Evolve

The pair tested here pairs a 14th-generation processor with a last-generation mainstream graphics card: the Intel Core i5-14600KF and the NVIDIA GeForce RTX 3060 Ti run Evolve, a 2015 FPS, across a matrix of three resolutions and four presets. The CPU has 14 physical cores (20 threads) with a 3.50 GHz base clock and a 5.30 GHz boost clock, while the GPU uses an 8 GB frame buffer on a 256-bit bus with 448.0 GB/s bandwidth. The data are true measured frames per second, not estimates. In Evolve, a 2015 release, the i5-14600KF sits at the 90th percentile of all measured CPUs in the database and the RTX 3060 Ti sits at the 59th percentile of all GPUs. The game yields a wide range of performance, from 32 FPS at 4K Ultra to 197 FPS at 1080p Low, which makes the resolution and setting tradeoffs the most substantive part of the analysis.

Resolution Scaling

Raising resolution from 1920×1080 to 3840×2160 inflates pixel count by a factor of 4, but the measured frame rate does not fall in a perfect linear ratio. At High settings the averages drop from 120 at 1080p to 80 at 1440p to 46 at 4K. The scale from 1080p to 1440p is about 0.67x the original frame rate (120 to 80), and from 1440p to 4K it is about 0.58x (80 to 46). This is a steeper fall at the upper end, which is the signature of the rendering channel becoming harder than the pixel-count mathematics. At Low settings the same resolution ladder moves from 197 → 132 → 75, a drop of 65% between 1080p and 4K anyway. The Ultra preset is the harshest: 83 → 56 → 32, meaning a 4K Ultra frame is less than half the frame delivered at 1080p Ultra. This pattern places the GPU as the dominant limiter when the game is run at its higher resolutions: the RTX 3060 Ti is designed to feed a 1080p panel at a high frame rate, and the measured numbers show it producing triple-digit averages only when it is fed the 1080p output, as in the Medium preset (153 FPS). At 4K, regardless of the preset, nothing reaches 100 FPS; the top 4K value is the Low preset’s 75, and the near 4K High average of 46 is actually less than half of the 1080p Low output.

The scale highlights where the CPU stops being the deciding factor. At 1440p and above, the gap between the High and Medium presets is narrower than the gap within the same resolution between the Low and Ultra settings. The constraining component is clearly the rendering hardware, not the processor.

FAQ

Q: What is the strongest resolution for this CPU and GPU play with at high detail in Evolve?

A: The highest average FPS measured across the whole matrix is 197 at 1080p Low. For a more direct look, the 1080p High average is 120, and 1440p High drops to 80, while 4K High settles at 46. The data shows that preserving a triple-digit average requires a resolution of 1080p or 1440p; the 4K is only at moderate frame rates.

Q: Why does the GPU appear to be the limit at 1080p as well?

A: The evidence is indirect but consistent with the numbers. At the Low preset, 1080p yields 197 FPS, while at 4K Low it drops to 75, yet the CPU’s score across the test suite is near the top (90th percentile). The framerate across settings declines rapidly without any corresponding drop in CPU processing requirement; the data shows no output near the CPU’s single-thread ceiling. In contrast, the GPU’s percent rank (59th percentile) and its scaling behavior of frame loss point to it as the constraining part when res is increased.

Q: Which settings produce a playable rate―60 to 100 FPS―at 4K?

A: At 4K there is no setting combination that breaks 100; but Medium yields an average of 58 with a recorded minimum of 50, High yields 46, and Low yields 75. Only Low is above 60. Thus, on average 60 is achieved only with Low, and Medium closely misses that threshold by 2 FPS.

Q: How does this i5-14600KF compare with the closest CPU alternatives according to its database rank?

A: The 14600KF has an average score of 49394 places it 90th percentile overall. Its nearest rival, AMD Ryzen 9 7900, scores 49228 (0.3% difference), and the AMD Ryzen 7 PRO 5755G is 49196 (0.4%) behind; the Intel Core Ultra 5 245 is lower by 0.8%, while only the AMD Ryzen AI Max+ 388 is faster by 0.8%. So the CPU sits essentially among a cluster of high-end processors, a nuance that is well reflected in its top decile percentinal.

Q: What does the rank of 1831 out of 3700 mean for this config in this game?

A: That rank, 1831st out of 3700 tested combos, proves it is mid-pack. Because 3700 tested combos are listed, the fraction that were faster is about half. With the GPU only at the 59th percentile and the CPU at the 90th, the GPU is the weak point in the pipeline, so the rank overweight near the middle rather than the top.

GPU Role

The RTX 3060 Ti is an 8GB GDDR6 card with a 256-bit bus, leading to 448.0 GB/s of memory bandwidth. The clock can be boosted to 1665 MHz from a base 1410 MHz. That combination of memory and clock rate is what delivers the 1080p High 120 FPS at 4K Low 75 and 197 at 1080p Ultra low. In the measured results, the GPU’s margin clearly decompresses as resolution goes up: At 1080p, the delta between High and Low is 77 FPS (120 vs 197), and at 4K that margin shrinks to 29 FPS (46 vs 75). That narrowing pattern is an artifact of the GPU hitting its throughput ceiling first when the pixel is large.

Doubling the resolution multiplies the pixel load by 4, and the GPU hardware must feed that load across its 4864 shading units, 152 texture mapping units, and 80 ROPs, as recorded. The 14.6 GB/s bandwidth (448 GB/s) is typical of this class but the measured framerates show that even a modest 8 GB frame buffer is not the limiting factor in these runs, because the Ultra setting at 1080p still produces 83 FPS, a strong number. The card’s power envelope is 200 W, which is not the physical limitation in this game. The data set underscores that the RTX 3060 Ti is a border card for high-rendered output: it comfortably covers 1080p and 1440p, but evolutions become only playable on Low or Medium at 4K.

How This Combo Ranks

In the comparative database, this configuration is assigned rank 1831 out of 3700 tested combos within Evolve. That world position means that about half the other combinations in the dataset deliver higher frame rates in this title. Because the CPU is at percentile 90 and the GPU at percentile 59, the rank is lower than the CPU alone would suggest, a direct proxy that the EVGA in this game is the darkness on the GPU side. A few other facts sharpen this: the GPU average benchmark score for the 3060 Ti is 16129, which is only 19% above the RTX 5700 XT (16014, 0.7% delta) and 0.8% above the GPU’s nearest competitor in average score (RX 9060). But in the actual measured framerates, the combo places in the mid-pack of the comparison – not in the upper 20 percent, which is the CPU’s rank. So the extra power of the CPU does not lift the envelope as much as the GPU enforces a lower ceiling.

No other combo's raw FPS numbers are given here, but the rank relative to the other 3,700 is exactly the requested stat. The conclusion is purely: the system is capable but not class-leading in Evolve as a configuration of the era, with the CPU somewhat underused.

CPU Role

The Core i5-14600KF has 6.5 cores? It is actually a 14 core / 20 thread Raptor Lake refresh, running at 3.50 GHz base and 5.30 GHz boost. The cache is L1 80 KB per core, L2 2 MB per core, and 24 MB shared L3. In the Evolve FPS data, this CPU delivers the measured 197 FPS at 1080p Low; the same CPU at 4K Low delivers 75, a 100% difference. Since the CPU is typically responsible for the upper bound of peak FPS on a per-thread basis, the 197 high result shows that at these conditions the GPU is not the bottleneck.

The spare headroom is visible in the scores: the average benchmark score is 49394, and the nearest rivals are all within 0.8% of the 14600KF’s composite. But in actual game scents, the GPU’s percentile is 59, which is roughly the 50th percentile of all cards, so the system runs in a CPU-light zone. Single-thread performance is high (PassMark single thread 4273) and multi-core is strong (Cinebench R23 multi 32544), but the game may not use more than 8 threads, given the release year. The CPU performance, though, is never an obstacle in any of the 12 test points in Evolve; the claimed bottleneck is measurable, as we found in the resolution scaling.

Measured FPS Breakdown

The exact rows, all measured:

• 3840×2160 – High: 46 avg; Low: 75 avg; Medium: 58 avg (min 50, max 67); Ultra: 32 avg.
• 2560×1440 – High: 80 avg; Low: 132 avg; Medium: 103 avg (min 87, max 118); Ultra: 56 avg.
• 1920×1080 – High: 120 avg; Low: 197 avg; Medium: 153 avg (min 130, max 176); Ultra: 83 avg.

The only preset that includes minimum and maximum values is Medium, at every resolution: a 4K Medium pushes to 50 percent to 67, a 1440p Medium goes 87–118, and a 1080p Medium alternatively 130–176. These min/max sets denote the narrowest range the 3060 Ti showed at this pace, and they demonstrate that the stable frame delivery at 1440p differs by about 31 frames between the worst and best frame. The largest gap is at 1080p, where the Medium spans 46 frames; that is a decent variance for a low-render-load scenario, presumably tied to scene intricacy. In general, dropping each setting tier yields higher figures – for example, going from High to Low yields 46% to 74% boosts.

Settings Recommendations

Given the measured data and the primary bottlenecks, the GPU is probably best served by the following choices. At 1080p, the High preset gives 120 average, which is significantly above the low-resolution floor, and Medium yields 153, a 27% increase but with a slightly lower effect. For normally a player using a 144Hz panel, Medium will exceed 90% of renders; Ultra, dropping to 83, fails to reach 100 on average. Based on the measured frames, use High at 1080p, as the next-lower setting is only takes 20% extra performance while raising the best or frametime.

At 1440p, the Medium preset at 103 FPS is the only tier (besides Low at 132) that breaks triple digits. High and Ultra at 80 and 56 are both well below the triple-digit marker. Medium is the best of quality that preserves an average above 100 in this steady. The data suggests that Ultra does not justify the 47-frame gap (103→56) from Medium to Ultra.

At 4K, the frame budget is strict: Ultra is unusable at 31 FPS average, and High (46) is a slight drop from Medium (58). The recommended gameplay would be Low (75 average) if competition is a priority, or Medium if you want better imagery and still get a 58 average (with a 50 minimum). Since this benchmark's data is clean, a 50 to 67 range for Medium means that some hard scenes drop to 50, but the visual quality is markedly better than Low. Between the two average values that are above 50 – Medium and Low – Medium is the only one hits near 60, so the median viable preset for 4K is Medium, or Low if the player prefers the highest average (75). The data lowest Ultra is not recommended on this GPU.

For the overall best experience, choosing the preset with the highest quality while maintaining a P50 average above 60, the answer is the Medium at 1440p (103 FPS) and High at 1080p (120 FPS). Those are the two settings that combine above 60 average with acceptable visual fidelity in this lineup.