SYSTEM ANALYZER

Rate My PC: AMD Ryzen AI Max+ PRO 395 + Intel Arc B390

Get a comprehensive performance analysis of your gaming rig with detailed benchmarks, bottleneck detection, and upgrade recommendations

87 / 100
HIGH-END

Power Build

Top 13% of systems. Excellent for 1440p Ultra or 4K High gaming.

1440p Ultra4K High

System Balance Analysis

CPU vs GPU performance ratio
Well Balanced
CPU
97%
VS
GPU
77%
PROCESSOR

AMD Ryzen AI Max+ PRO 395

80,762 Benchmark Score
Top 3% Market Ranking
View Full Specs →
GRAPHICS CARD

Intel Arc B390

1,482 Benchmark Score
Top 23% Market Ranking
View Full Specs →

Market Position

How your build compares to others
Budget
0-30
Mid-Range
30-60
High-End
60-85
Enthusiast
85-100
Your Build

Game Performance Benchmarks

Real-world 4K FPS in popular titles
View All Games →

Performance Insights

Tips to maximize your system

Strong Performance

Excellent for 1440p gaming. Most games will run at high/ultra settings smoothly.

Compatible Games See what you can play Compare CPUs Find upgrades Compare GPUs Find upgrades

Performance Tiers Explained

90-100

Ultimate

4K Ultra gaming, VR ready, ray tracing enabled, professional workloads

4K 60+ FPS VR Ready
70-89

High-End

1440p Ultra or 4K High settings, excellent for modern AAA titles

1440p Ultra 4K High
50-69

Mid-Range

1080p Ultra or 1440p Medium, great value for most gamers

1080p Ultra 1440p Med
30-49

Entry Level

1080p Medium settings, suitable for eSports and older titles

1080p Med eSports
0-29

Legacy

Basic gaming, older titles, consider upgrading for modern games

720p-1080p Low Older Games

The AMD Ryzen AI Max+ PRO 395 and Intel Arc B390 pairing represents a curious configuration: a top-tier 16-core mobile processor paired with an entry-level integrated GPU. This combination produces a system with extreme CPU compute capability but very limited graphics throughput. The following analysis relies exclusively on the provided benchmark data, which shows a CPU that ranks in the 95th percentile against all processors, while the GPU sits in the 9th percentile against all graphics cards. The combined system percentile is 52, indicating a build that is neither uniformly strong nor uniformly weak, but rather heavily skewed toward processor-heavy tasks.

Upgrade Path and Platform

The platform is built around the AMD Socket FP11, a mobile-specific socket that is not designed for user-level processor upgrades. The CPU supports LPDDR5X memory via a quad-channel bus, delivering 256.0 GB/s of memory bandwidth. This is a high-bandwidth configuration that benefits both CPU and GPU workloads, as the GPU memory is marked as "System Shared," meaning it draws from the same pool. ECC memory support is present, which is unusual for a mobile platform and suggests a workstation-oriented design. PCIe connectivity is limited to Gen 4 with 16 lanes from the CPU, which is sufficient for a single discrete GPU or fast NVMe storage, but the integrated nature of this GPU means the PCIe lanes are likely reserved for other peripherals.

The CPU has a TDP of 55 watts, while the GPU has an 80-watt TDP. There is no suggested PSU listed, but the combined thermal envelope of 135 watts, plus the rest of a laptop's components, means a capable power adapter is required. Because this is a laptop build class, the upgrade path is essentially closed. The user cannot swap the CPU or GPU. The sensible next upgrade, therefore, is not internal but external: an eGPU enclosure utilizing the available PCIe Gen 4 lanes, or a completely new system. The data does not support any other internal expansion, as the GPU is an IGP with no power connectors and a slot width of "IGP," meaning it is soldered to the motherboard. The memory bandwidth of 256.0 GB/s is a fixed asset; there is no way to increase it beyond the factory configuration.

Benchmark Performance

The CPU's benchmark scores are exceptional. In Cinebench R23, it achieves 35,061 points in multi-core and 2,012.5 points in single-core. The Passmark multithread score is 51,502, and the single-thread score is 4,078. The average benchmark score across all CPU tests is 80,762, placing it in the 95th percentile of all CPUs. Comparing to its nearest rivals, the AMD Ryzen 9 8940HX scores 81,103, which is 0.4% higher, and the Intel Xeon w5-3535X scores 81,115, also 0.4% higher. Conversely, the Intel Xeon 638 scores 80,723, which is essentially identical with a delta of 0%, and the Intel Xeon w5-2565X scores 80,671, which is 0.1% lower. This places the Ryzen AI Max+ PRO 395 in a tight cluster with workstation-class Xeon processors, indicating that its multi-threaded performance is on par with server-grade silicon.

The GPU presents a starkly different picture. Its only benchmark, 3DMark Steel Nomad DX12, yields a score of 1,482, placing it in the 9th percentile of all GPUs. The nearest rivals are all legacy or entry-level parts: the NVIDIA GeForce GT 520MX scores 1,463 (1.3% lower), the NVIDIA GeForce 800M scores 1,460 (1.5% lower), and the NVIDIA GeForce GT 625 OEM scores 1,446 (2.5% lower). The GPU is only 2.7% faster than the NVIDIA GeForce GT 710. This means the integrated Intel Arc B390 is not competitive with any modern discrete graphics solution. The combined picture is a system that can handle massive parallel processing on the CPU side but will be severely constrained in any graphics-intensive task.

Usage Scenarios

For high-refresh gaming, the data indicates this system is not suitable. The GPU's 9th percentile ranking and 1,482 3DMark score are far too low to drive high frame rates at any resolution above minimal settings. The CPU's 95th percentile performance does not compensate for the GPU bottleneck.

For streaming, the CPU is more than capable of encoding video, given its 16 cores and 32 threads, and the Passmark data encryption score of 32,840 suggests strong hardware encryption support. However, the GPU cannot handle the rendering load of modern games while streaming, so this scenario is limited to non-gaming streams or very light titles.

For video editing, the CPU's Cinebench R23 multi-core score of 35,061 and Passmark floating-point math score of 120,533 indicate excellent performance for rendering and exporting timelines. The GPU, however, is a bottleneck for effects that rely on GPU acceleration, such as real-time previews with heavy filters. The 7.680 TFLOPS FP32 performance is too low for smooth 4K timeline scrubbing.

For 3D rendering, the CPU is the star. The Cinebench scores and the Passmark integer math score of 190,720 suggest that CPU-based rendering in applications like Blender (Cycles) or V-Ray will be very fast. The GPU's 3DMark score of 1,482 suggests that GPU-based rendering is not a viable option, as it would be slower than even a decade-old discrete card.

For software development, the CPU is ideal. The 16 cores and 32 threads handle compilation tasks with ease, and the Passmark multithread score of 51,502 confirms strong parallel throughput. The data compression score of 639,356 indicates fast build artifact handling. The GPU is irrelevant for most development tasks, except for UI rendering, which is fine.

For student and office work, the CPU is overkill, but it ensures that any document, spreadsheet, or browser workload is handled instantly. The GPU is sufficient for basic 2D desktop rendering and video playback, but it is not suitable for any graphical design or CAD work. The system will excel at productivity but fail at any visual task.

Who Should Build It

This build targets a very specific user: a developer, researcher, or data analyst who needs extreme CPU compute power in a laptop form factor but does not require GPU acceleration. The 95th percentile CPU ranking and the 52nd combined percentile make this clear. The target industries are software engineering, financial modeling, scientific computing, and server administration, where the CPU's 16 cores and 32 threads are the primary workhorse. The ECC memory support further targets this group, as data integrity is critical in those fields.

For gamers, this is not the right system. The GPU's 9th percentile ranking is a hard stop for any modern title at 1080p with high settings. Content creators who rely on GPU-accelerated rendering will also be disappointed. However, for a student or small business user whose workload is purely CPU-bound—such as running complex simulations, compiling large codebases, or processing massive datasets—this system is a powerful mobile workstation. The 256.0 GB/s memory bandwidth is a significant asset for these tasks, as it reduces the time spent moving data between memory and CPU.

Balance and Bottleneck

The system is severely unbalanced. The CPU is in the 95th percentile, while the GPU is in the 9th percentile. This is a delta of 86 percentage points, creating a massive bottleneck in any GPU-dependent workload. In gaming, the GPU is the limiting factor. The CPU could easily feed a much faster GPU, but the 1,482 3DMark score means the frame rate will be low regardless of the CPU's capability. The 7.680 TFLOPS FP32 performance is the hard ceiling for graphics throughput.

In CPU-bound workloads, such as video encoding or 3D rendering on the CPU, the GPU is not a factor, and the CPU's 35,061 Cinebench R23 score shines. The memory bandwidth of 256.0 GB/s is also a strength, but it is shared with the GPU, meaning that graphics operations will consume memory bandwidth that could otherwise be used by the CPU. The data indicates that the GPU's memory is "System Shared," which is a common configuration for IGPs but one that reduces overall system performance when the GPU is active. The FPS scaling is not available in the data, but the percentile gap is evidence enough that the GPU will not keep up with the CPU in any graphical task.

FAQ

Q: What is the CPU's multi-core performance relative to its nearest competitors?

A: The CPU's average benchmark score is 80,762, which is 0.4% lower than the AMD Ryzen 9 8940HX and 0.4% lower than the Intel Xeon w5-3535X, but 0% and 0.1% higher than the Intel Xeon 638 and Intel Xeon w5-2565X, respectively.

Q: Is this system suitable for modern gaming?

A: No. The GPU scores 1,482 in 3DMark Steel Nomad DX12, placing it in the 9th percentile, which is too low for modern games at high settings.

Q: What memory type does this platform support?

A: It supports LPDDR5X memory via a quad-channel bus, with a bandwidth of 256.0 GB/s and ECC support.

Q: Can the GPU be upgraded?

A: No, the Intel Arc B390 is an IGP with a slot width of "IGP" and no power connectors, meaning it is integrated into the processor package and cannot be replaced.

Q: How does the CPU compare to a workstation-class Xeon?

A: The CPU's average benchmark score of 80,762 is essentially identical to the Intel Xeon 638's 80,723 (0% delta) and the Intel Xeon w5-2565X's 80,671 (0.1% delta).

Q: What is the combined system percentile?

A: The combined percentile is 52, indicating a system that is average overall due to the extreme imbalance between CPU and GPU performance.

Q: What is the GPU's FP32 floating-point performance?

A: The GPU has a FP32 performance of 7.680 TFLOPS, which is a measure of its single-precision compute capability.

Gaming Performance

The FACT PACK contains no measured FPS rows for this exact combination of CPU and GPU, so all frame rate discussions are estimates based on the benchmark scores. The GPU's 3DMark Steel Nomad DX12 score of 1,482, which places it in the 9th percentile, is the primary indicator. This score is comparable to the NVIDIA GeForce GT 710, which scores 1,443, only 2.7% lower. In practical terms, this means that for any modern game at 1080p with ultra settings, the frame rate would be in the single digits or low teens. At 720p with low settings, the frame rate might reach 30 FPS in older or less demanding titles, but it would still struggle with newer releases.

The CPU's 95th percentile ranking does not help in this scenario, as gaming performance is bound by the GPU's 7.680 TFLOPS FP32 throughput and 60.00 GPixel/s pixel rate. The 12 RT cores are present, but their performance is limited by the same low shader count. The data suggests that this is not a gaming system by any reasonable measure. The estimated frame rates would be unplayable for competitive gaming and only marginally acceptable for turn-based or strategy games that do not rely on fast rendering. Users should treat the GPU as a display adapter for productivity, not a gaming solution.

Build Overview

This is a laptop-class build featuring the AMD Ryzen AI Max+ PRO 395 CPU and the Intel Arc B390 GPU. The CPU is a 16-core, 32-thread processor based on the Zen 5 architecture, codenamed Strix Halo, manufactured on a 4 nm process by TSMC. The GPU is an integrated part of the Intel Panther Lake chip, using the Xe3-LPG architecture on a 3 nm Intel process. The combined percentile is 52, which places it in the middle of all systems, but this average is misleading. The CPU is in the 95th percentile, and the GPU is in the 9th percentile. This is a high-end CPU paired with an entry-level GPU, resulting in a system that is a compute monster but a graphics lightweight. The build class is laptop, meaning all components are soldered and non-upgradeable. The overall tier is "workstation-class CPU performance with integrated graphics limitations."

CPU Analysis

The AMD Ryzen AI Max+ PRO 395 has 16 cores and 32 threads, with a base clock of 3.00 GHz and a boost clock of 5.10 GHz. The TDP is 55 watts, which is low for a 16-core part, indicating efficient power management. The architecture is Zen 5, built on a 4 nm process at TSMC. The cache hierarchy includes 80 KB of L1 per core, 1 MB of L2 per core, and 64 MB of shared L3 cache. The memory support is LPDDR5X over a quad-channel bus, yielding 256.0 GB/s of bandwidth, which is crucial for feeding the 16 cores.

Benchmark results show a Cinebench R23 multi-core score of 35,061 and a single-core score of 2,012.5. The Passmark multithread score is 51,502, and the single-thread score is 4,078. The average benchmark score is 80,762, placing it in the 95th percentile. The nearest rival, the Intel Xeon 638, scores 80,723, a 0% delta, meaning the Ryzen is statistically tied with a Xeon workstation chip. The Passmark integer math score of 190,720 and floating-point math score of 120,533 show strong performance in both integer and floating-point workloads. The data encryption score of 32,840 and data compression score of 639,356 indicate robust throughput for security and archiving tasks. For real workloads, this CPU will handle any multi-threaded task with ease, from 4K video encoding to complex scientific simulations. The 55-watt TDP means it can sustain high clocks without excessive heat, making it suitable for thin laptops.

GPU Analysis

The Intel Arc B390 is an integrated GPU based on the Xe3-LPG architecture, part of the Panther Lake chip, manufactured on a 3 nm process at Intel. It has 1,536 shading units, 48 TMUs, and 24 ROPs, with 12 RT cores. The base clock is 300 MHz, and the boost clock is 2,500 MHz. The memory is marked as "System Shared" with no dedicated VRAM, meaning it uses the system's LPDDR5X memory via the CPU's quad-channel bus. The bandwidth is listed as "System Dependent," but the CPU's 256.0 GB/s is the theoretical maximum. The pixel rate is 60.00 GPixel/s, and the texture rate is 120.0 GTexel/s. The FP32 performance is 7.680 TFLOPS, and FP16 is 15.36 TFLOPS with a 2:1 ratio.

The only benchmark score is 1,482 in 3DMark Steel Nomad DX12, placing it in the 9th percentile. This is a very low score, comparable to the NVIDIA GeForce GT 520MX (1,463) and GT 710 (1,443). The 12 RT cores are present, but their performance is limited by the low overall shader throughput. For rendering, this GPU is not suitable for any modern 3D application. The FP32 performance of 7.680 TFLOPS is about 1/10th of what a modern discrete GPU offers. The 80-watt TDP is also high for an IGP, suggesting it may consume significant power under load, but the performance does not justify it. The GPU supports DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4, so it has the API features, but not the hardware muscle to use them effectively. In practical terms, this GPU is fine for basic desktop rendering, 2D applications, and video playback, but it will struggle with any 3D workload, including gaming and GPU-accelerated rendering.