Counter-Strike: Global Offensive

Counter-Strike: Global Offensive is an older title, but this specific pairing of the Intel Core i5-13400F and the NVIDIA GeForce RTX 5090 produces wildly different experiences depending on resolution and settings. The data shows a system that is almost absurdly overqualified for 1080p but which still hits a wall at 4K Ultra. The measured frames per second reveal a game that is far more sensitive to settings changes than to resolution scaling, and the bottleneck shifts dramatically as you move up the quality presets.

Resolution Scaling

The FPS drop from 1080p to 4K is steep but not uniform across settings, and the pattern tells you exactly what is limiting performance. At Low settings, the average frame rate falls from 436 FPS at 1920x1080 to 311 FPS at 2560x1440, then to 164 FPS at 3840x2160. That is a 62% reduction from 1080p to 4K, which is a massive hit, but the absolute numbers remain so high that the experience is still fluid. At Medium settings, the drop is similar in percentage: from 345 FPS at 1080p to 219 FPS at 1440p, then to 116 FPS at 4K. The 4K Medium figure represents a 66% loss from the 1080p Medium result, showing that the GPU is being pushed harder as pixel count rises.

The High preset tells a slightly different story. At 1080p High, the system produces 267 FPS. Moving to 1440p High drops it to 170 FPS, and 4K High lands at 90 FPS. The scaling here is more linear in terms of pixel count, but the key observation is that the gap between 1440p and 4K is 80 FPS, which is larger than the 97 FPS gap between 1080p and 1440p. This suggests the GPU is starting to feel the load at 4K High, but it is still nowhere near its limits at lower resolutions. Ultra settings are where the wall appears: 176 FPS at 1080p, 112 FPS at 1440p, and just 59 FPS at 4K. The 4K Ultra result is the only measured point that falls below 60 FPS, indicating that this is the first configuration where the combination of high pixel count and heavy shading overwhelms the hardware.

What the resolution data makes clear is that the RTX 5090 is not the limiting factor at 1080p or even 1440p. At Low settings, the jump from 436 FPS to 311 FPS when going from 1080p to 1440p is a 29% drop, which is less than the 47% drop from 1440p to 4K. This implies that at lower resolutions, the CPU is holding the frame rate back, and only when the GPU workload increases enough at 4K does the graphics card become the primary constraint. The i5-13400F, with its 10 cores and 16 threads, is strong enough to push high frame rates in this older engine, but even it has a ceiling. The data shows that the most sensible way to play at 4K is to avoid Ultra entirely, as the difference between High and Ultra at that resolution is a 31 FPS loss, which is a 34% reduction for visual gains that are often marginal in a competitive shooter.

Settings Recommendations

The measured rows provide a clear hierarchy of performance, and the best preset depends entirely on your target refresh rate and resolution. At 1920x1080, every setting is playable, but Low is the only one that breaks 400 FPS, hitting 436 FPS. Medium is not far behind at 345 FPS, and High drops to 267 FPS. For players with a 240Hz or 360Hz monitor, Low is the only setting that guarantees you stay above the refresh ceiling in most scenes, as the 436 FPS average gives you headroom. High and Ultra at 1080p are still fast at 267 FPS and 176 FPS respectively, so they are viable for 144Hz displays, but they leave performance on the table for a game where frame rate often trumps fidelity.

At 2560x1440, Low produces 311 FPS, which is still excellent for high-refresh gaming. Medium drops to 219 FPS, and High to 170 FPS. The data indicates that Medium is the sweet spot for 1440p if you want a balance between visual quality and speed, because it maintains an average above 200 FPS while providing better shadows and textures than Low. High is also acceptable for 144Hz panels, but the 49 FPS loss from Medium to High is significant. Ultra at 1440p yields 112 FPS, which is playable but not ideal for a competitive FPS where higher frame rates reduce input latency.

At 3840x2160, the choices narrow considerably. Low is the only preset that stays above 150 FPS, hitting 164 FPS. Medium drops to 116 FPS, which is still smooth on a 120Hz display, and the measured minimum of 98 FPS with a maximum of 133 FPS shows some variance but no catastrophic dips. High at 4K produces 90 FPS, which is borderline for 90Hz or 100Hz displays but will feel less responsive on a 120Hz panel. Ultra at 4K is a non-starter at 59 FPS, as it falls below the 60 FPS threshold that most players consider the minimum for smooth gameplay. The recommendation from the data is clear: for 4K, use Low if you prioritize frame rate above all else, or Medium if you want a decent visual upgrade while staying above 100 FPS. High is only for players who value image quality over competitive edge, and Ultra should be avoided entirely.

GPU Role

The NVIDIA GeForce RTX 5090 is a 32 GB GDDR7 monster with a 512-bit memory bus and 1.79 TB/s of bandwidth, but in Counter-Strike: Global Offensive, those specifications are largely wasted. The GPU’s boost clock of 2407 MHz and base clock of 2017 MHz are high, yet the benchmark results show that the card is rarely the bottleneck except at 4K Ultra. The GPU’s own benchmark scores are impressive — a passmark G3D score of 39650 and a percentile rank of 92 among all GPUs — but the game’s performance is dictated more by the CPU at lower resolutions. The fact that 1080p Low produces 436 FPS while 4K Low produces 164 FPS indicates that the GPU is capable of far more than the game demands at 1080p, and it only starts to flex its muscles when the pixel count quadruples.

The VRAM capacity of 32 GB is irrelevant here, as the game is from 2012 and will not come close to using even a fraction of that memory. The memory bandwidth of 1.79 TB/s is also overkill for a game that is not texture-heavy by modern standards. What matters more is the GPU’s raw pixel throughput, and the data shows that the RTX 5090 can handle 4K High at 90 FPS and 4K Medium at 116 FPS. The gap between Medium and High at 4K is 26 FPS, which suggests that the High preset introduces rendering costs that the card cannot fully offset at that resolution. The GPU’s role in this combo is to provide headroom at 1080p and 1440p, where it is never the limiting factor, and to become the primary driver only when you push to 4K with moderate settings. The measured results show that the RTX 5090 is not the component holding back performance in most scenarios — it is the CPU.

How This Combo Ranks

This combination of the Intel Core i5-13400F and the NVIDIA GeForce RTX 5090 ranks 3293 out of 3719 tested combos in this game, placing it in the bottom 11% of all configurations. That is a surprisingly low rank for such high-end hardware. The data indicates that the ranking is not a reflection of poor absolute performance — the measured FPS numbers are high — but rather a consequence of the game’s age and the fact that many other combos, likely with faster CPUs, achieve even higher frame rates. The i5-13400F has a percentile rank of 77 against all CPUs, meaning it is better than 77% of processors, but in a game like CS:GO, the CPU is the critical component, and the 13400F is not at the top of the heap.

The nearest rivals to the CPU in synthetic benchmarks are the AMD Ryzen 9 6900HS with a deltaPct of 0, the Intel Core 5 120 with a deltaPct of -0.3, and the AMD Ryzen 5 5600X3D with a deltaPct of -0.3. These are all close in average benchmark score to the 13400F, which has an average benchmark score of 25292. The fact that the CPU is competitive with these parts but the combo ranks so low suggests that the GPU is not the differentiator here. In a game that is CPU-bound at lower resolutions, the RTX 5090 does not boost the combo’s rank because the frame rate is capped by the processor’s ability to feed the game loop. The rank of 3293 out of 3719 is a clear signal that if you are building for CS:GO specifically, this pairing is not optimal — a faster CPU with a mid-range GPU would likely rank higher because the game does not leverage the RTX 5090’s strengths.

The GPU itself ranks in the 92nd percentile among all GPUs, with an average benchmark score of 79842, and its nearest rivals include the NVIDIA Tesla P100 PCIe 16 GB with a deltaPct of 0.3 and the AMD Radeon RX 6850M XT with a deltaPct of 1.1. These are data-center and mobile parts, respectively, which shows that the RTX 5090 is in a class of its own for synthetic compute. However, in CS:GO, that compute power translates to nothing meaningful because the game’s engine is not designed to use it. The combo rank is the most telling statistic in this dataset: it proves that pairing a top-tier GPU with a mid-range CPU in a CPU-bound game results in a lower ranking than pairing a top-tier CPU with a weaker GPU.

CPU Role

The Intel Core i5-13400F is the true workhorse in this pairing for CS:GO, and its specifications explain the measured results. With 10 cores, 16 threads, a base clock of 2.50 GHz, and a boost clock of 4.60 GHz, the 13400F is a solid mid-range processor, but it is not a flagship. The CPU’s benchmark scores reflect this: a Cinebench R23 multicore score of 22604 and a single-core score of 3191. The single-core performance is what matters most in CS:GO, as the game’s engine is known to rely heavily on one or two threads. The 3DMark single-thread score of 960 and the 2-thread score of 1880 show that the processor has decent single-threaded throughput, but it is not exceptional.

The data shows that at 1080p Low, the CPU is the limiting factor, as the frame rate of 436 FPS is likely near the engine’s practical ceiling for this processor. The fact that 1440p Low produces 311 FPS, which is a 29% drop, suggests that the CPU is still able to feed the GPU at that resolution, but the GPU is starting to take on more work. The CPU’s 20 MB of shared L3 cache and 1.25 MB per-core L2 cache are adequate, but they do not provide the same advantage as the larger caches on some rival chips. The nearest CPU rival, the AMD Ryzen 5 5600X3D, has a deltaPct of -0.3, meaning it scores nearly identically in synthetic benchmarks, but the 3D V-Cache on that chip typically gives it an edge in games. The 13400F does not have that advantage, which is why the combo rank is so low.

In this game, the CPU’s role is to determine the maximum achievable frame rate, and the data shows that the 13400F caps out around 436 FPS at 1080p Low. The RTX 5090 is never the bottleneck at 1080p, as evidenced by the fact that lowering settings from High to Low increases FPS from 267 to 436. That 169 FPS jump is entirely due to reducing CPU workload, not GPU workload. The CPU’s 65W TDP and 10 nm process node are irrelevant to gaming performance, but its clock speeds are not. The boost clock of 4.60 GHz is high enough for decent frame rates, but it is not in the same league as a flagship CPU that could push past 500 FPS. For CS:GO, the 13400F is a capable but not exceptional partner for the RTX 5090, and the benchmark results make it clear that the CPU is the component that holds this combo back from a higher rank.

FAQ

Q: Why does this combo rank so low (3293 out of 3719) despite having an RTX 5090?

A: The rank is determined by the combination’s performance in this specific game, and CS:GO is CPU-bound at lower resolutions. The i5-13400F, while strong, limits the maximum frame rate, and the RTX 5090’s power is not fully utilized. The data shows that the CPU has a percentile rank of 77 among all CPUs, but in a game that relies on single-thread performance, it is not enough to place the combo higher.

Q: Is the RTX 5090’s 32 GB of VRAM useful for this game?

A: No. The game is from 2012 and does not use significant VRAM. The measured FPS at 4K High is 90, and at 4K Ultra is 59, which shows that the GPU is not limited by memory capacity but by rendering workload. The 32 GB GDDR7 with 1.79 TB/s bandwidth is overkill for a game that likely uses less than 4 GB.

Q: What is the best setting for 1440p gaming with this combo?

A: Medium is the most balanced preset at 2560x1440, producing an average of 219 FPS with a minimum of 186 FPS and a maximum of 252 FPS. Low is faster at 311 FPS, but Medium offers a visual upgrade while staying well above 144Hz. High drops to 170 FPS, which is still playable but leaves less headroom for a 165Hz monitor.

Q: Can this combo handle 4K gaming at 120 FPS?

A: Only at Low settings, which produce 164 FPS at 3840x2160. Medium drops to 116 FPS, which is just below 120 FPS, and High falls to 90 FPS. Ultra is not viable at 59 FPS. If you have a 120Hz 4K display, you would need to use Low or accept a frame rate below your refresh rate with Medium.

Q: Why does lowering settings from High to Low at 1080p increase FPS from 267 to 436?

A: Because the CPU is the bottleneck at 1080p. Lowering settings reduces the CPU’s workload, allowing it to feed the GPU faster. The RTX 5090 is not the limiting factor at this resolution, as evidenced by the massive frame rate gain. The 169 FPS increase shows that the i5-13400F is the component that dictates performance at 1080p.

Q: Is the i5-13400F a good match for the RTX 5090 in other games?

A: The data here only covers CS:GO, but the synthetic benchmarks show the CPU is competitive with parts like the AMD Ryzen 9 6900HS and the Intel Core 5 120, with deltaPct values of 0 and -0.3 respectively. In more GPU-bound games, the RTX 5090 would likely be the limiting factor, making the 13400F a better match. However, in CS:GO, the CPU holds the combo back, as shown by the low rank.