SYSTEM ANALYZER

Rate My PC: AMD Ryzen AI Max 390 + NVIDIA GeForce RTX 3070 Mobile

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

93 / 100
ULTIMATE READY

Apex Performer

Top 7% of systems. Capable of 4K Ultra gaming and advanced rendering.

4K 60+ FPSVR ReadyRay Tracing

System Balance Analysis

CPU vs GPU performance ratio
Well Balanced
CPU
95%
VS
GPU
91%
PROCESSOR

AMD Ryzen AI Max 390

56,273 Benchmark Score
Top 5% Market Ranking
View Full Specs →
GRAPHICS CARD

NVIDIA GeForce RTX 3070 Mobile

20,534 Benchmark Score
Top 9% 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

Optimal Performance

Your system is in the top tier. You can run any modern game at maximum settings.

4K Gaming Ready

Consider a 4K 144Hz monitor to fully utilize your hardware capabilities.

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

# FAQ

Q: What is the performance tier of the AMD Ryzen AI Max 390 processor?

A: The CPU sits at the 91st percentile among all tested processors, placing it firmly in the high-end mobile segment. Its average benchmark score of 56273 puts it statistically on par with the AMD Ryzen AI 9 HX PRO 470 (56306, 0.1% faster) and the AMD Ryzen Threadripper PRO 3955WX (56555, 0.5% faster), while edging out the Intel Core i9-14900HX by 0.5% and the Intel Core 7 253PQE by 0.6%.

Q: How does the NVIDIA GeForce RTX 3070 Mobile compare to its closest rivals?

A: The GPU holds the 65th percentile among all graphics cards, with an average benchmark score of 20534. It trades blows with the Intel Arc B570 (20556, 0.1% ahead) and Intel Arc A750 (20582, 0.2% ahead), while leading the NVIDIA Quadro M4000M by 0.3% and the AMD Radeon R9 M390X by 0.6%. The differences are within a single percentage point, indicating near-identical overall compute performance.

Q: What is the combined performance percentile for this laptop pairing?

A: The combined percentile for the Ryzen AI Max 390 and RTX 3070 Mobile is 78, which reflects a system that is substantially above average but not at the extreme flagship tier. The CPU is the stronger component by a wide margin, given its 91st percentile rank versus the GPU's 65th.

Q: What are the key specifications of the AMD Ryzen AI Max 390?

A: This is a 12-core, 24-thread Zen 5 processor built on TSMC's 4 nm process, with a base clock of 3.20 GHz and a boost clock of 5.00 GHz. It features 64 MB of shared L3 cache, 80 KB of L1 cache per core, and 1 MB of L2 cache per core. The chip supports LPDDR5X memory over a quad-channel bus with 256.0 GB/s of bandwidth and has a 55 W TDP.

Q: What are the key specifications of the NVIDIA GeForce RTX 3070 Mobile?

A: The GPU is built on the GA104 chip using Samsung's 8 nm process, with 17,400 million transistors. It offers 5120 shading units, 160 texture mapping units, 80 ROPs, 40 ray tracing cores, and 160 tensor cores. Clock speeds are 1110 MHz base and 1560 MHz boost, with 8 GB of GDDR6 memory on a 256-bit bus providing 448.0 GB/s of bandwidth. The TDP is 115 W.

Q: Is there measured frame rate data for this specific CPU-GPU combination?

A: No. The data shows no measured FPS rows for this exact pairing. All frame rate discussions must be treated as estimates derived from the benchmark scores of the individual components, not from direct gaming tests of this configuration.

Q: What socket and platform does the Ryzen AI Max 390 use?

A: The processor uses AMD Socket FP11 and is part of the Strix Halo platform (codename). It supports PCIe Gen 4 with 16 lanes from the CPU, includes a Radeon 8050S integrated graphics solution, and features ECC memory support. The production status is active, with a release date of January 5, 2025.

# Usage Scenarios

High-refresh gaming: The RTX 3070 Mobile's PassMark G3D score of 15309 and 3DMark Steel Nomad DX12 score of 2380 indicate it can handle competitive esports titles at high frame rates, though the 65th percentile GPU rank suggests that reaching maximum refresh rates at high settings in demanding AAA games will require compromises. The CPU's single-thread score of 5091 in Cinebench R23 ensures the processor will not bottleneck frame generation in most scenarios, but the GPU is the limiting factor for 144 Hz-plus gaming at high detail.

Streaming: The combination of a 12-core CPU with 24 threads and a GPU with 160 tensor cores provides strong headroom for simultaneous gaming and encoding workloads. The CPU's PassMark multithread score of 41737 and data compression score of 487145 indicate that software encoding will not cripple gaming performance, while the tensor cores enable hardware-accelerated encoding paths. The 256.0 GB/s memory bandwidth further supports smooth multitasking between the game, encoder, and streaming software.

Video editing: The Ryzen AI Max 390 excels in this workload, with a Cinebench R23 multicore score of 36064 that places it in the 91st percentile of all CPUs. The GPU's 15.97 TFLOPS FP32 performance and 448.0 GB/s memory bandwidth accelerate effects, color grading, and timeline rendering. The combination of 64 MB of L3 cache and quad-channel LPDDR5X memory ensures large media files and preview caches remain responsive during scrubbing and export.

3D rendering: CPU-based rendering will be exceptionally strong given the multicore performance, with the PassMark floating point math score of 90594 and integer math score of 146519 demonstrating substantial compute throughput. GPU-accelerated rendering will be moderate, as the RTX 3070 Mobile's 40 RT cores and 160 tensor cores provide capable but not class-leading ray tracing and denoising performance. The 8 GB VRAM may limit scene complexity for very large projects.

Software development: The 24 threads and 64 MB L3 cache handle compilation workloads effectively, with the PassMark extended instructions score of 38716 indicating strong SIMD throughput. The data encryption score of 25097 and random string sorting score of 53113 show balanced performance for build tools, code analysis, and version control operations. The 4 nm Zen 5 architecture at 5.00 GHz boost provides snappy single-threaded response for IDE interactions.

Student and office work: This system is substantially over-spec for typical productivity tasks, but the headroom ensures future-proofing. The Radeon 8050S integrated graphics provide a fallback if the discrete GPU is disabled for battery savings, and the CPU's PassMark single-thread score of 4028 handles spreadsheet, document, and web browsing workloads with ease. The 55 W CPU TDP and efficient 4 nm process contribute to reasonable power characteristics for a laptop class device.

# Benchmark Performance

The AMD Ryzen AI Max 390 delivers outstanding processor performance, ranking at the 91st percentile among all CPUs with an average benchmark score of 56273. In Cinebench R23, the multicore score reaches 36064, while the single-core score hits 5091, demonstrating a balanced architecture that excels in both heavily threaded and latency-sensitive tasks. The PassMark multithread score of 41737 confirms the strong scaling across 12 cores and 24 threads, while the integer math score of 146519 and floating point math score of 90594 show robust ALU and FPU throughput. Data compression at 487145 and encryption at 25097 further illustrate the chip's versatility across different workload types.

The NVIDIA GeForce RTX 3070 Mobile performs at the 65th percentile, with an average benchmark score of 20534. Its 3DMark Steel Nomad DX12 score of 2380 reflects modern DirectX 12 gaming capability, while Geekbench OpenCL (92939) and Vulkan (86768) scores indicate strong general compute performance. The PassMark G3D score of 15309 is the most relevant gaming metric, and the GPU compute score of 6827 shows reasonable, though not exceptional, compute throughput. The nearest rivals—Intel Arc B570, Intel Arc A750, NVIDIA Quadro M4000M, and AMD Radeon R9 M390X—are all within 0.6% of the RTX 3070 Mobile's average score, placing it in a very tight competitive band.

The combined picture is a system where the CPU is a top-tier mobile processor, outperforming the Intel Core i9-14900HX by 0.5% and matching desktop-class workstation chips like the Threadripper PRO 3955WX. The GPU is competent but mid-pack, roughly equivalent to desktop Intel Arc discrete cards. This asymmetry means the system's overall 78th combined percentile is driven significantly by the CPU's strength, with the GPU pulling the average down. For FPS estimation, no measured frame rates exist for this pairing; however, based on the GPU's 3DMark and PassMark scores relative to its peers, expect playable frame rates at medium-to-high settings in modern titles, with the CPU never being the bottleneck.

# Who Should Build It

Gamers seeking high-refresh 1080p experiences will find the RTX 3070 Mobile sufficient for maintaining elevated frame rates in competitive titles, though the 65th percentile GPU rank means settings may need tuning for the most demanding releases. The CPU's 91st percentile ensures frame pacing remains consistent, and the combined 78th percentile system percentile indicates a well-above-average gaming laptop.

Content creators working with video and photo editing are the primary beneficiaries of this configuration. The CPU's Cinebench R23 multicore score of 36064 and PassMark multithread score of 41737 accelerate exports, encodings, and batch processing, while the GPU's 8 GB VRAM and 448.0 GB/s bandwidth handle GPU-accelerated effects and previews. The 256.0 GB/s system memory bandwidth further aids large project files.

Software developers and data analysts will appreciate the 24 threads and strong integer math performance (146519) for compilation, testing, and data manipulation. The ECC memory support adds reliability for long-running compute tasks, and the PCIe Gen 4 x16 GPU interface ensures fast data transfer for CUDA-accelerated workloads.

Students and professionals in engineering or scientific fields can leverage the CPU's floating point math score of 90594 for simulations and the GPU's CUDA cores for parallel compute, though the 8 GB VRAM may cap the size of datasets that can be processed on the GPU.

Small business workstations that require a single laptop for mixed workloads—spreadsheets, presentations, light 3D modeling, and occasional video calls—will find this system far more capable than needed, offering years of headroom before performance becomes a concern. The active production status of the CPU and the mature driver stack for the GPU make this a reliable choice for longevity.

# CPU Analysis

The AMD Ryzen AI Max 390 is a 12-core, 24-thread processor built on the Zen 5 architecture, codenamed Strix Halo, using TSMC's 4 nm process node. It operates with a base clock of 3.20 GHz and a boost clock of 5.00 GHz, with a 55 W TDP. The cache hierarchy includes 80 KB of L1 per core, 1 MB of L2 per core, and a substantial 64 MB of shared L3 cache, which is critical for reducing memory latency in multi-threaded workloads.

Benchmark results show exceptional multicore performance, with a Cinebench R23 multicore score of 36064 placing it at the 91st percentile. This is stronger than the Intel Core i9-14900HX (56004 average, 0.5% slower) and nearly identical to the AMD Ryzen AI 9 HX PRO 470 (56306, 0.1% faster). The single-core Cinebench R23 score of 5091 indicates strong per-thread performance, essential for gaming and lightly threaded applications. The PassMark single-thread score of 4028 corroborates this.

Real-world workload analysis: the data compression score of 487145 and encryption score of 25097 suggest robust performance for archiving, database operations, and secure communications. The extended instructions score of 38716 shows strong SIMD throughput for scientific and multimedia code. The find prime numbers score of 316 is modest, indicating that certain integer-heavy algorithms may not scale as well, but the integer math score of 146519 offsets this concern for general compute. The floating point math score of 90594 is solid for physics simulations and 3D rendering.

The memory subsystem is a key strength: quad-channel LPDDR5X support with 256.0 GB/s bandwidth and ECC capability makes this CPU suitable for memory-intensive workloads like large dataset processing and virtual machines. The integrated Radeon 8050S GPU provides a fallback for display output and light graphics tasks, which is useful for power saving or troubleshooting.

# GPU Analysis

The NVIDIA GeForce RTX 3070 Mobile is built on the GA104 chip using Samsung's 8 nm process, with 17,400 million transistors on a 392 mm² die. It features 5120 shading units, 160 TMUs, and 80 ROPs, with clock speeds of 1110 MHz base and 1560 MHz boost. The memory subsystem consists of 8 GB of GDDR6 on a 256-bit bus, delivering 448.0 GB/s of bandwidth at 14 Gbps effective speed.

The GPU's benchmark scores place it at the 65th percentile, with an average score of 20534. The 3DMark Steel Nomad DX12 score of 2380 reflects modern DirectX 12 Ultimate (12_2) gaming performance, while the Geekbench OpenCL (92939) and Vulkan (86768) scores show strong cross-API compute capability. The PassMark G3D score of 15309 is the primary gaming indicator, and the DirectX 11 score of 138 and DirectX 12 score of 64 from PassMark suggest varying performance across different API versions, with DirectX 12 being the strongest relative to peers.

The 40 RT cores and 160 tensor cores enable hardware-accelerated ray tracing and DLSS, respectively, although the 65th percentile rank indicates these features perform at a mid-range level. The pixel rate of 124.8 GPixel/s and texture rate of 249.6 GTexel/s are moderate for the class, supporting 1080p and 1440p gaming with adjusted settings. The FP32 performance of 15.97 TFLOPS (with 1:1 FP16) is comparable to desktop mid-range cards, making this GPU suitable for CUDA-accelerated compute tasks like video transcoding and machine learning inference within the 8 GB VRAM limit.

Nearest rivals include the Intel Arc B570 (20556, 0.1% faster), Intel Arc A750 (20582, 0.2% faster), NVIDIA Quadro M4000M (20480, 0.3% slower), and AMD Radeon R9 M390X (20662, 0.6% faster). The near-identical scores indicate that the RTX 3070 Mobile is positioned in a crowded performance band, where driver optimization and specific game engines will determine real-world differences rather than raw compute capability.

# Upgrade Path and Platform

The AMD Ryzen AI Max 390 uses AMD Socket FP11, which is a mobile-specific socket with no upgrade options; the CPU is soldered to the motherboard. The platform supports LPDDR5X memory in a quad-channel configuration with 256.0 GB/s bandwidth, and ECC memory is supported for error-correcting workloads. PCIe Gen 4 is available with 16 lanes from the CPU, which is fully utilized by the RTX 3070 Mobile's PCIe 4.0 x16 interface.

The GPU's TDP is 115 W, while the CPU has a 55 W TDP, giving a combined thermal budget of approximately 170 W for the primary compute components. The suggested PSU field is null, but the power connectors are listed as "None," indicating the GPU draws power through the motherboard or proprietary laptop power delivery rather than standard PCIe power connectors. This is typical for a laptop class device, so PSU headroom is not a user-serviceable consideration.

A sensible next upgrade for this platform would be increasing memory capacity, as the quad-channel LPDDR5X configuration benefits from having all channels populated to maintain the 256.0 GB/s bandwidth. Storage upgrades via PCIe Gen 4 NVMe drives are also viable, though the 16 lanes are shared between the GPU and other devices. The integrated Radeon 8050S provides a backup display output if the discrete GPU fails or is disabled for power savings.

The production status of the GPU is end-of-life, meaning driver support may taper off over time, while the CPU remains active. Future driver updates for the RTX 3070 Mobile may focus on stability rather than performance gains, so users should set expectations accordingly. The platform itself is current as of the CPU's 2025 release date, ensuring software compatibility for several years.

# Build Overview

This is a laptop-class system pairing the AMD Ryzen AI Max 390 (a 12-core, 24-thread Zen 5 mobile processor) with the NVIDIA GeForce RTX 3070 Mobile (an Ampere-based discrete GPU). The CPU is a high-end mobile part at the 91st percentile, while the GPU is a mid-range mobile part at the 65th percentile, resulting in a combined percentile of 78.

The CPU is manufactured on TSMC's 4 nm process, featuring a 5.00 GHz boost clock, 64 MB of L3 cache, and quad-channel LPDDR5X memory support with 256.0 GB/s bandwidth. The GPU is built on Samsung's 8 nm process with 5120 shading units, 8 GB of GDDR6 memory, and 40 RT cores. The pairing represents a workstation-class CPU with a mainstream gaming GPU, creating a system that excels in processor-intensive tasks while offering competent but not flagship-level graphics performance.

The overall tier, based on the 78th combined percentile, places this system in the upper-middle range of laptop configurations. It outperforms the majority of consumer laptops but trails dedicated high-end gaming or creator machines that would pair similarly strong CPUs with top-tier GPUs. The data shows no measured FPS for this combination, so all gaming expectations must be inferred from the individual component benchmark scores.

This configuration is best suited for users who prioritize CPU-heavy workloads—such as software development, video editing, 3D rendering, and scientific computing—while still wanting a capable discrete GPU for moderate gaming and GPU-accelerated tasks. It is not optimized for maximum gaming frame rates, but it offers a balanced feature set for a professional or power-user laptop.

# Balance and Bottleneck

The performance asymmetry between the CPU and GPU is pronounced: the Ryzen AI Max 390 sits at the 91st percentile among all CPUs, while the RTX 3070 Mobile rests at the 65th percentile among GPUs. This 26-percentage-point gap means the processor is the dominant component in this pairing, and in most gaming scenarios the GPU will be the limiting factor.

For CPU-bound workloads—such as code compilation, video encoding, 3D rendering, and data processing—the processor's 24 threads and 36064 Cinebench R23 multicore score ensure that the system delivers top-tier performance. The GPU's compute capabilities (15.97 TFLOPS FP32) are sufficient to accelerate some tasks, but the 8 GB VRAM and 65th percentile rank mean that GPU-bound operations will cap overall throughput in compute-heavy applications.

In gaming, the RTX 3070 Mobile's PassMark G3D score of 15309 and 3DMark Steel Nomad score of 2380 indicate that frame rates will be determined by the GPU at 1080p and 1440p resolutions. The CPU's single-thread score of 5091 (Cinebench R23) is more than adequate to feed the GPU in most titles, so the bottleneck will manifest as GPU utilization reaching 100% while CPU utilization stays lower. At lower resolutions or with reduced graphics settings, the CPU could become the limiting factor in esports titles, but this is unlikely given the high single-core performance.

The memory system is well-balanced: the CPU's 256.0 GB/s LPDDR5X bandwidth and the GPU's 448.0 GB/s GDDR6 bandwidth are both substantial, minimizing memory-related bottlenecks. The 64 MB L3 cache further reduces CPU memory stalls, ensuring that data feeds the GPU efficiently. However, the GPU's 8 GB VRAM could be a constraint in modern AAA games at high textures, potentially causing stuttering or reduced detail settings.

Overall, this system is CPU-overprovisioned and GPU-adequate. For users who run mixed workloads—gaming plus productivity—the balance is acceptable, but for gaming-first users, the GPU will consistently be the performance ceiling. The 78th combined percentile reflects this: the system is above average overall, but its gaming performance will be closer to the 65th percentile GPU rank than the 91st percentile CPU rank.

No measured FPS data exists for this exact pairing, so the above bottleneck analysis is derived from the benchmark score distributions. The deltaPct values among nearest rivals—all within 0.6%—suggest that small driver or thermal variations will have a larger impact on real-world frame rates than the raw component differences.