SYSTEM ANALYZER

Rate My PC: AMD Ryzen AI Max+ 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+ 395

77,740 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

# Balance and Bottleneck

The pairing of the AMD Ryzen AI Max+ 395 with the Intel Arc B390 presents one of the most lopsided performance balances in the database. The CPU sits at the 95th percentile among all processors, while the GPU rests at just the 9th percentile among all graphics cards. This is not a subtle gap; it is a chasm. The data strongly indicates that the GPU is the limiting factor in virtually every graphics-bound workload, while the CPU operates with massive headroom.

Benchmark results confirm this interpretation. The processor's average benchmark score is 77,740, placing it within 1.6% of the AMD Ryzen Threadripper PRO 9945WX and 1.7% above the Intel Core i9-14900K. Meanwhile, the GPU's average score of 1,482 in 3DMark Steel Nomad DX12 places it just 1.3% ahead of the NVIDIA GeForce GT 520MX and 2.7% ahead of the GeForce GT 710. The GPU's performance tier aligns with entry-level discrete graphics from over a decade ago, while the CPU competes with modern flagship desktop processors.

In gaming workloads, the bottleneck is unequivocal. The CPU can feed frames far faster than the GPU can render them. The processor's single-core performance, demonstrated by a Geekbench single-core score of 2,464 and a Cinebench R23 single-core score of 2,044, is more than sufficient for any modern game engine. The GPU, however, with only 7.680 TFLOPS of FP32 throughput and a 60.00 GPixel/s pixel rate, will constrain frame rates well below what the CPU could theoretically support.

For productivity and compute workloads, the balance shifts. The CPU's 16 cores and 32 threads, combined with 64 MB of shared L3 cache, drive multi-threaded tasks like rendering and compilation with exceptional force. The GPU's contribution to these workloads is minimal, given its low compute throughput. The PassMark multi-thread score of 54,934 and integer math score of 200,665 suggest that CPU-bound tasks will see near-full utilization of the processor's capabilities, while the GPU idles or contributes marginally.

The memory subsystem further complicates the balance. The CPU supports quad-channel LPDDR5X with 256.0 GB/s of bandwidth, an enormous figure for a mobile platform. The GPU, however, uses system-shared memory with bandwidth described as "System Dependent." This means the GPU's effective memory performance will vary based on the system configuration, but it will never approach the dedicated bandwidth of a discrete graphics card. In practice, the shared memory architecture means the CPU and GPU compete for the same bandwidth pool, potentially reducing performance in mixed workloads.

# Benchmark Performance

The AMD Ryzen AI Max+ 395 delivers exceptional processor benchmarks. In Cinebench R23, it scores 35,314 in multi-core and 2,044 in single-core. The multi-core result places it in the upper echelon of mobile processors, rivaling desktop-class chips. The Geekbench scores reinforce this: 20,992 multi-core and 2,464 single-core. The PassMark suite shows strong results across varied workloads, including 690,650 in data compression, 354,55 in data encryption, 128,682 in floating point math, and 200,665 in integer math.

The CPU's percentile ranking of 95 among all CPUs means it outperforms approximately 95% of processors in the database. Its nearest rivals provide context: the Intel Core i9-14900K scores 79,097 on average, which is 1.7% higher, while the AMD Ryzen Threadripper PRO 9945WX scores 76,513, just 1.6% lower. The AMD EPYC Embedded 8224P and AMD Ryzen 9 8945HX trail by 1.6% and 2.0% respectively. This places the Ryzen AI Max+ 395 firmly in flagship territory, despite its 55 W TDP.

The Intel Arc B390 presents a stark contrast. Its single benchmark result, a 3DMark Steel Nomad DX12 score of 1,482, places it at the 9th percentile among all GPUs. The nearest rivals are all entry-level NVIDIA parts: the GeForce GT 520MX (1,463, 1.3% slower), the GeForce 800M (1,460, 1.5% slower), the GeForce GT 625 OEM (1,446, 2.5% slower), and the GeForce GT 710 (1,443, 2.7% slower). The GPU's average score of 1,482 is its only data point, making its performance characterization straightforward: it is an entry-level integrated graphics solution.

The combined percentile for this CPU+GPU pairing is 52, which reflects the averaging of the two components' very different performance tiers. The combined picture is one of extreme asymmetry: a top-tier processor paired with a bottom-tier GPU. This configuration will excel in CPU-intensive tasks and struggle in graphics-intensive applications. The data suggests this pairing is designed for compute-focused workloads where GPU acceleration is secondary, rather than for gaming or GPU-accelerated rendering.

# FAQ

Q: How does the CPU compare to the Intel Core i9-14900K?

A: The Ryzen AI Max+ 395 has an average benchmark score of 77,740, which is 1.7% lower than the Core i9-14900K's 79,097. This places the two processors in the same performance class, despite the AMD chip operating at a 55 W TDP in a mobile form factor.

Q: What is the GPU's performance tier relative to other graphics cards?

A: The Arc B390 sits at the 9th percentile among all GPUs. Its 3DMark Steel Nomad score of 1,482 is just 1.3% higher than the GeForce GT 520MX and 2.7% higher than the GeForce GT 710, placing it in entry-level territory.

Q: Does the CPU support ECC memory?

A: Yes, the Ryzen AI Max+ 395 supports ECC memory. It also supports quad-channel LPDDR5X memory with a bandwidth of 256.0 GB/s.

Q: What socket does the CPU use?

A: The Ryzen AI Max+ 395 uses AMD Socket FP11. It is based on the Zen 5 architecture with the codename Strix Halo, manufactured on a 4 nm process by TSMC.

Q: How much cache does the CPU have?

A: The CPU has 80 KB of L1 cache per core, 1 MB of L2 cache per core, and 64 MB of shared L3 cache. The total L3 cache is shared across all 16 cores.

Q: What is the GPU's architecture and process node?

A: The Intel Arc B390 uses the Xe3-LPG architecture, part of the Arc Graphics-M generation based on Panther Lake. It is manufactured on a 3 nm process by Intel.

Q: Is the GPU's memory dedicated or shared?

A: The GPU uses system-shared memory. Its memory size, type, bus width, and bandwidth are all described as "System Shared" or "System Dependent," meaning it relies on the system's LPDDR5X memory rather than having dedicated VRAM.

# Gaming Performance

No measured FPS data exists for this exact CPU+GPU combination. The FACT PACK contains no measuredFpsUltraByGame entries for this pairing, so all gaming performance figures discussed here are estimates derived from the benchmark scores. The data is not measured; it is inferred from component-level benchmarks.

Based on the GPU's 3DMark Steel Nomad DX12 score of 1,482, which places it at the 9th percentile, gaming performance will be extremely limited. The GPU's FP32 throughput of 7.680 TFLOPS and pixel rate of 60.00 GPixel/s are characteristic of entry-level integrated graphics. At 1080p with ultra settings, most modern games would likely run at low frame rates, potentially below 30 FPS in demanding titles. At 1440p or 4K, playable frame rates are unlikely for anything beyond esports titles or older games.

The CPU's gaming capability is not the constraint. Its single-core performance, with a Cinebench R23 score of 2,044 and Geekbench score of 2,464, is more than adequate for gaming. The bottleneck is entirely the GPU. For games that are CPU-bound or less graphically demanding, such as strategy games or indie titles, the system may perform acceptably. However, the GPU's nearest rivals include the GeForce GT 710 and GT 520MX, which are not considered gaming-capable by modern standards.

The shared memory architecture compounds the issue. The GPU's bandwidth is "System Dependent," meaning it shares the CPU's 256.0 GB/s quad-channel LPDDR5X bandwidth. While this bandwidth is generous for a mobile platform, the GPU's access to it will be constrained by the system's memory controller and the CPU's competing demands. Estimates suggest that gaming performance will be comparable to very old discrete GPUs, suitable for light or legacy titles but inadequate for current AAA releases at acceptable quality settings.

# Who Should Build It

This configuration targets users who prioritize CPU performance above all else and require graphics only for basic display output or light acceleration. The Ryzen AI Max+ 395's 95th percentile CPU ranking makes it an excellent choice for content creators working in CPU-bound applications like video encoding, 3D rendering, and software compilation. The 16 cores and 32 threads, combined with 64 MB of L3 cache, handle multi-threaded workloads with flagship-class performance.

Software developers will find this pairing valuable. The CPU's strong single-threaded performance (Geekbench 2,464 single-core) speeds up compilation and code analysis, while the multi-threaded score of 20,992 accelerates parallel build processes. The PassMark extended instructions score of 56,346 indicates robust SIMD and vector processing capability, beneficial for scientific computing and data analysis.

Students and professionals in engineering, finance, or data science fields who run CPU-intensive simulations or statistical analyses will benefit from the processor's performance. The 55 W TDP means this performance comes in a power-efficient package, suitable for laptop deployments where thermal and power constraints are significant.

Small business workstations handling database operations, spreadsheet analysis, or virtualization can leverage the CPU's multi-threading. The PassMark data compression score of 690,650 and random string sorting score of 76,177 suggest strong performance in data manipulation tasks. However, any workload requiring GPU acceleration, such as machine learning training or GPU-based video encoding, will be severely limited by the Arc B390's 9th percentile performance.

Gamers should not consider this build for modern gaming. The GPU's performance tier is simply too low for acceptable frame rates in current titles. This is not a gaming configuration; it is a CPU-centric compute platform with basic graphics capability.

# GPU Analysis

The Intel Arc B390 is an integrated GPU based on the Xe3-LPG architecture, manufactured on Intel's 3 nm process. It operates with a base clock of 300 MHz and a boost clock of 2500 MHz. The GPU has 1,536 shading units, 48 texture mapping units, and 24 raster output units. It includes 12 ray tracing cores, though their performance is constrained by the overall compute throughput.

Memory is entirely system-shared. The GPU has no dedicated VRAM; its memory size, type, and bus width are all "System Shared." Memory bandwidth is "System Dependent," meaning it varies based on the host system's configuration. In this pairing, the GPU shares the CPU's quad-channel LPDDR5X bandwidth of 256.0 GB/s. This shared architecture means the GPU's effective memory performance will be lower than dedicated graphics memory due to contention with the CPU.

Compute performance is modest. The GPU delivers 7.680 TFLOPS of FP32 performance and 15.36 TFLOPS of FP16 performance (at a 2:1 ratio). The pixel rate is 60.00 GPixel/s, and the texture rate is 120.0 GTexel/s. These figures place the GPU in entry-level territory. The 3DMark Steel Nomad DX12 score of 1,482 confirms this, ranking at the 9th percentile and within 1.3% of the GeForce GT 520MX.

The GPU supports modern APIs including DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4. This means it is feature-complete in terms of API support, but the raw performance is insufficient to leverage these features at playable frame rates in demanding applications. The ray tracing cores exist but will provide minimal practical benefit given the overall compute limitations.

For rendering workloads, the GPU's 7.680 TFLOPS of FP32 compute is insufficient for real-time 3D rendering or GPU-accelerated rendering in applications like Blender or Octane. The 15.36 TFLOPS of FP16 performance could assist in some machine learning inference tasks, but the shared memory architecture and lack of dedicated tensor cores limit its utility. The GPU is best suited for basic display output, video playback, and light 2D acceleration.

# CPU Analysis

The AMD Ryzen AI Max+ 395 is a 16-core, 32-thread processor based on the Zen 5 architecture with the codename Strix Halo. It operates with a base clock of 3.00 GHz and a boost clock of 5.10 GHz. The processor is manufactured on TSMC's 4 nm process with a die size of 2x 70.6 mm². Its 55 W TDP is remarkably low for the performance it delivers.

The cache hierarchy is substantial: 80 KB of L1 cache per core, 1 MB of L2 cache per core, and 64 MB of shared L3 cache. The large shared L3 cache benefits multi-threaded workloads by reducing memory latency and improving data locality across cores. The CPU supports quad-channel LPDDR5X memory with 256.0 GB/s of bandwidth and ECC memory support.

Benchmark results demonstrate flagship-class performance. The Cinebench R23 multi-core score of 35,314 places the CPU within 1.7% of the Intel Core i9-14900K's average score of 79,097 (comparing average benchmark scores). The Geekbench multi-core score of 20,992 and single-core score of 2,464 indicate strong performance across both parallel and single-threaded workloads. The PassMark multi-thread score of 54,934 and single-thread score of 4,147 confirm the processor's balanced capability.

The CPU's 95th percentile ranking among all processors means it outperforms the vast majority of CPUs in the database. Its nearest rivals are all high-end desktop or server processors: the Threadripper PRO 9945WX (1.6% lower), EPYC Embedded 8224P (1.6% lower), and Ryzen 9 8945HX (2.0% lower). Only the Core i9-14900K scores higher, by 1.7%. This is exceptional for a mobile processor with a 55 W TDP.

Real-world workload implications are significant. The PassMark data encryption score of 35,455 and extended instructions score of 56,346 indicate strong cryptographic and SIMD performance. The floating point math score of 128,682 and integer math score of 200,665 show robust compute capability for scientific and financial applications. The find prime numbers score of 303 is modest, suggesting some workloads may not scale perfectly with core count.

# Upgrade Path and Platform

The CPU uses AMD Socket FP11, a mobile-specific socket with no traditional upgrade path within the same platform. The processor is not multiplier-unlocked, limiting overclocking potential. The platform supports quad-channel LPDDR5X memory with a maximum bandwidth of 256.0 GB/s and ECC support. This is a soldered or embedded platform characteristic of high-end mobile workstations.

PCIe support is Gen 4 with 16 lanes available from the CPU. This provides adequate bandwidth for NVMe storage and external GPU connections, though the integrated GPU does not use PCIe lanes. Memory is system-shared between the CPU and GPU, which simplifies the platform but limits GPU memory performance.

The GPU's TDP is 80 W, and it has no power connectors, as it is an integrated part of the platform. The suggested PSU is not specified, but the combined CPU TDP of 55 W and GPU TDP of 80 W suggest a modest total system power draw. For a laptop platform, this is reasonable, though the actual power delivery depends on the specific system design.

A sensible next upgrade would focus on the GPU, but this is not possible in a laptop build. The integrated nature of the Arc B390 means it cannot be replaced. For users requiring better graphics performance, the only upgrade path is a different system entirely. Within the same platform, upgrades are limited to memory capacity (within the LPDDR5X standard) and storage (via PCIe Gen 4 NVMe).

For users considering this platform for future-proofing, the CPU's performance headroom is substantial. The 16 cores and 32 threads will remain relevant for years, and the 64 MB L3 cache helps with evolving workloads. However, the GPU's entry-level performance will increasingly lag as software demands more graphics compute. The platform is best suited for users who prioritize CPU performance and accept the GPU as a basic display output solution.

# Build Overview

This is a laptop-class build combining the AMD Ryzen AI Max+ 395 processor with the Intel Arc B390 integrated GPU. The build class is explicitly "laptop," and the GPU is designated as an IGP (Integrated Graphics Processor) with a slot width of "IGP" and no power connectors. This is a unified platform where the CPU and GPU share the same memory subsystem and power delivery.

The overall tier of this build, based on the combined percentile of 52, places it in the middle of the database. However, this average masks the extreme disparity between components. The CPU at the 95th percentile is a flagship-class processor, while the GPU at the 9th percentile is entry-level. The combined percentile of 52 reflects the arithmetic mean of these two very different performance tiers.

The CPU is a 16-core Zen 5 processor with a 55 W TDP, manufactured on TSMC's 4 nm process. It uses AMD Socket FP11 and supports quad-channel LPDDR5X memory with 256.0 GB/s bandwidth. The GPU is an Intel Arc B390 based on Xe3-LPG architecture, manufactured on Intel's 3 nm process, with 1,536 shading units and a boost clock of 2500 MHz.

This pairing is unusual in that it combines a top-tier CPU from AMD with an integrated GPU from Intel. This is not a typical configuration; most AMD platforms would pair with AMD Radeon graphics. The inclusion of the Intel Arc B390 suggests a system designed for specific workloads where CPU performance is paramount and GPU acceleration is secondary or unnecessary.

The build's overall character is that of a powerful compute workstation in a laptop form factor. It excels at CPU-bound tasks like compilation, data analysis, and multi-threaded rendering. It struggles with GPU-bound tasks like gaming, GPU-accelerated machine learning, or real-time 3D visualization. The 52nd combined percentile accurately reflects this split personality: above average overall, but with significant limitations in graphics-heavy applications.

# Usage Scenarios

High-Refresh Gaming: This is not viable. The GPU's 3DMark Steel Nomad score of 1,482 places it at the 9th percentile, and its nearest rivals are the GeForce GT 520MX and GT 710. Frame rates in modern games would be very low, even at 1080p with reduced settings. High-refresh displays would be wasted on this configuration.

Streaming: CPU-based streaming is feasible. The 16-core processor with a Cinebench R23 multi-core score of 35,314 can handle game capture and encoding via software x264. However, the GPU cannot run modern games at playable frame rates, so streaming gameplay is impractical. Streaming desktop content or light applications would work well.

Video Editing: CPU-based editing is strong. The processor's multi-threaded performance and 64 MB L3 cache handle timeline scrubbing and effects processing efficiently. However, GPU-accelerated effects and encoding are limited by the Arc B390's 7.680 TFLOPS of FP32 compute. Export times will be CPU-bound and acceptable, but GPU-accelerated previews will be slow.

3D Rendering: CPU rendering is excellent. The 16 cores and 32 threads provide substantial compute for render engines like Blender Cycles or V-Ray CPU. The PassMark floating point math score of 128,682 indicates strong numerical performance. GPU rendering is not viable due to the GPU's low compute throughput and shared memory constraints.

Software Development: This is a primary use case. The CPU's single-core Geekbench score of 2,464 and multi-core score of 20,992 accelerate compilation, testing, and code analysis. The extended instructions score of 56,346 supports SIMD-heavy workloads. Developers building large codebases will see significant time savings.

Student and Office Work: This is overkill for typical office tasks. The CPU provides enormous headroom for document processing, spreadsheet analysis, and web browsing. The PassMark data compression score of 690,650 and random string sorting score of 76,177 indicate rapid data manipulation. The low GPU performance is irrelevant for these tasks. Students in engineering or data science fields will benefit from the CPU's compute power for simulations and statistical analysis.