Wolverine Hits 20 FPS Peaks On PC While Most Play Runs 6 To 9

By Indie Kings | October 4, 2026

Updated October 4, 2026: A new test reported by VideoCardz shows Marvel's Wolverine reaching around 20 FPS in lighter scenes on a Ryzen 7 9800X3D plus RTX 4080 using Senaxx's September 30 KytyPS5 build. Most gameplay in the same capture runs closer to 6 to 9 FPS, with heavier sequences lower, so the peak is not the sustained rate.

Image: White wireless game controller. Credit: Wikimedia Commons, public domain, Evan-Amos.

The photograph above shows an Xbox 360 controller. It is not Wolverine gameplay and it is not the emulator. It is a generic controller illustration only.

This is experimental open source PS5 emulation research. The game in this report is a commercial PS5 title. This article links no emulator download, no build, no guide, no video, and no game files of any kind.

20 FPS Peaks, 6 To 9 Sustained

The headline figure is a peak of around 20 FPS in lighter scenes. The figure comes from a single test video by Unreal Gaming PC, as reported by VideoCardz by WhyCry. The test hardware shown is a Ryzen 7 9800X3D plus GeForce RTX 4080.

The sustained figure is lower than the peak. Most sections in the same capture run closer to 6 to 9 FPS. Heavier sequences drop even lower than that range.

The distinction between peak and sustained matters for readers. A peak describes a brief favorable moment in a light scene. A sustained rate describes what the same test holds across combat and playable sections.

The capture reportedly includes cutscenes, combat and playable sections. Lighter scenes occasionally reach around 20 FPS. Typical play stays in the single digits.

The result is still far from playable by normal standards. Playable implies stable input response and stable frame pacing across scenes. This test shows sustained rendering with brief faster periods, not stable play.

Flagship hardware does not change that reading. The Ryzen 7 9800X3D plus RTX 4080 is a high end PC combination. Even on that combination, the reported typical rate is 6 to 9 FPS.

The table below restates the figures exactly as reported. Each value carries its standing because this is one test video. No broader benchmark claim is supported.

The peak of around 20 FPS is therefore a light scene number on flagship hardware. It is not presented here as playable or typical. The typical rate in this capture remains 6 to 9 FPS.

Attribution stays with the stated chain. VideoCardz by WhyCry reported the test. Unreal Gaming PC produced the test video using Senaxx's KytyPS5. A second video by PSXTech was also noted in the same report.

Naming KytyPS5 in prose is permitted in this article. No link to any emulator download, build, guide, video, or game file is included. That boundary is intentional for a commercial PS5 title.

Readers should treat 20 FPS as a progress marker. It shows rendering can briefly reach higher in easy scenes. It does not show that the game holds that rate where action is dense.

The practical takeaway is narrow. Rendering is sustained across sections that previously did not run. Speed remains low across most of the capture.

This section establishes the core split. Peak performance touches around 20 FPS. Sustained performance sits at 6 to 9 FPS with heavier scenes lower.

Any comparison with other frame rate reports should wait for matched builds and matched hardware. Different builds and different hardware make such comparisons non direct. That caution is stated explicitly because casual comparisons would mislead.

The article does not state PC requirements. No minimum or recommended specification can be derived from one capture. Any performance prediction beyond these reported figures is marked [NEEDS VERIFICATION].

The Build And The Capture

The build used is Senaxx's KytyPS5, released September 30. That build date matters because the project warns that behavior changes between builds. A result on one build is not a result for every build.

Unreal Gaming PC is named here as the tester behind the test video. Senaxx is named as the builder whose KytyPS5 release was used. PSXTech is named because a second video was noted in the same VideoCardz report.

No video is linked in this article. No test frames are reposted. The discussion below describes the reported contents without reproducing them.

The capture reportedly shows cutscenes, combat and playable sections. That breadth matters because menu only progress would be a smaller claim. Playable sections mean the game is rendering interactive sequences, even at low speed.

The test video has 10,645 views at the time of the report as cited in the provided facts. View count does not validate the frame rate. It only shows the size of the audience for this single capture.

The prior range was roughly 3 to 7 FPS. Those earlier figures came from different builds and different hardware. They are therefore a non direct comparison with the current 6 to 9 FPS typical range.

Non comparability needs plain language. Same game plus different emulator build plus different PC can shift results for many reasons. Without matched controls, a speed increase cannot be measured as a clean percentage.

The direction of travel is still visible. Earlier work reached menus in late September. Then first in game rendering appeared at a few FPS. Now sustained rendering appears with brief 20 FPS periods.

Direction is not the same as rate stability. The current capture still spends most time at 6 to 9 FPS. Heavier sequences fall below that band.

Capture conditions add further limits. Recording, overlay software, driver version, operating system settings, and background load can all affect a single run. None of those variables are established here beyond the named CPU, GPU, and build.

The article therefore treats the capture as one data point. One data point can document that a scene reached around 20 FPS. One data point cannot establish average performance for the game.

Build dependence is a central warning. The KytyPS5 project describes compatibility as limited. It warns that behavior changes between builds.

That warning means readers should not assume the September 30 behavior persists in later or earlier builds. Each release can alter what boots, what renders, and how fast it runs.

This is experimental open source PS5 emulation research. The research character explains rapid change and uneven results. It also explains why one video is news but not a buying signal.

The commercial status of the game sets the linking rule used here. Sony and Marvel are named only as rights holders connected to Marvel's Wolverine as a Sony published title. No game files are linked or described as obtainable.

Unreal Gaming PC, Senaxx, Sony, and Marvel are named without hyperlinking. That choice keeps attribution in prose while avoiding download, store, or video paths. The only external link in this article appears in the Sources note for the VideoCardz report.

The build and capture section closes on a simple point. A September 30 build rendered cutscenes, combat and playable sections on the stated hardware. Peaks touched around 20 FPS in lighter scenes while most sections stayed much lower.

16GB And Unclear Why

The overlay reportedly shows KytyPS5 allocating close to the full 16GB VRAM on the RTX 4080. That is an overlay reading from the same test. It is not a confirmed statement of what the game needs.

The cause is unclear. It is not known how much the game itself needs versus emulator side memory behavior. That uncertainty is stated directly because memory readings invite speculation.

High allocation does not prove high requirement. Emulation layers can reserve, mirror, or duplicate resources during translation. Unreleased or experimental paths can hold buffers longer than a native port would.

The 16GB figure also belongs to the test card. The RTX 4080 in this test carries 16GB VRAM. Near full therefore means near the limit of that card in that run.

Readers should not convert the overlay into a specification. This article states no PC requirements. A single allocation reading cannot set a VRAM recommendation.

Memory behavior can also vary by scene. Cutscenes, combat and open playable sections may allocate differently. The report does not break the reading down scene by scene.

Memory behavior can vary by build as well. A September 30 release may manage memory differently from earlier or later work. Build dependence applies to memory as well as speed.

The responsible reading is therefore narrow. In this capture, the overlay showed near full 16GB use. Why it showed that use remains unresolved.

Future testing could separate game need from emulator overhead if detailed allocation data were published. That data is [NEEDS VERIFICATION] and is not present in the current report. Without it, no allocation breakdown is offered here.

Driver and tool effects are also unknown. Overlays can report allocated memory rather than actively used memory. Reserved pools can look full while working sets are smaller.

The article does not resolve which case applies. It records the reading and the uncertainty together. That pairing avoids turning an interesting overlay number into a false requirement.

VRAM pressure may matter more if later builds change allocation strategy. Lower overhead could free headroom on the same card. Higher accuracy could instead raise overhead before speed improves.

Either path is possible in experimental work. Which path KytyPS5 takes on this title is [NEEDS VERIFICATION]. No prediction is made here.

The memory point also reinforces the peak versus sustained theme. A light scene may hit a higher frame rate even while allocation stays high. Frame rate and memory readings do not move in lockstep.

For now, the 16GB observation stands as an overlay reading with unclear cause. It is worth noting because near full VRAM on a 16GB card is unusual to see. It is not worth converting into guidance.

Menu To Gameplay In Days

The progress arc moved quickly from menus to rendered play. In late September, prior VideoCardz coverage described menu level progress via KytyPS5. Days later, first in game rendering appeared at only a few FPS.

The current test shows sustained rendering with brief 20 FPS periods. Cutscenes, combat and playable sections all appear in capture. That is a wider footprint than menu only work.

Speed remains the limiting fact. Sustained rendering does not mean smooth rendering. Most sections run closer to 6 to 9 FPS, with heavier sequences lower.

The table below places the current report in sequence. Standing is included so readers see what each row represents. Playable state is listed explicitly as not reached.

Rapid early movement is common in emulation when a title first becomes renderable. Menus can unlock, then scenes can appear, then more scenes can hold together. Each step can arrive in days.

Later gains are often slower because accuracy, stability, and speed must improve together. Rendering a scene is one hurdle. Holding frame time stable through combat is another.

The current milestone is therefore best read as sustained but slow. More of the game draws than before. Little of it runs at a rate that resembles normal play.

Prior VideoCardz reports anchor the earlier rows. The current VideoCardz report anchors the latest row. That chain keeps the arc tied to published coverage rather than rumor.

The prior speed band deserves repetition. Earlier results near 3 to 7 FPS used different builds and hardware. They cannot be subtracted cleanly from current numbers to claim a fixed speedup.

Fans of Marvel's Wolverine may find the pace encouraging. Sony published context makes the title high profile, and any rendered gameplay draws attention. Encouragement should still stay separate from playability.

Insomniac is not named as a source in the supplied facts beyond the title connection, so no developer claim is attached here. The technical claims all point to the test setup and the emulator build.

The arc also shows why build labels matter. Late September menu behavior, first rendering days later, and a September 30 build with peaks near 20 FPS are three distinct software moments. Collapsing them into one trend line would hide the build changes.

What comes next is not established. Whether later builds raise sustained rates, stabilize heavy scenes, or reduce memory allocation is [NEEDS VERIFICATION]. The article makes no forecast.

The honest summary is short. Menu to gameplay in days is real progress. Playable remains distant.

Limited By Its Own Readme

KyTyPS5 is open source for Windows and Linux with experimental macOS support. That platform note comes from the project description in the supplied facts. It is included so readers understand the research setting.

The project describes compatibility as limited. Limited compatibility means many titles may not boot, may render partly, or may behave oddly. One title rendering does not imply broad support.

The project also warns that behavior changes between builds. That warning has direct relevance to this test. A September 30 release is a snapshot, not a stable target.

Build snapshots can fix one scene and regress another. They can improve speed in a light area while heavy areas stay slow. They can alter memory patterns without changing visible output.

Readers who follow emulation research will recognize this pattern. Early compatibility lists move often. Results are tied to exact revisions and exact settings.

The article therefore names the exact build used. Senaxx's KytyPS5 September 30 release is the build in this capture. Other builds may differ.

Open source status does not imply ease of use. Source availability means the work can be inspected and built by those with the skills to do so. This article provides no download, no build, and no guide.

Platform support also deserves care. Windows and Linux are the stated primary targets, with macOS described as experimental. Experimental support usually means lower expectations and more rough edges.

None of that diminishes the research interest. An open project rendering a modern commercial PS5 title at all is notable. Notable and playable are different categories.

The limited compatibility warning helps set that boundary. Readers should expect uneven behavior across titles and across builds. A single success does not rewrite the compatibility picture.

The build dependent warning helps with a second boundary. Readers should not treat the 20 FPS peak as a durable property of the emulator. It is a property of one build on one PC in selected scenes.

This is experimental open source PS5 emulation research. That sentence is repeated deliberately because it frames every number in the article. Research numbers move, regress, and depend on context.

The game remains a commercial PS5 title. That status is why this article links no emulator download, no build, no guide, no video, and no game files of any kind. Naming KytyPS5 in prose is permitted, linking is not included.

The section takeaway is plain. The project itself says compatibility is limited and behavior changes between builds. This test should be read inside those limits.

What Is Not Established

Many natural questions remain unanswered by a single test video. Settings are not established beyond the named CPU, GPU, and build. Resolution, internal scaling, vsync state, and capture overhead are [NEEDS VERIFICATION].

Stability is not established. The report describes sustained rendering, not crash free completion. Session length, save behavior, and repeatability are [NEEDS VERIFICATION].

Audio is not established in the supplied facts. No audio claim is made here. Visual rendering alone cannot confirm full scene behavior.

Input latency is not established. Frame rate gives only part of the feel of play. Frame pacing and input delay would need separate measurement.

Comparability with prior runs is explicitly non direct. Different builds and hardware make the older 3 to 7 FPS band a rough reference, not a baseline. Any speedup math from those numbers would mislead.

VRAM need is not established. Near full 16GB allocation is an overlay reading. How much the game needs versus emulator side memory behavior is unclear.

Broader hardware scaling is not established. One flagship PC cannot show how slower CPUs or smaller VRAM pools would behave. This article states no PC requirements.

Future progress is not established. Whether peaks broaden into higher sustained rates is [NEEDS VERIFICATION]. Experimental builds can move in either direction.

The table above keeps gaps visible. Each row separates what the report shows from what remains unknown. That separation protects readers from overreading one video.

Attribution remains tight. Every performance figure traces to the Unreal Gaming PC capture via VideoCardz by WhyCry. Every historical row traces to prior VideoCardz coverage.

The PSXTech second video is noted as additional context. It does not supply separate figures in this article. Without verified numbers, it stays a mention rather than a data source.

The article also avoids a requirements box. Requirements imply tested breadth across hardware. One test on a Ryzen 7 9800X3D plus RTX 4080 cannot support that breadth.

Predictions are marked where they appear. Any sentence about what later builds might do is [NEEDS VERIFICATION]. The default position is that later behavior is unknown.

The gap list is a strength, not a weakness. It shows where more testing would help. Matched builds, matched settings, and longer sessions would each add signal.

Until such testing is published, the record is narrow. Around 20 FPS peaks in lighter scenes, 6 to 9 FPS typical, lower in heavy scenes, near full 16GB shown on the overlay, all from one September 30 build test.

FAQ

Did Marvel's Wolverine hit 20 FPS on PC?
Around 20 FPS was reached as a peak in lighter scenes in one KytyPS5 test on a Ryzen 7 9800X3D plus RTX 4080, as reported by VideoCardz by WhyCry. Most sections stayed at 6 to 9 FPS.

Is 20 FPS the normal frame rate in this test?
No. The same capture runs closer to 6 to 9 FPS across most sections, with heavier sequences lower. The 20 FPS figure is a brief light scene peak.

What build was used for the test?
Senaxx's KytyPS5 September 30 release was used in the Unreal Gaming PC capture. Behavior can change between builds, so other releases may differ.

Why did VRAM show near full 16GB?
The overlay showed allocation near the full 16GB on the RTX 4080, but the cause is unclear. It is not established how much the game needs versus emulator side memory behavior.

Is the game playable through emulation now?
No. Sustained rendering with brief peaks is progress, but 6 to 9 FPS typical with lower heavy scenes is still far from playable.

Can older FPS results be compared directly?
No. Prior results near 3 to 7 FPS used different builds and hardware, so they are a non direct comparison. Matched controls would be needed for a valid speedup claim.

Bottom Line

The current state is sustained but slow rendering with brief highlights. Around 20 FPS appears in lighter scenes on flagship hardware. Most play sits at 6 to 9 FPS with heavier scenes lower.

The build context keeps the result bounded. Senaxx's September 30 KytyPS5 release produced this capture on a Ryzen 7 9800X3D plus RTX 4080. Another build or another PC could show different behavior.

The VRAM reading stays an open question. Near full 16GB allocation is reported from the overlay. Whether that reflects game need or emulator behavior is unclear.

The arc from menu to gameplay in days is genuine research movement. Late September menus led to first rendering at a few FPS, then to this sustained capture. Playable state has not been reached.

Readers should keep peak and sustained separate. Peaks describe what light scenes can briefly touch. Sustained rates describe what combat and playable sections actually hold.

Sources: VideoCardz by WhyCry, Marvel's Wolverine now reaches 20 FPS on PCs, published Oct 4, 2026 09:28 GMT, https://videocardz.com/newz/marvels-wolverine-now-reaches-20-fps-on-pcs. Underlying test work is attributed in prose to Unreal Gaming PC using Senaxx's KytyPS5, with a second video noted from PSXTech, and rights context to Sony and Marvel without hyperlinking.