SYSTEM ANALYZER

Rate My PC: Intel Core Ultra 5 235A + Intel Arc A380

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

90 / 100
ULTIMATE READY

Apex Performer

Top 10% 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
94%
VS
GPU
86%
PROCESSOR

Intel Core Ultra 5 235A

48,201 Benchmark Score
Top 6% Market Ranking
View Full Specs →
GRAPHICS CARD

Intel Arc A380

8,558 Benchmark Score
Top 14% 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

The Intel Core Ultra 5 235A paired with the Intel Arc A380 is a desktop combination defined by asymmetry: a 90th-percentile CPU from Intel's Arrow Lake generation attached to a 44th-percentile discrete GPU from the Alchemist family. The combined percentile of 67 reflects that split. This page examines what that imbalance means in practice, using only the benchmark data available — and it should be stated up front that no measured FPS rows exist for this exact pairing, so all frame-rate discussion below is estimated from the benchmark scores rather than directly observed.

GPU Analysis

The Arc A380 is a compact Alchemist part built on TSMC's 6 nm process, with 7,200 million transistors packed into a 157 mm² die at a density of 45.9 million transistors per square millimetre. Its core configuration is entry-level by modern standards: 1024 shading units, 64 texture mapping units, 32 render output units, and 8 ray-tracing cores. There are no tensor cores listed in the specification, which immediately constrains its usefulness for accelerated AI inference workloads.

The memory subsystem is the card's most obvious structural limit. Six gigabytes of GDDR6 sits on a 96-bit bus, delivering 186.0 GB/s of bandwidth. That bus width and capacity ceiling matter more than raw shading throughput in modern rendering, because once a scene's working set exceeds the frame buffer, performance falls off a cliff rather than degrading gracefully. For viewport work and older engines the capacity is workable; for texture-heavy modern titles at higher resolutions, it is the first wall the card hits.

Clocks are unusual in how narrow the range is: a 2000 MHz base and a 2050 MHz boost, with memory running at 1937 MHz (15.5 Gbps effective). The card barely boosts above its base, which suggests its scores reflect near-constant operation rather than burst behaviour. Throughput figures follow directly: 65.60 GPixel/s of pixel fill, 131.2 GTexel/s of texture fill, 4.198 TFLOPS FP32, and 8.397 TFLOPS FP16 at a 2:1 ratio. These are the numbers of a card designed for display output and light compute, not competitive rendering.

The benchmark picture confirms that. A 3DMark Steel Nomad DX12 score of 808 is very low in absolute terms — this test is designed to stress modern GPUs, and a sub-1000 result places the A380 firmly in the entry tier. Passmark G3D lands at 6252, with a GPU compute score of 2762. The API-specific Passmark results are striking: DirectX 12 at 35, DirectX 11 at 38, DirectX 10 at 37, and DirectX 9 at 73. The older-API score being roughly double the newer ones is a pattern worth questioning — it may reflect driver optimisation priorities on Alchemist, where legacy paths were in some cases stronger than advanced ones. Either way, the implication for rendering is that the card copes better with older graphics APIs than with modern ones.

Geekbench results tell a similar story: 38,224 in OpenCL and 36,736 in Vulkan. These are consistent with each other, suggesting the card behaves predictably across compute interfaces rather than being wildly optimised for one. Against its nearest rivals — the AMD FirePro W5170M (average score 8595, only 0.4 percent ahead), the Radeon HD 8870M (8462, the A380 is 1.1 percent ahead), the GeForce MX330 (8458, 1.2 percent behind), and the Radeon 880M (8436, 1.4 percent behind) — the A380 sits in a cluster of mobile and integrated-class parts. Its average benchmark score of 8558 and 44th percentile position against all GPUs confirm it as a lower-mid-tier card, effectively at the level of integrated graphics with a modest dedicated frame buffer.

The card is end-of-life, with Battlemage listed as its successor. It draws 75 W, occupies a dual-slot footprint, requires a single 8-pin connector, and connects over PCIe 4.0 x8. For rendering, the practical verdict from the data: expect capable display output and light compute, but the Steel Nomad score of 808 signals that heavy modern real-time rendering is outside its comfort zone.

Benchmark Performance

The CPU side of this build posts strong numbers. In Cinebench R23, the Core Ultra 5 235A scores 32,633 multi-core and 4,607 single-core. In R20 the equivalent figures are 13,705 and 1,934; in R15, 3,289 and 464. The single-core results in particular are high-end territory — a 4,607 R23 single-core score indicates snappy, responsive per-thread behaviour, while the multi-core scores reflect the 14-core, 14-thread configuration working without hyper-threading-style duplication.

Passmark adds breadth. Single-thread scores of 4,557 appear twice (single_thread and singlethread), consistent with the Cinebench single-core strength. Multithread lands at 38,392, physics at 2,437, integer math at 88,626, floating point at 118,778, data compression at 393,800, encryption at 30,136, extended instructions at 31,625, prime number finding at 392, and string sorting at 49,489. The compression and encryption numbers are particularly relevant for development and file-handling workloads, and both are solid for this class.

The CPU's average benchmark score of 48,201 places it in the 90th percentile of all CPUs, and its nearest rivals are remarkably close: the AMD Ryzen AI Max PRO 380 averages 48,171 (0.1 percent behind), the Intel Core Ultra 5 245HX averages 48,287 (0.2 percent ahead), the AMD EPYC 4345P averages 48,470 (0.6 percent ahead), and the AMD Ryzen 9 3900 averages 47,918 (0.6 percent behind). A spread of barely more than half a percent either way means this chip performs in a statistical dead heat with parts spanning mobile AI processors to workstation-class EPYC silicon. That is an interesting finding: the 235A punches at a level usually associated with higher-positioned products.

The combined picture, then, is a 67th-percentile desktop. The GPU's average score of 8,558 versus the CPU's 48,201 quantifies the imbalance. In any workload that leans on both — gaming above all — the GPU is the ceiling. In CPU-bound workloads, the system performs like a 90th-percentile machine.

FAQ

Q: Does this build have measured frame-rate data for specific games?

A: No. The database contains no measured FPS entries for this exact CPU-GPU combination, so all frame-rate discussion on this page is estimated from benchmark scores rather than observed results.

Q: How strong is the CPU compared to the wider market?

A: It sits in the 90th percentile of all CPUs, with an average benchmark score of 48,201 — statistically tied with the AMD Ryzen AI Max PRO 380, Intel Core Ultra 5 245HX, AMD EPYC 4345P, and AMD Ryzen 9 3900, all within about half a percent.

Q: Is the Arc A380 good for modern gaming?

A: The data suggests limited capability. A 3DMark Steel Nomad DX12 score of 808 and a Passmark G3D score of 6252 place it in the 44th percentile of all GPUs, in the same cluster as integrated-class rivals like the Radeon 880M and GeForce MX330.

Q: How much memory does the GPU have, and why does it matter?

A: Six gigabytes of GDDR6 on a 96-bit bus with 186.0 GB/s of bandwidth. The capacity is the practical constraint: modern texture-heavy scenes at higher resolutions can exceed it, causing sharp performance drops rather than gradual slowdowns.

Q: What power supply does this build need?

A: The GPU's suggested PSU is 250 W, with the card itself drawing 75 W via a single 8-pin connector. The CPU adds a 65 W TDP. A modest PSU covers this configuration with headroom.

Q: Can the CPU be overclocked?

A: No — the multiplier is locked on this model.

Q: What is the CPU's launch MSRP?

A: The launch MSRP was $269.

Gaming Performance

As noted, no measured FPS data exists for this pairing in the database — every figure in this section is an estimate derived from the benchmark scores, not a measured result.

What do the scores imply? The Passmark G3D score of 6252 and the Steel Nomad result of 808 anchor expectations. Parts in this GPU's rival cluster — the Radeon 880M, the MX330, the HD 8870M — are components associated with entry-level and thin-and-light gaming, not high-refresh or high-resolution play. The estimate, therefore, is playable frame rates at lower resolutions and reduced settings in most titles, with older or lighter engines running comfortably and demanding modern games requiring substantial setting reductions.

The API breakdown adds nuance. With Passmark DirectX 9 at 73 versus DirectX 12 at 35, the card's relative strength in older APIs suggests that legacy and well-optimised titles will feel disproportionately smoother than the newest releases. Ray tracing, with only 8 RT cores and a 65.60 GPixel/s fill rate, should be considered off the table for real-time use.

The 6 GB frame buffer sets the resolution ceiling. At higher resolutions, texture data will exceed capacity before the shaders themselves run out of work — meaning the estimate is that performance will not scale gracefully as resolution rises. For this pairing, the practical gaming estimate is lower-resolution play at moderate settings, where the 14-core CPU will rarely be the limiting factor.

Balance and Bottleneck

This is an unambiguously GPU-limited build for graphics workloads. The evidence is in the percentiles: a 90th-percentile CPU against a 44th-percentile GPU, producing a combined percentile of 67. In any frame-rate-bound scenario, the GPU is the constraint by a wide margin — the CPU's R23 multi-core score of 32,633 and Passmark multithread score of 38,392 represent far more simulation and draw-call throughput than the A380 can convert into pixels.

Flip the workload, and the picture inverts completely. For compiling code, running multi-threaded batch processing, or any compute task that fits in system memory, the system behaves like a 90th-percentile machine with 14 threads, 4,557 Passmark single-thread performance, and DDR5 bandwidth of 102.4 GB/s through a dual-channel bus. The CPU will not be the bottleneck in these scenarios for years of typical software growth.

The 6 GB GPU frame buffer introduces a secondary, capacity-based bottleneck that is independent of raw GPU speed — even in workloads the shaders could handle, memory exhaustion will intervene first. The one genuinely balanced workload class for this pairing is display output and light compute: the card's dual-slot format, 222 mm length, and triple DisplayPort 2.0 plus HDMI output make it a multi-monitor driver, a role where its 44th-percentile standing is irrelevant.

Who Should Build It

The data points to a specific user profile. Students and office users get the most coherent package: the 90th-percentile CPU delivers responsive application performance (4,607 R23 single-core) and the GPU provides display connectivity — one HDMI 2.1 and three DisplayPort 2.0 outputs — with enough capability for casual or older games. Software developers benefit directly: 14 threads, a Passmark multithread score of 38,392, and 393,800 in data compression make compilation and file operations quick, and the locked multiplier is irrelevant to this audience.

Content creators are a split case. Video editors working with moderate projects will find the CPU's 32,633 R23 multi-core score capable, though the GPU's 6 GB buffer and 2,762 compute score rule out heavy GPU-accelerated effects. 3D artists doing viewport modelling can use the card, but final rendering on it — a Steel Nomad score of 808 — is not realistic; CPU rendering at the 90th percentile is the viable path. Small business workstations are arguably the ideal fit, where multi-monitor output, ECC absence being acceptable, and strong general-purpose compute matter more than graphics horsepower.

Gamers should look carefully. At lower resolutions in modest titles, the pairing works; gamers targeting high refresh rates or higher resolutions in modern AAA titles should treat the GPU as the component to replace.

CPU Analysis

The Core Ultra 5 235A is a Core Ultra Series 2 desktop processor built on the Arrow Lake architecture (codename Arrow Lake-S), manufactured by TSMC on a 3 nm process. The die measures 243 mm² and carries 17,800 million transistors. It offers 14 cores and 14 threads — a one-thread-per-core design — with a 3.40 GHz base clock boosting to 5.00 GHz. The TDP is 65 W, the multiplier is locked, and the integrated graphics are Arc Xe-LPG with 24 execution units, relevant here as a fallback display solution.

Cache is generous and well-distributed: 192 KB of L1 per core, 3 MB of L2 per core, and 24 MB of shared L3. That per-core L2 allocation benefits workloads with localised data access — compilation, simulation, and game logic all fit this pattern. Memory support is DDR5 over a dual-channel bus at 102.4 GB/s; ECC is not supported.

What do the benchmark scores mean for real workloads? The single-core results — 4,607 in R23, 1,934 in R20, 464 in R15, and 4,557 in Passmark single-thread — describe a processor that feels fast in everything user-facing: application launches, browser responsiveness, editor interactions. The multi-core results (32,633 in R23, 38,392 Passmark multithread) describe genuine throughput for sustained parallel work. The floating point score of 118,778 and integer score of 88,626 cover scientific and general computation respectively, while the physics score of 2,437 and encryption score of 30,136 round out a balanced profile. Notably, the 5.00 GHz boost on 14 threads without SMT-style doubling means the multi-core scores are earned by physical cores alone — an architecture that relies on clock speed and cache rather than thread inflation. Against rivals like the Core Ultra 5 245HX and Ryzen 9 3900, it holds its ground within a fraction of a percent.

Build Overview

This is a desktop-class build combining Intel's Core Ultra 5 235A (Arrow Lake, Socket 1851, released with a launch MSRP of $269) with the Intel Arc A380 (Alchemist, DG2-128, end-of-life, launch MSRP of 149 USD). Both components come from Intel, making this an all-Intel platform — CPU, iGPU, and dGPU.

The overall tier, judged from percentiles, is upper-mid: a 67th-percentile combined system built from a 90th-percentile CPU and a 44th-percentile GPU. That composition defines the build's identity — it is a compute-forward machine with display-grade graphics, rather than a balanced gaming rig. The CPU is in Active production status while the GPU is end-of-life, which is worth noting for anyone planning long-term platform strategy.

Usage Scenarios

High-refresh gaming: Not this build's role. With GPU estimates derived from a 6252 Passmark G3D score and 808 Steel Nomad result — and again, no measured FPS data exists for this pairing — high-refresh expectations are unsupported by the numbers. The CPU could feed a much faster GPU.

Streaming: The CPU side is well-suited: 14 threads and an R23 multi-core score of 32,633 leave headroom for encoding alongside gameplay. The GPU side is the constraint — game rendering and encode simultaneously would strain a 6 GB, 44th-percentile card.

Video editing: Feasible at moderate scale. The CPU's multi-core throughput and 102.4 GB/s of DDR5 bandwidth carry timeline scrubbing and export, while the GPU's limited compute score (2,762 Passmark GPU compute) and 6 GB buffer cap heavy effects and acceleration.

3D rendering: Viewport work is acceptable; final-frame GPU rendering is not, given 4.198 TFLOPS FP32 and a Steel Nomad score of 808. CPU-based rendering at the 90th percentile is the realistic path, and the 24 MB of shared L3 supports it.

Software development: A strong fit. A Passmark single-thread score of 4,557, multithread score of 38,392, and 393,800 in data compression make compilation and repository operations quick. The locked multiplier does not impede development use.

Student and office work: Arguably the ideal scenario. Fast single-core responsiveness, multi-monitor support via three DisplayPort 2.0 and one HDMI output, modest 75 W GPU and 65 W CPU power draw, and capable casual gaming performance from the benchmark estimates.

Upgrade Path and Platform

The platform foundation is forward-looking: Intel Socket 1851, DDR5 support over a dual-channel bus, and PCIe Gen 5 with 20 CPU lanes. The Arc A380 itself connects over PCIe 4.0 x8, so it does not exploit the Gen 5 backbone — which is precisely why the GPU is the obvious upgrade target. Any faster card dropped into this slot hierarchy gains a 90th-percentile CPU and 20 Gen 5 lanes to work with, and the data suggests the CPU could drive a substantially stronger GPU before becoming the limiting factor.

Power headroom is generous. The GPU's suggested PSU is 250 W against its 75 W draw, and the CPU adds a 65 W TDP — a modest total that leaves substantial room for a much more demanding graphics card without PSU replacement, depending on the installed unit's rating. The GPU's compact 222 mm length and dual-slot width also mean most upgrade cards will fit where the A380 did, though buyers should verify case clearance for larger replacements qualitatively.

The sensible next upgrade, given the data, is the GPU: replacing the end-of-life A380 addresses the build's 44th-percentile ceiling while preserving its 90th-percentile CPU foundation, PCIe Gen 5 expansion, and DDR5 memory platform. The reverse — upgrading the CPU first — would yield almost nothing in graphics-bound scenarios, since the 235A already outclasses the A380 by every percentile measure in this fact set.