A game built on Epic’s Unreal Engine 5 is supposed to lean on the graphics card. Gears of War E-Day apparently didn’t get that memo. According to a report published by Tom’s Hardware on October 1, 2026, the studio’s loading screens and early gameplay push CPU usage to 80-90% even with a monitoring overlay running in the background, a pattern more associated with simulation-heavy strategy titles than a cover-based shooter. Tom’s Hardware ran the game across 25 desktop CPUs from Intel and AMD and then dug into a feature called Low Core Mode, finding that it made things slower on every single chip tested, including a quad-core budget part that lost more than half its frame rate.
The result is an unusual data point in a year where GPU shortages and price spikes have dominated hardware headlines (shattered.io has tracked the RTX 5090 disappearing from US retail and GPU prices climbing across Europe). Here, the bottleneck isn’t the graphics card at all. It’s the processor, and specifically how many cores Unreal Engine 5 can actually put to work.
What Tom’s Hardware Found Benchmarking Gears of War E-Day
Tom’s Hardware’s test article, published October 1, 2026, set out to answer a simple question: how much CPU does Gears of War E-Day actually need? The outlet ran the same test scene and settings across 25 desktop processors spanning current and recent Intel and AMD lineups. During loading sequences, with a hardware monitoring overlay open, CPU utilization sat between 80% and 90%, a figure the outlet called out as unusually high for a shooter built on an engine best known for leaning on the GPU.
That utilization pattern matters because it changes who should worry about a CPU upgrade before buying this game. Most Unreal Engine 5 titles released over the past two years, including the engine migrations covered in our report on Unity 7’s move toward Unreal-based rendering, still bottleneck primarily on the graphics card at 1440p and above. Gears of War E-Day flips that assumption, at least during the loading and early-mission window Tom’s Hardware measured.
Separately, the Japanese outlet GIGAZINE also covered the game’s CPU behavior, adding to the pile of outlets flagging the same pattern around the October 1 timeframe. Neither outlet’s reporting included a full breakdown of every one of the 25 CPUs tested, the exact graphics card paired with them, or the resolution used for each result, so this piece sticks to the figures both outlets actually confirmed rather than guessing at the rest.
Why Unreal Engine 5 Makes This Game Unusually CPU-Bound
Unreal Engine 5 games typically spend most of their CPU budget on three things: world streaming, physics, and the animation and AI logic running behind the camera. Lumen (dynamic global illumination) and Nanite (virtualized geometry) are mostly GPU workloads, which is why so many UE5 releases scale with graphics card tier rather than core count. The PC requirements Remedy published for Control Resonant followed that pattern closely, with storage and VRAM dominating the conversation rather than processor class.
Gears of War E-Day’s loading-screen CPU spike suggests something different is happening under the hood, likely tied to how the game streams assets and spins up world state before handing control to the player. Tom’s Hardware didn’t publish the exact systems or subsystems responsible, so any specific explanation beyond “the loading and early-game path is CPU-heavy” would be speculation. What is confirmed is the symptom: a sustained 80-90% CPU load reading during that window, observed consistently enough to become the headline finding of the piece.
The 25-CPU Test Lineup: Intel and AMD Head to Head
Running 25 CPUs through a single test scene is a serious undertaking, and it puts Gears of War E-Day in the same benchmarking spotlight as the wider CPU market battles playing out this year. AMD and Intel have been trading blows across desktop, laptop, and server lines, from the EPYC Venice comparisons against Xeon in the data center to Panther Lake’s laptop debut alongside a 10% CPU price increase. A single game forcing a 25-chip shootout, spanning both vendors, is a rare and useful real-world stress test rather than a synthetic benchmark.
Tom’s Hardware’s reporting confirms the lineup included Intel’s Core Ultra 5 225, Core Ultra 5 245K, Core Ultra 5 250K Plus, Core Ultra 7 265K, and Core Ultra 7 270K Plus, plus the older quad-core Core i3-14100 and AMD’s Ryzen 7 9800X3D. The outlet did not publish the complete 25-chip roster, the test-system specification, or the graphics card used, so readers should treat any claim about the other 18 or so CPUs as unconfirmed until Tom’s Hardware’s full piece or a follow-up fills in those gaps.
Ryzen 7 9800X3D Hits Full Utilization at 1080p
Among the chips named in the report, AMD’s Ryzen 7 9800X3D, long considered the gaming benchmark reference point thanks to its 3D V-Cache, was identified as reaching full CPU utilization at 1080p. That is a notable result on its own. The 9800X3D typically has so much headroom in most games that reviewers use it specifically to remove the CPU as a variable and isolate GPU performance. Gears of War E-Day apparently saturates it anyway, at least during the loading sequence Tom’s Hardware measured with its overlay open.
Tom’s Hardware’s report didn’t publish a frame-rate figure for the 9800X3D specifically, so this analysis isn’t able to say how that utilization translated into playable performance. What’s confirmed is the utilization reading itself: full saturation on a chip that rarely gets fully taxed by games at 1080p, a resolution where the bottleneck almost always sits with the processor rather than the graphics card.
What Is Low Core Mode, and Why Does It Exist
Low Core Mode is a setting Tom’s Hardware tested separately, running the identical scene and settings a second time with it switched on. The name implies it’s meant to help systems with fewer CPU cores, likely by trimming background threads, reducing streaming parallelism, or scaling down simulation workloads so a four- or six-core chip isn’t overwhelmed. That’s the logical intent behind a setting with this name, though the developer’s own description of what Low Core Mode actually changes under the hood wasn’t part of the confirmed reporting, so it’s worth treating the mechanism as inferred rather than officially documented.
What makes the finding newsworthy is that the mode didn’t behave as its name suggests. Instead of being neutral-to-positive for low-core systems and a non-factor for high-core systems, Tom’s Hardware found it reduced performance across the entire 25-CPU lineup, high-core and low-core chips alike.
Low Core Mode Benchmark Results: Losses Across the Board
Tom’s Hardware published five specific Low Core Mode comparisons on Intel’s Core Ultra line, each showing a stock frame rate, a Low Core Mode frame rate, and the resulting percentage drop. Every single one lost performance, and the losses got worse rather than better as the chips moved up the Core Ultra stack, which is the opposite of what you’d expect if the mode were simply redistributing work more efficiently for lower core counts.
| CPU | Stock FPS | Low Core Mode FPS | Change |
|---|---|---|---|
| Intel Core Ultra 5 225 | 116.8 | 100.3 | -14.1% |
| Intel Core Ultra 5 245K | 138.7 | 111.1 | -19.9% |
| Intel Core Ultra 5 250K Plus | 144.3 | 113.4 | -21.4% |
| Intel Core Ultra 7 265K | 147.5 | 114.4 | -22.4% |
| Intel Core Ultra 7 270K Plus | 151.6 | 120.2 | -20.7% |
Read across the table and a pattern jumps out: the Core Ultra 7 265K, a more capable chip than the Core Ultra 5 225, actually lost a larger share of its performance with Low Core Mode enabled (-22.4% versus -14.1%). If the setting were purely a concession for weaker hardware, you’d expect the opposite curve. Instead, every tested Intel chip in this list gave up between 14% and 22% of its frame rate for a feature ostensibly designed to help, not hurt, CPU-constrained systems.
The Intel Core i3-14100 Edge Case: 89 FPS to 36 FPS
The most dramatic number in Tom’s Hardware’s report didn’t come from the Low Core Mode toggle at all. It came from a separate experiment on the quad-core Intel Core i3-14100, where the outlet reduced the chip from four active cores down to two. Average performance fell from 89 FPS to 36 FPS, a 59% decrease, the steepest drop recorded anywhere in the test.
That figure underlines just how far Gears of War E-Day leans on having at least four real cores available. A 59% frame-rate collapse from cutting two cores is the kind of result normally reserved for heavily multithreaded productivity software, not a third-person shooter. It also reinforces why Low Core Mode performing worse, not better, on constrained hardware is the headline of this story rather than a footnote: the i3-14100 result shows the game genuinely needs the cores it’s given, which makes a feature that trims core usage anyway look counterproductive.
The One Case Where Low Core Mode Actually Helped
Tom’s Hardware’s reporting included one exception worth flagging precisely because it complicates the “always worse” headline. In at least one test configuration, enabling Low Core Mode resolved an asset-streaming issue paired with outright game freezing, and once that stall was gone, performance in that specific case improved by 30%. The outlet didn’t specify which CPU or test pass produced that result, so it should be read as a situational fix for a stability bug rather than a general performance recommendation.
That nuance matters for anyone deciding whether to flip the setting. If a system is experiencing freezing or stuttering tied to asset streaming, Low Core Mode may be worth testing as a stability workaround. For everyone else, based on the five Core Ultra data points above and the quad-core result, leaving it off and running stock settings produced the higher frame rate in every confirmed case.
Table: What Tom’s Hardware Confirmed vs. What’s Still Unverified
Given how much attention a 25-CPU benchmark generates, it’s worth being explicit about where the confirmed reporting ends and speculation would begin. This site is not going to fill those gaps with invented numbers.
| Detail | Status | What’s known |
|---|---|---|
| Number of CPUs tested | Confirmed | 25 desktop CPUs from Intel and AMD |
| Full CPU model list | Not disclosed | Only 7 specific models named in available reporting |
| Graphics card used | Not disclosed | No GPU model specified in available reporting |
| Resolution for every result | Partially disclosed | 1080p confirmed for the Ryzen 7 9800X3D utilization finding only |
| CPU utilization during loading | Confirmed | 80-90% with monitoring overlay open |
| Low Core Mode net effect | Confirmed | Reduced performance on all 25 tested CPUs |
| Named on-record quote from developer | Not available | No attributed quotation in available reporting |
CPU Utilization Overlay: Why 80-90% Stands Out
Running a monitoring overlay like RivaTuner Statistics Server or HWInfo alongside a game typically adds a small amount of CPU overhead itself, which is part of why Tom’s Hardware’s own measurement conditions matter here. Even accounting for that overhead, a sustained 80-90% utilization reading during loading is high enough to flag a genuine CPU-bound sequence rather than measurement noise. Most modern shooters built on Unreal Engine 5 sit well below that threshold during loading screens, since asset streaming is typically disk- and GPU-decompression bound rather than CPU bound.
For players who already use a hardware monitoring tool, this is an easy thing to verify independently once the game is available: watch the overlay during the loading transition into a mission and see whether utilization spikes the way Tom’s Hardware described. It’s a good reminder that community benchmarking and overlay screenshots are likely to keep generating headlines around this title well past its launch week.
How to Check Your Own Core Scaling Before Buying a CPU
Readers trying to figure out whether their own CPU is likely to be a bottleneck can run a simple core-count experiment similar in spirit to what Tom’s Hardware did with the i3-14100, using Windows’ built-in affinity controls rather than BIOS core disabling. This won’t replicate Low Core Mode specifically, since that’s a setting inside the game, but it’s a reasonable way to see how a title’s engine scales with fewer logical processors.
:: Check how many physical cores and logical processors Windows sees
wmic cpu get NumberOfCores,NumberOfLogicalProcessors
:: Launch a game restricted to 2 logical processors to approximate a low-core test
start /affinity 3 "" "C:\Path\To\Game.exe"The affinity mask in the second command (3 in hexadecimal, meaning the first two logical processors) is a rough approximation, not an exact match for how a game engine schedules threads internally. It’s a starting point for curious players, not a substitute for the kind of controlled, repeatable testing Tom’s Hardware ran across its 25-chip lineup.
How This Compares to Other Recent Unreal Engine 5 Releases
Gears of War E-Day’s CPU appetite stands out partly because of what else has shipped on Unreal Engine 5 this year. CD Projekt Red’s Witcher 3 Remastered leaned heavily on GPU driver optimization to fix its performance complaints, not CPU tuning. Remedy’s Control Resonant, by its own published specs, asks more of storage and VRAM than of the processor. Even Epic’s own engine roadmap, with Unity 7’s decision to render through Unreal rather than compete with it, has been framed around graphics fidelity rather than CPU-side simulation work.
That makes Gears of War E-Day something of an outlier in the current UE5 catalog: a title where the processor, not the graphics card, decides whether the loading screen and early mission run smoothly. For a franchise built around large-scale combat encounters and destructible environments, a CPU-heavy simulation layer isn’t entirely surprising in hindsight, but it cuts against the general assumption that “Unreal Engine 5 game” automatically means “GPU-bound game.”
Historical Context: CPU-Bound Console Shooters Aren’t New
The original Gears of War trilogy, built on Epic’s Unreal Engine 3, actually has a history of CPU-sensitive moments too, particularly during large set-piece battles with heavy physics and destructible cover. What’s different this time is the scale of the public benchmarking: a 25-CPU shootout from a major outlet, published the same week the game’s CPU demands became apparent, is a far more rigorous and visible spotlight than anything the earlier games in the series received at launch.
It’s also a reminder that CPU bottlenecks in gaming tend to resurface every few years, usually tied to a specific engine feature rather than a steady industry trend. World streaming, destructible geometry, and large enemy counts have each produced their own CPU-bound outliers in the past, and this is shaping up to be this generation’s version of that pattern, at least during the specific loading sequence Tom’s Hardware measured.
Market Impact: What This Means for PC Gamers Shopping for a CPU
For anyone building or upgrading a PC specifically with Gears of War E-Day in mind, the practical takeaway from Tom’s Hardware’s reporting is straightforward: more cores help, and the Low Core Mode toggle is not a free performance boost for budget builds. It’s also a reminder that frame-generation and upscaling technologies like Nvidia’s DLSS don’t offset a CPU-side bottleneck the way they can smooth out GPU-limited scenes, since the slowdown Tom’s Hardware measured happens before the GPU ever becomes the limiting factor. Given the pricing pressure already hitting AMD’s processor lineup and Intel’s own Core Ultra prices climbing alongside Panther Lake’s launch, buyers weighing a quad-core or low-end hex-core chip specifically for this game should treat the i3-14100’s 59% collapse as a worst-case data point rather than an outlier to dismiss.
It’s also a data point retailers and system builders are likely to cite in marketing for pre-built gaming PCs ahead of the game’s wider release, especially since the Ryzen 7 9800X3D result shows even a top-tier gaming CPU gets pushed to full utilization. Expect “optimized for Gears of War E-Day” messaging to show up in CPU bundle listings once the title is more widely available, in the same way the Witcher 3 Remastered driver saga generated a wave of AMD-specific marketing earlier this year.
Competitive Comparison: Intel Core Ultra vs. AMD in This Workload
The confirmed Low Core Mode figures only cover Intel’s Core Ultra 5 and Core Ultra 7 chips, so a direct apples-to-apples Intel-versus-AMD Low Core Mode comparison isn’t possible from the available reporting. What is clear is that AMD’s flagship gaming chip, the Ryzen 7 9800X3D, was singled out for reaching full utilization at 1080p, a result that stands on its own regardless of the Low Core Mode question. Combined with AMD’s continued 3D V-Cache advantage in cache-sensitive workloads and Intel’s Core Ultra line carrying higher core counts at the top end, this looks like a workload where raw core count and cache size both matter, rather than one vendor having a clean advantage.
That nuance tracks with the broader 2026 CPU market story, where AMD and Intel have traded leads depending on the workload rather than either vendor sweeping every category, a dynamic also visible in AMD’s EPYC Venice server chip comparisons against Intel Xeon.
Predictions: Where This Story Goes From Here
- Expect a developer response or patch note addressing Low Core Mode once the full community benchmark cycle plays out after wider release, given how clearly it underperforms its own name across every confirmed test.
- Other outlets will likely publish their own CPU benchmark passes on Gears of War E-Day in the days following the Tom’s Hardware report, filling in the remaining CPUs from the 25-chip lineup that weren’t individually named.
- System builders and PC OEMs will probably start referencing this benchmark in marketing copy for gaming desktops, similar to how specific game benchmarks have been used to sell CPU upgrades in past console-to-PC ports.
- Budget-tier quad-core buyers are likely to see more scrutiny of “minimum spec” CPU recommendations for this title specifically, given the magnitude of the i3-14100 result.
- If the CPU-heavy loading behavior persists into the full release, expect comparisons to other upcoming Unreal Engine 5 titles to increasingly ask about CPU scaling up front rather than assuming GPU-bound performance by default.
What Tom’s Hardware’s Report Didn’t Cover
It’s worth restating plainly: the available reporting does not include the complete list of all 25 tested CPUs, the test-system specification, the graphics card paired with the CPUs, every result’s exact resolution, or pricing for the hardware involved. No directly attributed on-record quotation from a developer, Microsoft, or The Coalition appeared in the available coverage either. Any article claiming otherwise about this specific benchmark should be treated with skepticism until a fuller breakdown is published.
That gap is itself useful information. It tells you that the headline numbers, the 14-22% Low Core Mode losses on Core Ultra chips and the 59% collapse on the i3-14100, are solid, specific, and attributable, while everything else about the test methodology remains an open question for now.
Frequently Asked Questions
Is Gears of War E-Day a CPU-intensive game?
Based on Tom’s Hardware’s October 1, 2026 benchmark, yes, at least during loading sequences, where CPU utilization measured 80-90% with a monitoring overlay running, and where even AMD’s Ryzen 7 9800X3D reached full utilization at 1080p.
What is Low Core Mode in Gears of War E-Day?
It’s a setting Tom’s Hardware tested that, based on its name, appears intended to help systems with fewer CPU cores. In testing, it reduced frame rates across all 25 CPUs tried, including high-core-count Intel Core Ultra chips, rather than only affecting lower-core systems.
Should I enable Low Core Mode?
Based on the confirmed Core Ultra results, leaving it off produced higher frame rates in every documented case. The one exception reported was a situation where it fixed an asset-streaming freeze and then delivered a 30% improvement, so it may be worth testing only if you’re experiencing stuttering or freezing tied to loading.
How many CPUs were benchmarked?
Tom’s Hardware tested 25 desktop CPUs from Intel and AMD. The outlet’s available reporting names seven specific models: the Intel Core Ultra 5 225, Core Ultra 5 245K, Core Ultra 5 250K Plus, Core Ultra 7 265K, Core Ultra 7 270K Plus, Core i3-14100, and AMD’s Ryzen 7 9800X3D.
What happened with the Intel Core i3-14100?
Tom’s Hardware reduced the quad-core chip from four active cores to two and recorded average performance falling from 89 FPS to 36 FPS, a 59% decrease, the single largest drop in the entire test.
Does this mean Gears of War E-Day needs a high-end CPU to run well?
The confirmed data points to processor headroom mattering more than it does in most recent Unreal Engine 5 releases, particularly during loading. Full minimum and recommended CPU specifications from the developer were not part of the available reporting as of this article’s publication.
Is Gears of War E-Day GPU-bound or CPU-bound?
Based on the confirmed findings, the loading sequence and early gameplay window tested by Tom’s Hardware leaned CPU-bound, which is unusual for a game built on Unreal Engine 5. Graphics card behavior for the rest of the game wasn’t part of the available reporting.
Which outlets reported on this benchmark?
Tom’s Hardware published the primary 25-CPU benchmark and Low Core Mode investigation on October 1, 2026. GIGAZINE also covered the game’s CPU behavior around the same timeframe.
