (1:1). The Intel part has a pixel rate of 60.00 GPixel/s, a texture rate of 120.0 GTexel/s, and FP32 of 7.680 TFLOPS with FP16 at 15.36 TFLOPS (2:1).

The power figures differ. The AMD part has a TDP of 28 W, while the Intel part has a TDP of 80 W. Both parts use no power connectors. The Intel part is listed as an IGP with a slot width of IGP and a bus interface of IGP, while the AMD part does not list a slot width or bus interface.

The display outputs differ. The AMD part lists 1x USB Type-C, while the Intel part lists portable device dependent outputs. Both parts support DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. The release dates differ, with the AMD part released on 2025-07-08 and the Intel part on 2026-01-26. Neither part lists a launch MSRP in the database.