Evolve

Evolve, the 2015 FPS, sits near the older end of the benchmark database, and pairing it with a modern mobile platform — the AMD Ryzen 9 8945HX and the NVIDIA GeForce RTX 5070 Mobile — produces a dataset that raises interesting questions about where the limits actually lie. The measured rows span three resolutions and four presets, and the spread between them tells a story that is less about raw hardware capability and more about how a decade-old title interacts with a contemporary GPU. The following analysis walks through the scaling data, the component roles, and where this specific combination lands among all tested pairings.

Resolution Scaling

The headline scaling behavior is straightforward: at Low settings the combo moves from 187 FPS at 1920x1080 to 125 FPS at 2560x1440 to 71 FPS at 3840x2160. That is a 33 percent drop from 1080p to 1440p and a further 43 percent drop from 1440p to 4K. At High settings the pattern repeats: 114 FPS, then 76, then 43. Ultra runs 79, 53, and 30 FPS across the three resolutions.

What does that say about the limiting component? A GPU-bound title typically shows FPS falling almost in proportion to pixel count, and 4K is exactly four times the pixel load of 1080p. Here the Low-setting falloff from 187 to 71 is a 62 percent reduction — close to but not quite the theoretical GPU-bound maximum. The fact that 1080p never exceeds 187 FPS, even at the lowest preset, hints at a ceiling that is not purely GPU-related. Whether that ceiling comes from the engine's own frame pacing, driver overhead on an older title, or a CPU-side synchronization limit is not determinable from averages alone, but the flatness at the top end is worth noting.

The Ultra row is the sharpest indicator of GPU limitation. At 1080p Ultra the average is 79 FPS; at 4K Ultra it collapses to 30 FPS — a 62 percent drop that closely tracks the quadrupled pixel count. When the drop approaches the pixel-count ratio, the GPU is the constraint. When the drop is shallower — as at Low, where 1080p to 4K costs 62 percent as well but starts from a much higher base — the question becomes whether the CPU is quietly gating the top of the range. The Medium preset offers the only minimum/maximum data in the set: 47 to 64 FPS at 4K and 123 to 167 FPS at 1080p. That 4K Medium spread is a 1.36x ratio between floor and ceiling, which suggests frame delivery is reasonably consistent rather than wildly erratic.

FAQ

Q: Can this setup run Evolve at 4K?

A: Yes, with caveats. At 3840x2160 the combo averages 55 FPS on Medium with a minimum of 47, and 71 FPS on Low. High drops to 43 FPS and Ultra lands at 30 FPS, so 4K play is viable only if the Medium or Low preset is acceptable.

Q: What is the best frame rate this combo delivers in Evolve?

A: 1920x1080 at Low settings, averaging 187 FPS. The next highest is 1080p Medium at 145 FPS with a maximum recorded frame of 167.

Q: How does the RTX 5070 Mobile compare to rival GPUs in the database?

A: Its average benchmark score is 29928, within 0.1 percent of the GeForce RTX 3070 Ti and 0.5 percent ahead of the RTX 2080 Ti. It trails the Radeon RX 6800 by 0.6 percent and the RX 6700 by 1.7 percent. It sits in the 75th percentile against all GPUs.

Q: Is the Ryzen 9 8945HX strong enough for this game?

A: The data indicates a large surplus. The CPU sits in the 95th percentile against all CPUs, with a Passmark multithread score of 51405 and single-thread score of 3907 — comfortably beyond anything a 2015 title should demand.

Q: Where does this CPU/GPU combination rank for Evolve specifically?

A: Rank 2049 out of 3700 tested combinations, which places it in the top 56 percent of pairings recorded for this game.

Q: Does VRAM become a concern?

A: The RTX 5070 Mobile has 8 GB of GDDR7 on a 128-bit bus with 384.0 GB/s of bandwidth. For a 2015 title, the measured 4K rows show steady scaling rather than a cliff, so there is no evidence in this data of a VRAM wall.

GPU Role

The RTX 5070 Mobile is a Blackwell 2.0 part built on TSMC's 5 nm process, with 4608 shading units, 144 TMUs, 48 ROPs, 36 RT cores, and 144 tensor cores on the GB206 chip. Its boost clock is 1425 MHz from a 907 MHz base, and its FP32 throughput is 13.13 TFLOPS. The memory subsystem runs GDDR7 at 1500 MHz — 24 Gbps effective — across a 128-bit interface.

Against the wider GPU field, this card lands in the 75th percentile, and its rival cluster is telling: the RTX 3070 Ti, RTX 2080 Ti, RX 6800, and RX 6700 are all within roughly two percent of its average score. In other words, this is a card whose database performance profile resembles desktop hardware from a couple of GPU generations ago, compressed into a 50 W mobile envelope. That context makes the Evolve results easier to read: an older, modestly demanding FPS title is precisely the kind of workload where such a card should post comfortable numbers at mainstream resolutions — and it does, with 145 FPS at 1080p Medium and 97 FPS at 1440p Medium.

The 8 GB VRAM allocation deserves a closer look. Evolve shipped in 2015, when 4K gaming at High presets was a far more exotic proposition. The measured data shows 4K Ultra at 30 FPS — playable-adjacent but not comfortable — while 4K High at 43 FPS and 4K Medium at 55 FPS hold up better. If VRAM capacity were the binding constraint at 4K, one would typically expect High and Medium to degrade together; instead the smooth stair-step from 71 to 55 to 43 to 30 suggests the limit is compute and fill related rather than capacity related. The 68.40 GPixel/s pixel rate and 48 ROPs are the more plausible bottlenecks when the pixel count quadruples.

The GPU's API support — DirectX 12 Ultimate (12_2), Vulkan 1.4, OpenGL 4.6 — fully covers anything a 2015 title requires, so there is no compatibility angle to consider in interpreting these scores.

Measured FPS Breakdown

At 1920x1080: Low averages 187 FPS, Medium 145 FPS (minimum 123, maximum 167), High 114 FPS, and Ultra 79 FPS. The Medium min/max spread of 123–167 means even the slowest recorded frames stayed above 120, which is the strongest consistency evidence in the dataset.

At 2560x1440: Low averages 125 FPS, Medium 97 FPS (minimum 83, maximum 112), High 76 FPS, and Ultra 53 FPS. The Medium floor of 83 is notable — it remains above the 4K High average, and the 83-to-112 spread indicates the frame delivery at this resolution and preset is tight enough for a fast-paced shooter.

At 3840x2160: Low averages 71 FPS, Medium 55 FPS (minimum 47, maximum 64), High 43 FPS, and Ultra 30 FPS. Only the Medium row carries min/max data at 4K; the 47 FPS minimum sits close enough to the 55 FPS average that dips, when they occur, are moderate rather than severe.

Preset-to-preset cost is remarkably consistent. From Low to Ultra at 1080p, FPS falls from 187 to 79 — a 58 percent reduction for four quality tiers. At 1440p the same journey runs 125 to 53, also 58 percent. At 4K it is 71 to 30, a 58 percent reduction yet again. That identical proportional cost at every resolution is an interesting signature: it implies the presets scale almost purely with pixel work, with no single setting — such as a volumetric or shadow effect — disproportionately punishing any one resolution.

How This Combo Ranks

Among the 3700 CPU/GPU combinations tested in Evolve, this pairing ranks 2049th. That is a middling placement — roughly the top 56 percent — and at first glance it seems low for hardware this modern. The resolution is that the ranking pool presumably includes desktop flagships and newer GPUs whose advantage in an older, lighter title is proportionally smaller, while the RTX 5070 Mobile is a 50 W part competing in a field that includes much higher-power hardware.

Still, the rank invites questions. The CPU side is emphatically not the drag: the Ryzen 9 8945HX ranks in the 95th percentile of all CPUs, with an average benchmark score of 76212 that edges the Intel Core Ultra 9 285HX by 0.1 percent and even the Ryzen 9 9950X3D by 0.6 percent, while trailing the EPYC Embedded 8224P and Threadripper PRO 9945WX by 0.4 percent each. A processor trading blows with workstation-class silicon is not the reason this combo sits mid-pack. The GPU, at the 75th percentile and 1.7 percent behind the RX 6700, is the more plausible anchor on the rank. The combination of a top-tier mobile CPU with an upper-quartile mobile GPU produces, in this specific title, results that are good but not dominant — 4K High at 43 FPS is exactly the kind of number that caps a ranking even when 1440p results are excellent.

CPU Role

The Ryzen 9 8945HX is a Dragon Range part on the Zen 4 architecture, fabricated at 5 nm, with 16 cores and 32 threads, a 2.50 GHz base clock, and a 5.40 GHz boost clock. It carries 64 MB of L3 cache plus 1 MB of L2 per core, supports dual-channel DDR5 with 83.2 GB/s of bandwidth, and exposes 28 PCIe Gen 5 lanes from the CPU. Its 55 W TDP and Socket FL1 packaging mark it as a mobile flagship.

Its benchmark credentials are extensive: Cinebench R23 multi-core at 42713 and single-core at 6030; R20 at 17939 and 2532; R15 at 4305 and 607. Passmark multithread reaches 51405, single-thread 3907, integer math 195180, floating point 117453, physics 2168, and data compression 677755. For a game released in 2015, this is an enormous surplus of capability.

What does the Evolve data say about CPU involvement? The tell is the 1080p ceiling. At Low settings the average stops at 187 FPS despite the GPU's Low-preset workload being trivially light at that resolution. If the GPU were the sole limit, one might expect a higher 1080p ceiling before the resolution scaling kicked in. A 187 FPS cap on a processor with this pedigree suggests either engine-side limits or a frame-rate governor common to the title itself — a frequent characteristic of games from that era. The single-core strength (6030 in R23, 3907 in Passmark single-thread) means that even if Evolve leans on one strong thread, as older shooters often do, this CPU can feed it. There is no evidence in the measured rows of the CPU dragging any configuration below GPU-driven expectations.

It is also notably the multiplier is unlocked, and the integrated Radeon 610M graphics exist, though neither factor appears in the measured data for this game.

Settings Recommendations

For 1920x1080 play, High is the clear pick: 114 FPS average from a 2015 shooter is far beyond smooth, and stepping to Ultra costs 35 FPS for visual gains the database cannot quantify. Medium at 145 FPS with a 123 FPS floor is the choice for high-refresh displays.

At 2560x1440, Medium again makes the strongest case — 97 FPS average, 83 minimum, 112 maximum. High at 76 FPS is still comfortable for a competitive FPS, but the Medium row's minimum of 83 exceeds the High average, which for frame-time consistency is the more valuable property. Ultra at 53 FPS is achievable but hard to justify against those alternatives.

At 3840x2160 the decision tightens. Medium's 55 FPS average with a 47 FPS minimum is the balanced recommendation; Low's 71 FPS suits competitive priorities; High's 43 FPS sits below the threshold where a fast shooter feels fully fluid; Ultra's 30 FPS should be treated as a demonstration setting rather than a play setting. Given the identical 58 percent Low-to-Ultra cost curve at every resolution, dropping one preset tier buys a consistent proportional gain — roughly a fifth to a quarter more frames per tier — making tier selection the single most effective lever in this title on this hardware.