SYSTEM ANALYZER

Rate My PC: Intel Core Ultra 9 290K Plus + Intel Arc A770

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

97 / 100
ULTIMATE READY

Apex Performer

Top 3% 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
98%
VS
GPU
97%
PROCESSOR

Intel Core Ultra 9 290K Plus

84,003 Benchmark Score
Top 2% Market Ranking
View Full Specs →
GRAPHICS CARD

Intel Arc A770

68,809 Benchmark Score
Top 3% 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

# CPU Analysis

The Intel Core Ultra 9 290K Plus is a 24-core, 24-thread desktop processor built on Arrow Lake Refresh, Intel's second-generation Core Ultra architecture. It is manufactured on a 3 nm process at TSMC with 17,800 million transistors on a 243 mm² die. The chip runs at a 3.70 GHz base clock and boosts to 5.80 GHz, with a 125 W TDP. It uses the Intel Socket 1851 platform and supports dual-channel DDR5 memory with a bandwidth of 115.2 GB/s. ECC memory support is included, and the CPU provides 20 PCIe Gen 5 lanes. An integrated Arc Xe-LPG Graphics 64EU unit is present, and the multiplier is unlocked for overclocking.

The cache hierarchy consists of 192 KB of L1 per core, 3 MB of L2 per core, and 36 MB of shared L3. This is a substantial amount of on-die cache, particularly for a desktop chip without 3D V-Cache. In multi-threaded workloads, the 24 threads without hyper-threading represent a design choice that prioritizes efficiency and consistent scaling across physical cores. The absence of hyper-threading means each core is a full physical execution unit, which can be advantageous in workloads that are sensitive to core count rather than thread count.

Benchmark results place this CPU in the 96th percentile among all CPUs tested, with an average benchmark score of 84,003. The nearest rival is the Intel Core Ultra 9 285K, which scores 83,807 — a delta of just 0.2%. This means the 290K Plus is essentially a marginal refinement over its predecessor, not a generational leap. The AMD EPYC 4584PX scores 83,090 (1.1% behind), and the EPYC 9135 scores 82,980 (1.2% behind). Interestingly, the AMD EPYC 7F72 scores higher at 85,072, putting the 290K Plus 1.3% behind that server chip. These are tight margins, indicating that the 290K Plus sits at the very top of the desktop CPU hierarchy but does not dominate server-class silicon in raw multi-threaded throughput.

In single-threaded performance, the Cinebench R23 single-core score of 7,303 is exceptional. This is the kind of result that translates directly to snappy application responsiveness and strong performance in lightly-threaded tasks like web browsing, office productivity, and legacy software. The multi-core Cinebench R23 score of 51,731 shows that the 24 cores scale well under heavy loads, making this a serious contender for rendering and compilation workloads. The PassMark single-thread score of 4,823 reinforces the strong per-core performance, while the PassMark multi-thread score of 60,860 confirms that the chip leverages its core count effectively.

Data-centric workloads show mixed but generally strong results. PassMark data compression scores 698,346, which is a very high number and suggests excellent performance in file archiving and database operations. Data encryption scores 52,563, and extended instructions score 52,338 — both indicate robust SIMD and cryptographic throughput. Floating-point math scores 214,760, while integer math scores 166,194. The find prime numbers test scores 503, which is low relative to other integer operations, but this is a specialized test that often favors different microarchitectural traits. Random string sorting scores 79,744, and physics scores 3,315. The average benchmark score of 84,003 places this chip in elite company, with only the fastest server and workstation parts outperforming it.

# FAQ

Q: How does the Intel Core Ultra 9 290K Plus compare to its direct predecessor, the Core Ultra 9 285K?

A: The 290K Plus scores an average of 84,003 across all benchmarks, which is just 0.2% higher than the 285K's 83,807. This is a very small improvement, essentially a refresh rather than a new generation.

Q: Is this CPU better than high-end AMD EPYC server processors?

A: In average benchmark score, the 290K Plus at 84,003 trails the AMD EPYC 7F72, which scores 85,072 (1.3% higher). However, it beats the EPYC 4584PX (83,090, 1.1% lower) and the EPYC 9135 (82,980, 1.2% lower). So it is competitive with but not superior to the fastest EPYC parts.

Q: What is the memory bandwidth of this CPU and does it support ECC?

A: The 290K Plus supports dual-channel DDR5 memory with a bandwidth of 115.2 GB/s. It also supports ECC memory, which is a notable feature for workstation builds where data integrity is critical.

Q: Does this CPU have integrated graphics?

A: Yes, it includes an integrated Arc Xe-LPG Graphics 64EU unit. This is not a gaming-grade solution, but it can handle display output and basic graphics tasks without requiring a discrete GPU.

Q: What socket does this CPU use and can it be overclocked?

A: It uses Intel Socket 1851, and the multiplier is unlocked, meaning it can be overclocked. This is a desktop-class chip with a 125 W TDP.

Q: How fast is the CPU in single-threaded workloads?

A: The Cinebench R23 single-core score is 7,303, and the PassMark single-thread score is 4,823. These are top-tier numbers that indicate excellent performance in everyday tasks and lightly-threaded applications.

Q: What is the process node and transistor count?

A: The CPU is fabricated on a 3 nm process at TSMC, with 17,800 million transistors on a 243 mm² die. This is a advanced manufacturing process that contributes to the chip's efficiency and performance.

# GPU Analysis

The Intel Arc A770 is a desktop graphics card based on the DG2-512 chip and the Xe-HPG architecture, part of the Alchemist generation. It is built on a 6 nm process at TSMC with 21,700 million transistors on a 406 mm² die, yielding a transistor density of 53.4 million per mm². The card runs at a base clock of 2100 MHz and boosts to 2400 MHz. Memory operates at 2000 MHz with 16 Gbps effective speed. It has a 225 W TDP, requires a 1x 6-pin and 1x 8-pin power connector, and is a dual-slot card. The suggested PSU is 550 W. It connects via PCIe 4.0 x16 and offers display outputs of 1x HDMI 2.1 and 3x DisplayPort 2.0. The GPU is now end-of-life, with its successor being Battlemage, and it was released on 2022-10-11.

The memory subsystem is a strong point: 16 GB of GDDR6 on a 256-bit bus delivers 512.0 GB/s of bandwidth. This is a generous amount of VRAM for a card in this performance class, and the bandwidth is sufficient for high-resolution textures and compute workloads. The GPU has 4096 shading units, 256 texture mapping units, and 128 raster operation pipelines. It also includes 32 ray tracing cores, although dedicated tensor cores are not listed. Pixel rate is 307.2 GPixel/s, and texture rate is 614.4 GTexel/s. Compute performance is rated at 19.66 TFLOPS for FP32 and 39.32 TFLOPS for FP16 at a 2:1 ratio. API support includes DirectX 12 Ultimate (12_2), OpenGL 4.6, and Vulkan 1.4.

In benchmark results, the Arc A770 scores 2,969 in 3DMark Steel Nomad (DX12), 109,175 in Geekbench OpenCL, and 94,284 in Geekbench Vulkan. Its average benchmark score is 68,809, which places it in the 90th percentile among all GPUs. The nearest rivals are the NVIDIA CMP 90HX with an average score of 69,000 (0.3% higher), the AMD Radeon Instinct MI25 at 68,562 (0.4% lower), the AMD Radeon Pro WX 8200 at 69,870 (1.5% higher), and the NVIDIA Quadro P6000 at 69,986 (1.7% higher). These are all professional or mining-oriented cards, which says something about the Arc A770's positioning: it is a consumer card that trades blows with workstation-class hardware from the previous generation. The launch MSRP was 329 USD.

The ray tracing hardware, with 32 dedicated cores, allows for hardware-accelerated ray tracing in supported games, though the performance is not class-leading. The lack of tensor cores means that AI-accelerated features rely on the Xe-HPG architecture's general-purpose compute. For rendering workloads, the 16 GB VRAM is a significant advantage, allowing larger scenes and higher-resolution textures than cards with 8 GB. The 512.0 GB/s bandwidth supports demanding compute tasks, and the FP32 throughput of 19.66 TFLOPS is respectable for a card in this tier.

# Who Should Build It

The Intel Core Ultra 9 290K Plus combined with the Intel Arc A770 targets a specific niche: users who want top-tier CPU performance but are willing to compromise on GPU raw power in favor of Intel's ecosystem and VRAM capacity. The CPU's 96th percentile ranking among all CPUs means it is suitable for any CPU-bound workload. Content creators working with video editing, 3D rendering, or software compilation will find the 24 cores and 36 MB of L3 cache beneficial. The Cinebench R23 multi-core score of 51,731 indicates excellent scaling in tasks like Blender rendering or Premiere Pro exports.

The GPU's 90th percentile ranking and 16 GB of VRAM make it interesting for developers working with machine learning inference or compute tasks that require large memory footprints. The OpenCL score of 109,175 demonstrates strong compute capability. Software developers who compile large codebases will benefit from the CPU's multi-threaded performance, and the ECC memory support on the CPU side is a plus for long-running build servers or data processing tasks. Students in engineering or data science programs will find the CPU performance more than adequate for their coursework, though the GPU may be overkill for basic tasks.

Small business workstations that run virtualization, database management, or financial modeling will appreciate the CPU's 96th percentile performance and the GPU's compute capabilities. The GPU's 16 GB VRAM is particularly useful for CAD or GIS applications that load large datasets into video memory. Gamers at 1440p or 4K resolution who play less demanding titles will find the Arc A770 sufficient, but those targeting high refresh rates at 1080p may be disappointed. The CPU's single-threaded performance is excellent, so esports titles that rely on CPU speed will run well, but the GPU will become the limiting factor in graphically intensive games.

# Benchmark Performance

The combined picture of this build is a desktop system that sits at the 93rd percentile overall. The CPU is the dominant component, with an average benchmark score of 84,003 and a 96th percentile ranking. The GPU is comparatively weaker, with an average score of 68,809 and a 90th percentile ranking. The delta between CPU and GPU percentiles is significant: the CPU is among the best available, while the GPU is merely above average.

In CPU benchmarks, the Cinebench R15 multi-core score of 5,214 and single-core score of 736 show that the chip performs well in both threaded and single-threaded workloads. The Cinebench R20 scores of 21,727 (multi-core) and 3,067 (single-core) are strong. The R23 scores of 51,731 and 7,303 are the most indicative of modern rendering performance. PassMark results show a balanced profile across integer math (166,194), floating-point math (214,760), and extended instructions (52,338).

For the GPU, the Steel Nomad DX12 score of 2,969 is the only gaming-oriented benchmark, and it is modest. The Geekbench OpenCL score of 109,175 and Vulkan score of 94,284 indicate better compute performance than gaming performance. The nearest rivals for the GPU are all professional or mining cards, which suggests that its compute performance is competitive with older workstation GPUs but its gaming performance is not directly comparable to those parts.

When viewed together, this build is CPU-heavy. The 96th percentile CPU will rarely be the bottleneck in any workload, while the 90th percentile GPU will limit performance in GPU-bound tasks. For users who prioritize productivity and compute over gaming, this is a sensible pairing. For gamers, the GPU is the weaker link, and the CPU's power may go underutilized.

# Usage Scenarios

High-refresh gaming: At 1080p with competitive settings, the CPU's single-threaded strength will drive high frame rates, but the Arc A770's 90th percentile GPU performance will cap frame rates in most titles. Expect playable but not elite-level performance at high refresh rates.

Streaming: The 24-core CPU can handle encoding and streaming workloads without breaking a sweat. The Cinebench R23 multi-core score of 51,731 ensures that encoding overhead will not significantly impact game performance, though the GPU's modest gaming prowess may limit overall stream quality settings.

Video editing: The CPU's multi-core performance is excellent for export and rendering tasks. The GPU's 16 GB VRAM and 512.0 GB/s bandwidth will accelerate effects and timeline previews in applications that support Intel's GPU acceleration. Average benchmark scores suggest a capable editing workstation.

3D rendering: In CPU-based renderers like Blender Cycles, the 24-core processor with a 51,731 Cinebench R23 score will produce strong results. GPU-accelerated rendering will benefit from the Arc A770's 19.66 TFLOPS FP32 and 16 GB VRAM, though it will not match the fastest dedicated render cards.

Software development: Compilation tasks will fly with the 24 cores and 60,860 PassMark multi-thread score. The ECC memory support is a bonus for long-running builds. The GPU is largely irrelevant here, but its compute capability can accelerate test workloads.

Student and office work: This is overkill for basic tasks, but the 4,823 PassMark single-thread score means snappy application response. The integrated Arc Xe-LPG Graphics can handle display output, and the discrete GPU is unnecessary for office workloads but available for any compute assignments.

# Gaming Performance

No measured FPS rows exist for this exact CPU+GPU combination in the FACT PACK. All frame rate discussion below is estimated from the benchmark scores, not measured data. The GPU's 3DMark Steel Nomad score of 2,969 and its 90th percentile ranking among all GPUs provide a baseline for expectations. The CPU's 96th percentile ranking means it will not bottleneck the GPU in most games.

At 1080p ultra settings, the Arc A770 is likely to deliver playable frame rates in most titles, but it will not compete with higher-tier cards. The 16 GB VRAM is ample for 1080p textures, but the 19.66 TFLOPS FP32 throughput limits raw rasterization performance. At 1440p ultra, expect frame rates to drop further, and the GPU will become the clear bottleneck. At 4K ultra, the card will struggle in demanding titles, though less demanding games or those with upscaling may remain playable.

The GPU's nearest rivals are professional cards like the Quadro P6000 and Radeon Pro WX 8200, which are not gaming-focused. This suggests that the Arc A770's gaming performance is comparable to older workstation cards, which were not fast in games even when new. Ray tracing performance, with 32 RT cores, will be present but likely underwhelming in titles that use heavy ray tracing effects. The Vulkan score of 94,284 suggests that games using Vulkan may perform better than those using DX12, though the Steel Nomad DX12 score of 2,969 indicates baseline DX12 capability.

# Build Overview

This is a desktop build that pairs Intel's top-tier Core Ultra 9 290K Plus processor with Intel's Arc A770 discrete GPU. The CPU is an Arrow Lake Refresh part with 24 cores and 24 threads, a 5.80 GHz boost clock, and a 96th percentile ranking among all CPUs. The GPU is an Alchemist-generation Xe-HPG card with 16 GB of GDDR6 memory and a 90th percentile ranking among all GPUs. The combined percentile for this pairing is 93rd, which places it in the upper echelon of desktop systems.

The CPU is the star of this build. Its average benchmark score of 84,003 puts it just 0.2% ahead of the Core Ultra 9 285K and within 1.3% of the AMD EPYC 7F72. This is a processor that can handle almost any workload with authority. The GPU, while not weak, is clearly the secondary component. Its average score of 68,809 is competitive with older professional cards but not with modern high-end gaming GPUs.

This is not a balanced build in the traditional sense. It is a CPU-first system with a capable but not spectacular GPU. The 16 GB VRAM is a redeeming feature for compute workloads, but gamers may find the GPU limiting. The build class is desktop, and the overall tier, based on the 93rd combined percentile, is high-end but not flagship. For users who need maximum CPU performance and only moderate GPU power, this is a coherent choice. For gamers, the GPU should be upgraded or the CPU downgraded for better balance.

# Balance and Bottleneck

The bottleneck analysis for this build is straightforward: the CPU is far more powerful than the GPU. The CPU sits in the 96th percentile, while the GPU is in the 90th percentile. In CPU-bound workloads like compilation, rendering, or data processing, the CPU will be the limiting factor only if the workload exceeds its capabilities, which is rare. In GPU-bound workloads like gaming or GPU-accelerated rendering, the GPU will be the bottleneck almost every time.

Evidence from benchmark scores supports this. The CPU's Cinebench R23 multi-core score of 51,731 is exceptional, while the GPU's Steel Nomad DX12 score of 2,969 is modest. The CPU's PassMark multi-thread score of 60,860 dominates, while the GPU's OpenCL score of 109,175, while strong, is not in the same league relative to its peers. When the GPU is under load in a game, the CPU will have significant headroom, meaning frame rates will be dictated by the GPU.

In mixed workloads like video editing, the balance depends on the task. Exporting video uses the CPU heavily, and the 24 cores will accelerate this. Applying effects or using GPU-accelerated encoding will shift the load to the GPU, which will then become the limiting factor. For 3D rendering, CPU-based renderers will favor this build, but GPU-based renderers will expose the GPU's limitations.

The FPS scaling in gaming reflects this imbalance. Since no measured FPS data exists for this combination, estimates from benchmark scores indicate that the GPU will cap frame rates at levels consistent with a 90th percentile card. The CPU's performance will not be the constraint in games, meaning that upgrading the GPU would yield disproportionate FPS gains without changing the CPU. Conversely, upgrading the CPU would yield negligible gaming improvements. This is a build where the CPU is the long-term investment and the GPU is the short-term compromise.