Overpower

The Intel Core i5-14600K paired with the NVIDIA GeForce RTX 5070 Ti delivers a high-refresh-rate experience across the board in Overpower, but the data reveals a clear scaling hierarchy. At 1080p, the system is heavily CPU-bound at lower settings; the jump from Low to Ultra only costs 131.1 FPS, while the GPU still manages 126.5 FPS at the highest preset. Moving to 1440p, the average FPS drops by roughly 15-20% across all settings, indicating the GPU is beginning to take charge. At 4K, the scaling becomes dramatic: the RTX 5070 Ti loses 50.9% of its 1080p Ultra performance (from 126.5 to 94 FPS) and 59 FPS from its 4K Low result (198.7 to 94 FPS). This pattern shows that while the 14600K’s strong single-core performance (Geekbench single-core score of 2803, Cinebench R23 single-core of 4642) sustains high frame rates at lower resolutions, the 4K Ultra result of 94 FPS is firmly GPU-limited, as the card’s 16 GB of GDDR7 memory and 896.0 GB/s bandwidth are pushed to their limits.

Resolution Scaling

The FPS deltas between resolutions are the most telling metric for bottleneck analysis. From 1080p to 1440p, the average frame rate across all settings drops by 15.1 FPS (High), 24.6 FPS (Low), 14.8 FPS (Medium), and 17.9 FPS (Ultra). These are modest declines, suggesting the i5-14600K can still feed the GPU adequately at 1440p, though the gap widens at the Low preset, where the CPU is the primary limiter.

The real story emerges at 4K. Comparing 1440p to 4K, the average FPS falls by 18.6 (High), 33.6 (Low), 22 (Medium), and 14.6 (Ultra). The Low preset’s 33.6 FPS drop is notable because it shows that even with minimal graphical load, the GPU’s raw throughput (43.94 TFLOPS FP32) becomes the constraint at 4K. The Ultra preset’s smaller absolute drop (14.6 FPS) but lower absolute value (94 FPS) indicates that the game’s heavy effects at this setting are already saturating the GPU at 1440p, leaving less headroom for resolution scaling.

The percentage-based view is clearer: 4K Ultra delivers 74.3% of 1440p Ultra performance, while 4K Low delivers 85.5% of 1440p Low performance. The data suggests that Overpower is not purely GPU-bound at 4K, as the CPU’s 14 cores and 20 threads (with a 5.30 GHz boost clock) still contribute to frame pacing, but the RTX 5070 Ti is the definitive limiting factor at the highest resolution.

FAQ

*Q: What is the best resolution to play Overpower with this hardware?*

A: Based on the measured rows, 1440p Ultra (108.6 FPS) provides a strong balance, but 1080p Ultra (126.5 FPS) is the highest fidelity option that stays above 120 FPS. For pure smoothness, 1440p Medium (145.7 FPS) or 1080p Low (257.6 FPS) are viable, though the latter sacrifices visual quality.

Q: How does the RTX 5070 Ti compare to its nearest rival in this context?

A: The RTX 5070 Ti’s average benchmark score (52086) is nearly identical to the NVIDIA GeForce RTX 4080 (52085), with a deltaPct of 0%. This means the two cards offer statistically similar raw performance, so the measured FPS in Overpower should be representative of both.

Q: Is the CPU or GPU the bottleneck at 4K Ultra?

A: The data indicates the GPU is the primary constraint. The 4K Ultra result (94 FPS) is 32.5 FPS lower than 1440p Ultra (108.6 FPS), and the card’s 87th percentile ranking among all GPUs (vs. the CPU’s 93rd percentile) suggests the GPU is the weaker link in this pairing.

Q: Can the i5-14600K keep up with the RTX 5070 Ti at 1080p?

A: Yes, at 1080p Low the system hits 257.6 FPS, and even at 1080p High it reaches 148.4 FPS. The CPU’s Cinebench R23 multicore score (32881) and PassMark multithread score (38682) show ample headroom, though the 1080p Low result (257.6 FPS) likely approaches the CPU’s maximum frame generation limit.

Q: What is the performance penalty for using Ultra vs. High settings?

A: At 1080p, Ultra (126.5 FPS) is 21.9 FPS slower than High (148.4 FPS). At 1440p, the penalty is 24.7 FPS (108.6 vs. 133.3). At 4K, Ultra (94 FPS) is 20.7 FPS slower than High (114.7 FPS). The relative penalty shrinks at higher resolutions as the GPU becomes more saturated.

Q: How does the system perform at 4K Low compared to 1080p Ultra?

A: 4K Low (198.7 FPS) is 72.2 FPS faster than 1080p Ultra (126.5 FPS). This demonstrates that the GPU has substantial raw power (16 GB GDDR7, 896.0 GB/s bandwidth) but that Ultra settings introduce effects that disproportionately tax the rendering pipeline.

Settings Recommendations

The measured data points to High as the optimal preset for most users. At 1080p, High (148.4 FPS) offers a 21.9 FPS improvement over Ultra (126.5) while retaining most visual fidelity. At 1440p, High (133.3 FPS) is still well above 120 FPS and only 24.7 FPS behind Ultra (108.6), which falls below the high-refresh threshold. The pattern holds at 4K: High (114.7 FPS) is the only preset besides Low that stays above 100 FPS, making it the practical choice for 4K gaming on a 120Hz display.

For competitive play, Low at 1080p (257.6 FPS) or 1440p (232.3 FPS) maximizes frame rate, but the jump from Low to Medium (160.5 FPS at 1080p, 145.7 at 1440p) is substantial — a 97.1 FPS and 86.6 FPS penalty respectively — suggesting that Low removes significant particle or shadow effects. Medium offers a middle ground for 4K (123.7 FPS), which is 9 FPS faster than High (114.7) and keeps the experience fluid. Ultra is only recommended for 1080p (126.5 FPS) where the GPU can still maintain a triple-digit frame rate; at 1440p (108.6 FPS) and 4K (94 FPS), the visual gains are not worth the frame rate loss.

Measured FPS Breakdown

At 1920x1080, the system delivers 257.6 FPS on Low, 160.5 FPS on Medium, 148.4 FPS on High, and 126.5 FPS on Ultra. The largest single-step drop is from Low to Medium (97.1 FPS), indicating that Medium introduces significant GPU workload. High to Ultra is a modest 21.9 FPS drop, showing diminishing returns at the top end.

At 2560x1440, the averages are 232.3 (Low), 145.7 (Medium), 133.3 (High), and 108.6 (Ultra). The Low to Medium penalty is 86.6 FPS, slightly less than at 1080p, as the GPU becomes more of a factor. The High to Ultra drop is 24.7 FPS, similar to 1080p, suggesting these two settings have consistent relative costs.

At 3840x2160, the results are 198.7 (Low), 123.7 (Medium), 114.7 (High), and 94 (Ultra). The Low to Medium drop (75 FPS) is smaller than at lower resolutions, confirming the GPU is nearing its limits. The High to Ultra penalty (20.7 FPS) is the smallest of any resolution, and the 94 FPS Ultra result is the only sub-100 FPS average in the entire dataset.

GPU Role

The RTX 5070 Ti’s specifications directly explain its scaling behavior in Overpower. With 16 GB of GDDR7 memory on a 256-bit bus (896.0 GB/s bandwidth), the card has ample capacity for 4K textures, but the 94 FPS Ultra result indicates that memory bandwidth is not the limiting factor — compute is. The GPU’s 8960 shading units and 280 TMUs deliver 43.94 TFLOPS FP32, yet the 70 RT cores and 280 tensor cores suggest that Overpower’s Ultra preset may invoke ray-traced or AI-enhanced effects that these units must process.

The card’s clocks (2295 MHz base, 2452 MHz boost) are moderate for the 50-series, and its 300 W TDP is managed by a dual-slot cooler with a 16-pin connector. The measured results show that the GPU’s PassMark G3D score (32974) and 3DMark Steel Nomad DX12 score (6604) place it in the 87th percentile, which aligns with its 4K performance: strong but not class-leading. The 0% deltaPct versus the RTX 4080 confirms that this card is a direct competitor, and the FPS data reflects that parity — neither card would substantially outperform the other in Overpower. The GPU’s 16 GB VRAM is more than sufficient for the game’s 4K Ultra settings, as the 94 FPS result is compute-bound rather than memory-capacity-bound.