SYSTEM ANALYZER

Rate My PC: AMD Ryzen AI Max PRO 390 + Intel Arc B390

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

86 / 100
HIGH-END

Power Build

Top 14% of systems. Excellent for 1440p Ultra or 4K High gaming.

1440p Ultra4K High

System Balance Analysis

CPU vs GPU performance ratio
Well Balanced
CPU
96%
VS
GPU
77%
PROCESSOR

AMD Ryzen AI Max PRO 390

63,765 Benchmark Score
Top 4% 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 390 paired with the Intel Arc B390 is, on paper, a deeply lopsided combination: a 93rd-percentile mobile workstation processor bolted to a GPU that sits in just the 9th percentile of all graphics hardware. The data tells that story immediately. The CPU's average benchmark score of 63765 places it within a fraction of a percent of parts like the Intel Core i9-13900KS and AMD EPYC 7343, while the Arc B390's lone recorded 3DMark Steel Nomad score of 1482 puts it in the same bracket as entry-level integrated-class graphics from a much earlier era. This is a laptop-class pairing where the processor is overwhelmingly the stronger half, and every workload that leans on GPU throughput will feel that imbalance.

One important caveat up front: no measured FPS rows exist for this exact combination in the database — `dataIsMeasured` is false and the measured-FPS table is empty. All frame-rate discussion below is therefore estimated from the benchmark scores, not from captured gameplay data.

CPU Analysis

The Ryzen AI Max PRO 390 is a 12-core, 24-thread Zen 5 processor built on TSMC's 4 nm process, codenamed Strix Halo, and it runs on Socket FP11 as a mobile part. It has a base clock of 3.20 GHz and boosts to 5.00 GHz. Cache is generous for a mobile design: 80 KB of L1 per core, 1 MB of L2 per core, and a large 64 MB of shared L3. That L3 pool matters for both sustained throughput work and for feeding the integrated Radeon 8050S graphics — though in this pairing the Arc B390 is the GPU in question, and it draws from system memory rather than its own pool.

The benchmark scores define the character of this chip. In Cinebench R23 multi-core it posts 24828, with a single-core score of 1976; in the older R15 test those figures are 3918 multi-core and 303 single-core. The R23 multi-to-single ratio of roughly 12.6x against a 12-core/24-thread design indicates the chip scales well across its full thread count rather than throttling into a narrow thermal envelope — a common weakness in mobile silicon. The 55 W TDP is modest for this level of performance and explains why the part competes with desktop-flagship-class rivals in aggregate score.

PassMark results reinforce the picture. The multithread score of 42912 and single-thread score of 3967 sit alongside strong specialized results: 148508 in integer math, 94666 in floating point math, 506170 in data compression, and 26027 in encryption. The extended instructions score of 40888 shows the Zen 5 SIMD units are being exercised effectively. Random string sorting at 55248 and find-prime-numbers at 323 round out a profile that is uniformly capable rather than spiky — there is no workload category in the pack where the CPU scores disproportionately poorly.

In real terms, that means sustained multi-threaded workloads — code compilation, rendering, simulation, heavy data transformation — will run at a level comparable to a desktop i9-13900KS or an EPYC server chip, at a fraction of the power draw. Single-thread responsiveness is equally respectable at 1976 in R23, so lightly-threaded applications and general desktop interactivity will not feel like a compromise. The multiplier is locked, so clock tuning is off the table; the performance in the database is effectively the performance a buyer gets.

Benchmark Performance

The CPU sits at the 93rd percentile versus all CPUs in the database — elite territory for a mobile part. Its average benchmark score of 63765 lands it in a remarkably tight cluster of rivals: the Intel Core i9-13900KS at 64051 is just 0.4% ahead, the AMD EPYC 7343 at 64202 is 0.7% ahead, the AMD Ryzen AI 7 450G at 63331 is 0.7% behind, and the Intel Core Ultra 7 265HX at 63173 is 0.9% behind. Being statistically indistinguishable from a desktop i9 flagship and a server EPYC part, while drawing 55 W, is the single most striking fact in this dataset.

The GPU story is the opposite. The Arc B390's average benchmark score is 1482, derived from a single recorded test — 3DMark Steel Nomad at 1482 — and that places it in the 9th percentile of all GPUs. Its nearest rivals are the NVIDIA GeForce GT 520MX at 1463 (1.3% behind the Arc), the GeForce 800M at 1460 (1.5% behind), the GT 625 OEM at 1446 (2.5% behind), and the GT 710 at 1443 (2.7% behind). In other words, the B390's graphics throughput is comparable to basic display-output-class hardware, not gaming hardware.

The combined percentile of 51 reflects this tug-of-war: the CPU drags the pairing upward, the GPU drags it down, and the net result is an exactly middling aggregate. That 51st-percentile figure should not be read as "balanced" — it is the average of two extremes, and the practical experience of using this machine will swing dramatically depending on which component a given task touches.

FAQ

Q: How does the Ryzen AI Max PRO 390 compare to desktop CPUs?

A: Its average benchmark score of 63765 is within 0.4% of the Intel Core i9-13900KS (64051) and 0.7% of the AMD EPYC 7343 (64202), while operating at a 55 W TDP from the mobile Socket FP11.

Q: Is the Intel Arc B390 good for gaming?

A: No. It sits in the 9th percentile of all GPUs, with a 3DMark Steel Nomad score of 1482 — comparable to the NVIDIA GeForce GT 710 (1443) and GT 520MX (1463). Modern AAA gaming is outside its envelope.

Q: How many cores and threads does the CPU have, and what architecture is it?

A: 12 cores and 24 threads on AMD's Zen 5 architecture, codename Strix Halo, manufactured on TSMC's 4 nm node, with a 3.20 GHz base and 5.00 GHz boost clock.

Q: What memory does this platform support?

A: LPDDR5X in a quad-channel configuration with 256.0 GB/s of bandwidth, plus ECC support — a server-grade feature set for a mobile chip.

Q: Does the database contain measured game FPS for this pairing?

A: No. There are no measured FPS rows for this CPU+GPU combination; all frame-rate expectations discussed here are estimates derived from the benchmark scores.

Q: Can the CPU be overclocked?

A: No — the multiplier is locked, so the recorded performance figures represent the ceiling.

Q: What is the overall tier of this build?

A: It is a laptop-class pairing with a combined percentile of 51: a 93rd-percentile CPU matched to a 9th-percentile GPU.

Who Should Build It

The target user for this combination is unambiguous: someone whose workload is compute-bound, not graphics-bound. Developers are the clearest fit. A 24828 Cinebench R23 multi-core score and a 42912 PassMark multithread result mean compile times, containerized build pipelines, and local test suites complete at desktop-flagship speed. The strong encryption (26027) and compression (506170) PassMark scores also make this a credible machine for anyone handling encrypted volumes, large archives, or data pipelines on the move.

Content creators working in CPU-bound disciplines — audio production, photo editing, code development, spreadsheet-heavy analysis — get the full benefit of the 93rd-percentile processor. The 64 MB of shared L3 and quad-channel 256.0 GB/s memory bandwidth help with large working sets that would choke narrower memory subsystems. ECC support adds a data-integrity angle that small-business workstations and compute-cluster nodes sometimes require.

What this pairing is not for: gamers. A GPU in the 9th percentile with rivals like the GT 710 cannot deliver meaningful modern gaming performance at any resolution, let alone at high settings. Students in non-graphics fields, engineering teams running simulation code, and businesses deploying portable compute nodes are the realistic audience. Note also that the CPU's own integrated Radeon 8050S exists in the silicon; the presence of a discrete-class Arc B390 in the same system adds little on the benchmark evidence available, since the B390 scores like display-output hardware rather than a gaming accelerator.

Balance and Bottleneck

The bottleneck analysis is the shortest section of this page because the answer is singular: the GPU. A 93rd-percentile CPU feeding a 9th-percentile GPU is an 84-percentile gap between the two halves of this machine. Every graphics workload — 3D rendering, GPU compute, gaming — is limited entirely by the Arc B390, whose Steel Nomad score of 1482 is the sole relevant number. No amount of additional CPU headroom changes that; the processor could be halved in performance and the GPU would still be the limiter in any rendering task.

Conversely, every CPU-bound workload ignores the GPU almost entirely, and there the machine performs like a desktop workstation: within a fraction of a percent of an i9-13900KS. The combined 51st percentile is the arithmetic meeting point of these two realities and does not describe any single user experience well.

Since no measured FPS rows exist for this combination, FPS scaling across resolutions cannot be quoted numerically. Qualitatively from the benchmark scores: estimated frame rates in modern AAA titles would be low at 1080p on ultra settings and would remain low at reduced settings, because the B390's peer group — the GT 520MX, GeForce 800M, GT 625 OEM, GT 710 — consists of hardware that was never positioned for that class of workload. The CPU would register near-total utilization headroom in any game; the GPU would saturate immediately. Esports-class and older titles at modest settings are the realistic estimated ceiling, and even there expectations should be tempered.

Upgrade Path and Platform

The platform facts are unusual and worth stating plainly. The CPU uses AMD Socket FP11 — a mobile, soldered socket — and the GPU is an IGP-class part with no power connectors, bus interface listed as IGP, and system-shared memory. In a laptop build (`buildClass` is laptop), neither component is user-upgradeable in the conventional sense. The upgrade path is therefore the whole machine, not a part swap.

What the platform does offer on paper: PCIe Gen 4 with 16 lanes from the CPU, LPDDR5X quad-channel memory at 256.0 GB/s with ECC, and a total platform power envelope implied by a 55 W CPU TDP and an 80 W GPU figure. No suggested PSU value is provided, which is consistent with an integrated mobile design where power delivery is fixed by the chassis. Buyers should treat memory and storage configuration at purchase time as the only meaningful "upgrade decisions" available, since LPDDR5X is not socketed and the memory size is shared between CPU workloads and the B390's system-dependent graphics memory.

A sensible next upgrade, given the data, is a machine that keeps the processor class but raises the GPU class dramatically — the CPU side of this pairing is already elite, and the 9th-percentile graphics position is the only metric with substantial room to improve. Any successor system should be judged first on its GPU percentile.

Gaming Performance

To restate clearly: the database contains no measured FPS figures for this exact CPU+GPU combination, so the figures discussed here are estimates inferred from the benchmark scores, not captured results.

The estimate is not encouraging. The Arc B390's only recorded score, 1482 in 3DMark Steel Nomad, places it 9th-percentile overall and within a few percent of the GT 520MX (1463, 1.3% gap), GeForce 800M (1460, 1.5% gap), GT 625 OEM (1446, 2.5% gap), and GT 710 (1443, 2.7% gap). Hardware in that bracket is entry-level display-class silicon. Estimated modern AAA performance at 1080p ultra would fall well short of playable frame rates, and even aggressive setting reductions would likely only narrow the gap rather than close it, since the deficit is architectural rather than a matter of tuning.

The CPU contributes nothing to change this outcome — a 1976 R23 single-core score and 2773 PassMark physics result mean the processor delivers more frame-pacing and simulation capacity than the GPU can consume. Estimated best-case scenarios are lighter or older titles at reduced resolutions and settings. For gaming as a primary use case, the benchmark data supports a clear verdict: look elsewhere.

Build Overview

This is a laptop-class pairing: a mobile AMD Ryzen AI Max PRO 390 (Zen 5, Strix Halo, Socket FP11, 4 nm TSMC) combined with an Intel Arc B380-family GPU — specifically the B390, a Panther Lake, Xe3-LPG architecture part built on Intel's 3 nm process, released later than the CPU by roughly a year in the database records. The two components come from different vendors and different design philosophies, and the combined percentile of 51 captures the resulting tension.

Tier-wise, the pairing is best described as a portable compute platform with entry-level graphics. The CPU half is 93rd-percentile, workstation-grade, ECC-capable. The GPU half is 9th-percentile. Anyone evaluating this build should weight those two numbers according to their workload: compute-heavy users get near-top-tier hardware; graphics users get bottom-decile hardware. The machine's identity is the former.

GPU Analysis

The Intel Arc B390 is an integrated-class graphics part: 1536 shading units, 48 texture mapping units, 24 render output units, and 12 RT cores, on the Xe3-LPG architecture with the Panther Lake chip manufactured on Intel's 3 nm node. Its clock range spans a 300 MHz base to a 2500 MHz boost, with memory shared from the system pool — no dedicated VRAM, no fixed bus width, and bandwidth listed as system dependent.

Theoretical rates follow from those specs: 60.00 GPixel/s pixel fill, 120.0 GTexel/s texture fill, 7.680 TFLOPS FP32, and 15.36 TFLOPS FP16 at a 2:1 ratio. Modern API support is present — DirectX 12 Ultimate (12_2), Vulkan 1.4, OpenGL 4.6 — so software compatibility is not the limiting factor. The 80 W figure attached to the part is high for an IGP-class design and suggests the B390 is the graphics engine of a mobile platform rather than a standalone card; it has no power connectors and its slot width is listed as IGP.

For rendering workloads, the single benchmark datum is decisive: 1482 in 3DMark Steel Nomad, 9th percentile, with the GT 520MX and GT 710 as nearest peers. That places real-time 3D rendering, GPU-accelerated viewport work, and ray-traced rendering (despite the 12 RT cores) far outside this GPU's practical envelope. The system-shared memory design ties graphics bandwidth to the platform's 256.0 GB/s LPDDR5X pool, which is generous by iGPU standards and is the one structural advantage the B390 enjoys — but the benchmark score indicates the shader array and overall throughput, not memory, are the binding constraints. As a rendering engine this GPU is a baseline; as a display output device for a very fast mobile CPU, it is exactly what the data says it is.