SYSTEM ANALYZER

Rate My PC: Intel Core Ultra 5 125U + Intel Arc A350M

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

89 / 100
HIGH-END

Power Build

Top 11% of systems. Excellent for 1440p Ultra or 4K High gaming.

1440p Ultra4K High

System Balance Analysis

CPU vs GPU performance ratio
Well Balanced
CPU
85%
VS
GPU
92%
PROCESSOR

Intel Core Ultra 5 125U

20,826 Benchmark Score
Top 15% Market Ranking
View Full Specs →
GRAPHICS CARD

Intel Arc A350M

24,647 Benchmark Score
Top 8% 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 Intel Core Ultra 5 125U paired with the Intel Arc A350M is a laptop-class combination that sits at an interesting intersection in the data: a 74th-percentile CPU, a 70th-percentile GPU, and a combined 72nd percentile. That consistency across both components is unusual, and it frames every question this analysis will ask. What does a Meteor Lake mobile processor with a low-power discrete Alchemist GPU actually deliver together, and where does the pairing run out of headroom? One important caveat up front: no measured FPS rows exist for this exact combination — the FACT PACK contains no measuredFps data — so all frame-rate discussion below is estimated from the benchmark scores rather than directly observed.

CPU Analysis

The Core Ultra 5 125U is a 12-core, 14-thread Meteor Lake processor built on Intel's 7 nm process. The thread count lagging the core count by only two tells you the design is heterogeneous: some cores support hyperthreading and some do not, a hybrid layout typical of Meteor Lake's performance-plus-efficiency tile structure. The base clock sits at 1300.00 MHz with a boost clock of 4.30 GHz — a wide spread that reflects a 15 W TDP part engineered to burst high and settle low rather than sustain aggressive frequencies under prolonged load.

The cache hierarchy reinforces the mobile-efficiency brief: 112 KB of L1 per core, 2 MB of L2 per core, and 12 MB of shared L3. That shared L3 pool is modest by desktop standards, and it will show up most in latency-sensitive workloads with large working sets. Memory support is DDR5 on a dual-channel bus delivering 89.6 GB/s of bandwidth, though the pack notes this depends on the motherboard — or in this case, the laptop implementation, since the chip uses the Intel BGA 2049 socket and is soldered rather than socketed. ECC is not supported, which rules the part out of certain workstation deployments.

The benchmark scores tell a coherent story. Cinebench R23 multi-core lands at 9799 while single-core reaches 1561 — a roughly six-to-one scaling ratio that is respectable for a hybrid-core design at this power envelope. Cinebench R20 shows the same shape: 5982 multi-core versus 844 single-core. Cinebench R15 posts 1640 and 227. Geekbench mirrors this with 6308 multi-core and 1493 single-core. The single-core figures are the more impressive half of the picture; a 1561 in R23 single-core means snappy interactive performance — application launches, browser responsiveness, light compilation — despite the 15 W envelope.

PassMark results add texture. The single-thread score of 3296 is solid for a mobile part, while the multithread score of 17417 confirms competent but unspectacular parallel throughput. The integer math result of 57046 and floating point score of 41331 suggest balanced general-purpose compute. Data compression at 198858 and encryption at 12231 indicate the chip handles throughput-oriented tasks adequately. Random string sorting at 21872 and extended instructions at 11340 round out a profile that is efficient-breadth rather than raw-depth.

Against its nearest rivals, the picture is strikingly flat. The average benchmark score of 20826 sits within a fraction of a percent of the Intel Xeon 6325P (20821, deltaPct 0), the AMD EPYC 7J13 (20845, deltaPct -0.1), the Intel Core i5-12490F (20802, deltaPct 0.1), and the Intel Core Ultra 7 268V (20897, deltaPct -0.3). That clustering is curious: a 15 W mobile chip matching server and desktop parts on aggregate score is a data quirk worth questioning. Aggregate averages blend single-core and throughput-heavy subtests, and the 125U's strong per-thread results likely compensate for its lower sustained multi-core ceiling. The practical reading: per-core responsiveness comparable to those rivals, but sustained all-core workloads — long renders, extended compiles — would likely favor the higher-power parts in that list. The multiplier is locked, so there is no overclocking escape hatch.

Benchmark Performance

The headline numbers again: 74th percentile versus all CPUs, 70th percentile versus all GPUs, and a combined 72nd percentile. What does a 74th-percentile CPU score mean in practice? It places the 125U above roughly three quarters of the CPUs in the database — but the database includes many older and low-power parts, so the percentile flatters a mobile chip. The nearest-rival data anchors it: effectively tied with a Xeon, an EPYC, a desktop Core i5, and a higher-tier Core Ultra 7 on aggregate.

The GPU side is thinner on data but consistent. Geekbench OpenCL scores 24546 and Vulkan scores 24747, for an average of 24647 — a 70th-percentile result. The Vulkan and OpenCL scores being nearly identical suggests the Arc A350M behaves uniformly across both compute APIs, with no driver-side oddity skewing one path. Against rivals, it is a dead heat with the AMD Radeon RX 590 (24744, deltaPct -0.4), effectively tied with the NVIDIA RTX A5000 Mobile (24763, deltaPct -0.5), marginally ahead of the AMD Radeon RX 6600 XT (24442, deltaPct 0.8), and 1.5% ahead of the NVIDIA GeForce GTX 1630 (24277).

The combined picture is balance. Neither component drags the pairing dramatically, and the combined percentile of 72 lands almost exactly between the two individual percentiles. For a laptop build, that balance is arguably the most useful property in the data — this is not a fast CPU throttled by a weak GPU, nor a strong GPU starved by a slow CPU. It is uniformly mid-pack, which sets expectations for everything that follows.

Usage Scenarios

High-refresh gaming: The GPU data argues against this use case. A 70th-percentile GPU with 3.379 TFLOPS of FP32 and 4 GB of VRAM cannot push high-refresh rates at modern resolutions; expectations should center on playable frame rates at reduced settings, and even that is workload-dependent.

Streaming: The CPU's single-core strength (R23 single at 1561, Geekbench single at 1493) can handle the encoder-side and chat-overlay load, but with 15 W of total thermal budget shared between a 12-core CPU and a 25 W discrete GPU, sustained streaming plus gaming simultaneously would press the power envelope hard. The data suggests casual streaming is feasible; dedicated streaming rigs are not.

Video editing: Timeline scrubbing and preview playback depend heavily on single-thread responsiveness, and the 125U's per-core numbers support a smooth interactive editing experience. Final exports are a different question — a 9799 R23 multi-core score is capable but not fast for 4K timeline exports. Short-form and 1080p editing fits comfortably.

3D rendering: This is the weakest scenario in the data. Sustained multi-core throughput of this class, matched against server parts like the EPYC 7J13 only on blended averages, will not translate into quick production renders. The 24546 OpenCL score offers a GPU-assisted path in renderers that support it, but 4 GB of VRAM limits scene size.

Software development: The strongest CPU scenario. Compilation is a mix of single-core and bursty parallel work, and the 125U's single-thread results (PassMark 3296, Geekbench 1493) plus adequate multithreading (17417 PassMark) make it a competent developer machine. Containers and local test suites are within reach at moderate scale.

Student and office work: Ideal. Documents, browsing, video calls, and light photo work are dominated by single-core responsiveness and quick wake behavior, exactly where the 125U's benchmark profile is strongest. The 74th CPU percentile guarantees a fluid everyday experience.

Who Should Build It

Since this is a laptop-class pairing (buildClass: laptop), "build" means "buy the machine that implements it." The data points to students and general-purpose professionals first: the CPU's single-core scores deliver the responsiveness those users feel every day, and the combined 72nd percentile is more than sufficient for mainstream productivity suites.

Content creators working at 1080p scale benefit next — video editors cutting short-form content and photographers processing RAW files will find the CPU scores adequate and the GPU's 24647 average benchmark score a modest accelerator in supported applications. Game developers and 3D artists should look elsewhere; the rendering throughput and VRAM capacity in this data do not support that workload class.

Software developers doing general application work fit well, as argued above. Small business workstations for office and light data work also fit — the PassMark encryption score of 12231 and compression result of 198858 indicate competent handling of file-heavy tasks, though the absence of ECC support rules out deployments where memory integrity is mandatory. Gamers, finally, are only a partial fit: this pairing suits casual and esports-adjacent play at moderate settings, not enthusiast gaming at high resolutions.

Gaming Performance

To restate the caveat once, clearly: no measured FPS data exists for this exact CPU-plus-GPU combination — the FACT PACK contains no measuredFps rows, and dataIsMeasured is false. Everything in this section is estimated from the benchmark scores, not directly measured.

What do the scores imply? The Arc A350M's compute profile — 3.379 TFLOPS FP32, a 52.80 GPixel/s pixel rate, and a 105.6 GTexel/s texture rate — describes a GPU tuned for 1080p-class rendering at moderate settings. The 4 GB GDDR6 buffer is the binding constraint: modern titles at ultra settings routinely exceed that capacity, so texture pop-in and frame-time stutter become likely before raw compute does. Estimates for esports titles at 1080p would land in smooth territory; estimates for modern AAA titles at 1080p ultra would likely fall below comfortable thresholds, with medium settings the realistic target. The rival positioning — statistically tied with the Radeon RX 590 and slightly ahead of the GTX 1630 on aggregate scores — supports the same conclusion: entry-level 1080p gaming, settings dialed down for demanding titles. Ray tracing deserves a note: six RT cores exist, but the compute budget behind them makes RT practically off the table for smooth play.

GPU Analysis

The Arc A350M is an Alchemist-generation Xe-HPG part, the DG2-128 chip, manufactured by TSMC on a 6 nm process. The physical numbers are compact: 7200 million transistors on a 157 mm² die, a density of 45.9M per mm². The rendering resources consist of 768 shading units, 48 TMUs, 24 ROPs, and 6 RT cores. There is no tensor core count listed in the data.

Clocks run from a 1150 MHz base to a 2200 MHz boost, with memory at 1750 MHz — 14 Gbps effective — across a 64-bit bus carrying 4 GB of GDDR6. That 64-bit interface yields 112.0 GB/s of bandwidth, a narrow pipe that constrains how far the boost clock can be exploited in bandwidth-hungry scenes. The FP32 rate of 3.379 TFLOPS and FP16 rate of 6.758 TFLOPS (2:1) define the compute ceiling.

For rendering workloads, the benchmark interpretation is straightforward. The OpenCL score of 24546 and Vulkan score of 24747 place the GPU in the 70th percentile — competent mid-tier compute for a mobile part. What that means for real rendering: GPU-accelerated viewport work and effects processing are viable at moderate scene complexity, but the 4 GB VRAM ceiling caps asset sizes before compute does. The bus interface is PCIe 4.0 x8, adequate for a discrete mobile GPU at this performance class. API support is modern and broad — DirectX 12 Ultimate (12_2), Vulkan 1.4, and OpenGL 4.6 — which matters because Alchemist's architecture is designed around these newer APIs, and the near-identical Vulkan and OpenCL scores suggest it delivers on them evenly. Note the production status: the A350M is listed as end-of-life, so this pairing represents a fixed, non-evolving platform.

Upgrade Path and Platform

The upgrade story here is short by nature: this is a BGA 2049 soldered mobile processor, so there is no CPU socket to swap. The platform characteristics define the boundaries instead. Memory support is DDR5 on a dual-channel bus at 89.6 GB/s of bandwidth, dependent on the specific laptop implementation — buyers should verify whether a given machine ships with dual-channel populated configurations to capture that full figure. PCIe is Gen 4 with 12 lanes from the CPU, which covers the A350M's PCIe 4.0 x8 link plus a fast NVMe SSD.

Power headroom is defined by the two TDP figures: 15 W for the CPU and 25 W for the GPU. No suggested PSU value exists in the data — expected, since this is a battery-and-adapter laptop domain rather than an ATX one. The sensible "next upgrade" is therefore not a component swap but a whole-platform step: the data shows the natural successor within Intel's own stack would be something like the Core Ultra 7 268V, which appears among the CPU's nearest rivals with an average score of 20897 — a marginal 0.3% aggregate gain, which is itself telling. Moving meaningfully up the stack requires leaving this efficiency class entirely. The locked multiplier closes off any tuning path.

FAQ

Q: Is the Intel Core Ultra 5 125U good for gaming?

A: As a CPU, yes — its 74th percentile ranking and 1561 Cinebench R23 single-core score are more than sufficient. Gaming performance in this pairing is limited by the GPU, not the processor.

Q: How much VRAM does the Arc A350M have, and is it enough?

A: 4 GB of GDDR6 on a 64-bit bus with 112.0 GB/s of bandwidth. For 1080p gaming at moderate settings, workable; for ultra-settings modern titles or large 3D scenes, it is the primary constraint.

Q: How does this CPU compare to its rivals?

A: Its average benchmark score of 20826 is statistically tied with the Intel Xeon 6325P (20821), AMD EPYC 7J13 (20845), Intel Core i5-12490F (20802), and Core Ultra 7 268V (20897) — all within 0.3% on aggregate.

Q: Can this setup handle video editing?

A: Interactive editing, yes — single-core scores of 1493 (Geekbench) and 3296 (PassMark) drive smooth timeline work. Final renders will be moderate rather than fast, given the 9799 Cinebench R23 multi-core result.

Q: Is the FPS data in this analysis measured?

A: No. No measured FPS rows exist for this combination; all frame-rate discussion is estimated from benchmark scores.

Q: What is the launch MSRP of the CPU?

A: The launch MSRP is $363.

Q: Can the CPU be overclocked?

A: No — the multiplier is unlocked flag is false, so no multiplier-based tuning is available.

Balance and Bottleneck

The combined percentile of 72, sitting neatly between the CPU's 74 and the GPU's 70, describes a pairing with no gross imbalance on paper. But percentiles hide workload-specific dynamics, and the data lets us dig one level deeper.

For gaming, the GPU is the limiter. The A350M's 3.379 TFLOPS FP32, 112.0 GB/s of bandwidth, and 4 GB VRAM place a hard ceiling well below what the 125U's single-thread performance could otherwise feed. In CPU-bound scenarios — low-resolution esports settings, strategy games with heavy simulation — the 1561 R23 single-core score keeps frame delivery smooth; as resolution and settings climb, the GPU percentile of 70 becomes the binding constraint, and the narrow 64-bit memory bus plus limited VRAM will bite first.

For content creation, the picture inverts. GPU compute at the 24647 average score can accelerate supported filters and effects, but the CPU's blended-average parity with server-class rivals does not translate into sustained render throughput at a 15 W TDP. Long all-core tasks will thermally settle below peak boost — the 1300.00 MHz base clock is the realistic sustained anchor — making the CPU the limiter in extended renders while remaining the stronger component in interactive work.

Because no measured FPS rows exist, bottleneck evidence rests entirely on the score spread: the near-identical percentiles (74 versus 70) suggest a balanced pairing for mixed use, while the underlying spec asymmetry — strong single-core CPU, bandwidth-constrained GPU — predicts GPU-limited gaming and CPU-limited rendering. That is the signature of an efficiency-first laptop design, and the data is consistent with it.

Build Overview

This is a laptop-class pairing — buildClass confirms it — combining Intel's Meteor Lake efficiency architecture with Intel's entry-tier Alchemist discrete graphics. The Core Ultra 5 125U brings 12 cores, 14 threads, a 4.30 GHz boost, and a 74th-percentile aggregate that lands within a rounding error of desktop and server rivals on blended benchmarks, powered by just 15 W. The Arc A350M adds 768 shading units, 4 GB of GDDR6, and a 70th-percentile compute profile at 25 W. Both chips come from the same manufacturer, sharing a PCIe 4.0 connection.

Overall tier: upper-mid-pack, uniformly. A combined percentile of 72 with both components clustered around it describes a machine that is simultaneously adequate at everything and exceptional at nothing — a generalist. The data's clearest strengths are single-core responsiveness (Geekbench 1493, PassMark 3296, Cinebench R23 1561) and balanced API support on the GPU side; the clearest limits are sustained multi-core throughput, VRAM capacity, and a memory bus that constrains graphics bandwidth. For students, office professionals, developers, and 1080p-scale content work, the measured numbers support the pairing without reservation. For enthusiast gaming and heavy 3D rendering, the same numbers argue for looking further up the stack. The pairing is, in the most literal reading of the data, a 72nd-percentile machine — and every recommendation above follows from that single, consistent fact.