Thief

The Intel Core i5-14400F paired with the NVIDIA GeForce RTX 5070 delivers a performance profile in Thief that is heavily dictated by the game's age and the resolution targeted. This combination ranks in the 74th percentile of all tested combos, placing it well above the median but not at the absolute top. The data reveals a system that can obliterate 1080p and 1440p, but which begins to show its limits at 4K Ultra settings, where the GPU's 12 GB VRAM and the game's demanding shaders start to matter more than raw CPU throughput.

The RTX 5070 is a modern Blackwell 2.0 architecture GPU with a 2325 MHz base clock and a 2512 MHz boost clock. It features 6144 shading units, 192 TMUs, and 80 ROPs, with a pixel rate of 201.0 GPixel/s and a texture rate of 482.3 GTexel/s. Its 12 GB of GDDR7 memory on a 192-bit bus provides 672.0 GB/s of bandwidth. This hardware is well ahead of the game's 2014 release, and the measured FPS data confirms that the CPU is rarely the bottleneck except at the very lowest settings and resolutions.

Resolution Scaling

The most striking observation from the measured FPS data is how the frame rate scales as resolution increases. At 1080p with Low settings, the combo achieves a massive 253 FPS average. This drops to 162 FPS at 1440p and then to 86 FPS at 4K. This represents a 36% reduction from 1080p to 1440p, and a further 47% reduction from 1440p to 4K. The performance cliff between 1440p and 4K is particularly steep, indicating that the GPU is becoming the primary limiting factor at the higher pixel count.

For High settings, the scaling is similar but less extreme. The system hits 143 FPS at 1080p, 92 FPS at 1440p, and 49 FPS at 4K. The drop from 1080p to 1440p is 35.7%, and from 1440p to 4K is 46.7%. The pattern is consistent: moving to 4K roughly halves the frame rate compared to 1440p, regardless of the setting preset. This suggests that the RTX 5070's fillrate and memory bandwidth are being saturated by the 3840x2160 resolution, a workload that is far removed from the game's original target hardware.

At Medium settings, the FPS values are 189 at 1080p, 121 at 1440p, and 65 at 4K. The scaling from 1080p to 1440p is a 36% drop, and from 1440p to 4K is a 46.3% drop. The consistency of these percentage drops across different settings is a strong indicator that resolution scaling is a pure GPU function here. The CPU, with its 10 cores and 16 threads, is not the constraining factor in these transitions.

The Ultra preset, which is the most demanding, shows a different kind of scaling. The FPS is 105 at 1080p, 67 at 1440p, and 36 at 4K. The drop from 1080p to 1440p is 36.2%, and from 1440p to 4K is 46.3%. While the percentages are similar, the absolute values are much lower. At 4K Ultra, the system is only producing 36 FPS, which is below the threshold for smooth gameplay. This is the only configuration where the combo fails to deliver a playable experience, and it clearly points to the GPU being overwhelmed by the combination of high resolution and high settings.

The data suggests that for a smooth 60 FPS experience, the user should target High settings at 1440p, where the system delivers 92 FPS, or Low settings at 4K, where it delivers 86 FPS. The 1440p High configuration seems to be the sweet spot, offering a balance between visual fidelity and performance. The 4K resolution is only viable at Low settings, which undermines the purpose of running at that resolution.

GPU Role

The NVIDIA GeForce RTX 5070 is the dominant component in this pairing. Its 12 GB of GDDR7 memory, while not massive by modern standards, is sufficient for a 2014 title like Thief at most settings. The 672.0 GB/s of memory bandwidth is a significant asset, as it helps to feed the GPU's 80 ROPs at high resolutions. The data shows that at 4K, the GPU is the limiting factor, and the FPS drops are directly attributable to the render load.

The GPU's clock speeds are moderate for the 50-series, but they are more than adequate for this game. The boost clock of 2512 MHz is what allows the card to achieve the high frame rates seen at 1080p and 1440p. The GPU's FP32 performance of 30.87 TFLOPS is a theoretical figure that has little bearing on a game from 2014, but it does indicate that the card is not lacking in raw compute power. The pixel rate of 201.0 GPixel/s is a more relevant metric, as it directly relates to the 4K performance.

The RTX 5070's nearest rivals in the benchmark database are the AMD Radeon Pro 580, the AMD Radeon Pro WX 7100, and the AMD Radeon Pro 5300. The performance deltas are minimal: 0.1% ahead of the Pro 580, 0.8% ahead of the WX 7100, and 1.2% behind the Pro 5300. These are professional workstation cards, not gaming GPUs, which highlights the fact that the RTX 5070's gaming performance is not being compared to its true peers in this dataset. The card sits in the 82nd percentile of all GPUs, which is a strong showing, but the rivals listed are not indicative of its actual gaming competition.

The 12 GB VRAM is not a bottleneck in this game. At 4K Ultra, the FPS drops to 36, but this is more likely due to the shader complexity and ROP throughput than memory capacity. The game's maximum VRAM usage is not listed in the data, but given the game's age, it is unlikely to exceed 4 GB even at 4K. The GPU's memory type, GDDR7, is a modern standard that offers high bandwidth, but it is not a factor in this game's performance.

The data shows that the RTX 5070 is a capable card for Thief, but its performance is not exceptional when compared to the top tier of the database. Its percentile rank of 82 is good, but it is not in the top 10% of GPUs. The card's performance is consistent with its position in the market, offering strong 1440p performance and acceptable 1080p performance, but struggling at 4K with high settings.

How This Combo Ranks

The combination of the Intel Core i5-14400F and the NVIDIA GeForce RTX 5070 ranks 965 out of 3720 tested combos in Thief. This places it in the 74th percentile, meaning it outperforms roughly three-quarters of all other combinations tested. This is a solid ranking, but it is not indicative of a top-tier system. The rank is a composite measure that takes into account the average FPS across all resolutions and settings, and it suggests that while this combo is fast, it is not the fastest.

The CPU's percentile rank is 83, and the GPU's percentile rank is 82. When combined, the system's rank of 965 out of 3720 is slightly lower than what you might expect from these individual components. This is likely due to the fact that Thief is a CPU-light game, and the i5-14400F's performance is not being fully utilized at lower resolutions. At 1080p Low, the system produces 253 FPS, which is likely close to the CPU's maximum frame output. However, at 4K Ultra, the GPU is the bottleneck, and the CPU's performance is irrelevant.

The combo rank is a useful metric for comparing systems, but it does not tell the whole story. A system with a weaker GPU but a stronger CPU might rank higher in this game because Thief relies on single-threaded performance. The i5-14400F has a boost clock of 4.70 GHz, which is high, but its Cinebench R23 single-core score of 3055 is not exceptional. The system's performance is therefore a balance between the CPU's single-threaded capabilities and the GPU's raw throughput.

The data shows that the combo is well-balanced for 1440p gaming, but it is not optimized for 4K. The 965 rank out of 3720 is a good result, but it is not a top-500 finish. Users looking for the highest possible frame rates in Thief would need a more powerful GPU, such as one with a higher pixel rate and more ROPs. The RTX 5070 is a mid-range card, and its performance in this game reflects that.

Measured FPS Breakdown

The measured FPS data provides a detailed picture of the combo's performance across different resolutions and settings. At 1080p, the system is capable of very high frame rates. With Low settings, it averages 253 FPS, which reflects the GPU's ability to handle the game's low-resolution shaders. Medium settings reduce this to 189 FPS, with a minimum of 161 FPS and a maximum of 218 FPS. High settings produce 143 FPS, and Ultra settings produce 105 FPS. These numbers indicate that at 1080p, the system is never CPU-bound, and the GPU is the sole determinant of performance.

At 1440p, the frame rates are still high, but the GPU begins to work harder. Low settings average 162 FPS, Medium settings average 121 FPS with a minimum of 103 FPS and a maximum of 139 FPS, High settings average 92 FPS, and Ultra settings average 67 FPS. The jump from High to Ultra at 1440p is a 27.2% reduction in FPS, which is a significant cost for the extra visual fidelity. The 67 FPS at Ultra is still playable, but it is close to the 60 FPS threshold.

At 4K, the performance drops significantly. Low settings average 86 FPS, Medium settings average 65 FPS with a minimum of 55 FPS and a maximum of 74 FPS, High settings average 49 FPS, and Ultra settings average 36 FPS. The Medium setting is the only one that provides a consistently playable experience, with a minimum FPS of 55. High and Ultra settings are not recommended for 4K, as they produce frame rates below 60 FPS. The data shows that 4K is only viable at Low settings, where the game still looks good but loses many of its visual enhancements.

The minimum FPS values are only provided for the Medium settings. At 1080p Medium, the minimum is 161 FPS, at 1440p Medium it is 103 FPS, and at 4K Medium it is 55 FPS. These minimums indicate that the game is stable and does not have significant frame time spikes. The lack of minimum FPS data for other settings suggests that the game is more consistent at Medium, which is a common pattern for titles that are not heavily optimized for modern hardware.

FAQ

Q: What is the best resolution and settings combination for a consistent 60 FPS experience?

A: The data shows that 1440p High settings produce an average of 92 FPS, which is well above 60 FPS. For a more demanding visual experience, 1440p Ultra averages 67 FPS, which is also above 60 FPS. At 4K, only Low settings (86 FPS) and Medium settings (65 FPS) exceed 60 FPS on average.

Q: How does the combo perform at 4K Ultra settings?

A: At 3840x2160 with Ultra settings, the system averages 36 FPS. This is below the 60 FPS threshold and is not recommended for smooth gameplay. The minimum FPS for this configuration is not listed, but the average suggests significant frame pacing issues might be present.

Q: Is the CPU or GPU the limiting factor in this game?

A: The data indicates that the GPU is the limiting factor at all resolutions above 1080p. At 1080p, the CPU might be a slight bottleneck at Low settings, where the system reaches 253 FPS. However, the CPU's high boost clock of 4.70 GHz ensures it is not a major constraint.

Q: How does this combo compare to other tested systems?

A: The combo ranks 965 out of 3720 tested combinations, placing it in the 74th percentile. This means it is faster than 74% of all other tested systems, but it is not in the top tier. The CPU ranks in the 83rd percentile, and the GPU ranks in the 82nd percentile.

Q: What is the frame rate difference between High and Ultra settings at 1440p?

A: At 2560x1440, High settings average 92 FPS, while Ultra settings average 67 FPS. This is a 27.2% reduction in average FPS when moving from High to Ultra. The visual difference is present but may not justify the performance cost for all users.

Q: Can this system handle the game at 1080p with a high refresh rate monitor?

A: Yes. At 1920x1080, the system averages 143 FPS on High settings and 105 FPS on Ultra settings. This is sufficient for a 144 Hz monitor, although the 253 FPS at Low settings would be needed for a 240 Hz monitor.

CPU Role

The Intel Core i5-14400F plays a supporting role in this game, but it is a capable one. With 10 cores and 16 threads, based on a hybrid architecture that is not fully detailed in the data, it provides ample processing power. The base clock of 2.50 GHz and boost clock of 4.70 GHz are high enough to avoid bottlenecking the GPU at most settings. The CPU's 20 MB of shared L3 cache is beneficial for game logic, but the data shows that the GPU is the primary driver of FPS.

The i5-14400F's benchmark scores are solid. Its Cinebench R23 multi-core score is 21645, and its single-core score is 3055. These scores indicate a strong mid-range CPU. In Geekbench, it scores 11268 multi-core and 2058 single-core. The PassMark multithread score is 25470, and the single-thread score is 3701. These are all good numbers, but they are not exceptional for a 14th-gen Core i5.

The CPU's nearest rivals in the database include the AMD Ryzen 7 PRO 8840U, the Intel Core i9-11900, the AMD Ryzen 7 6800HS, and the Intel Core i7-12800H. The performance deltas are small: the i5-14400F is 0.1% ahead of the Ryzen 7 PRO 8840U, 0.2% ahead of the Core i9-11900, 0.2% behind the Ryzen 7 6800HS, and 0.5% ahead of the Core i7-12800H. This shows that the CPU is right in the middle of a pack of similar-performance processors.

In Thief, the CPU's role is to feed the GPU with draw calls and process game logic. The game is not CPU-intensive, and the i5-14400F is more than capable of handling it. The data shows that at 1080p Low, the system reaches 253 FPS, which is likely near the CPU's maximum frame output. If this FPS were higher, the CPU would be the bottleneck, but it is not. The CPU's performance is therefore adequate, but not the star of the show.

The CPU's 65 W TDP is low, which is good for cooling and power consumption. It supports DDR4 and DDR5 memory, but the data does not specify which type was used in the test. The CPU's PCIe Gen 5 support is also present, but it is not a factor in this game. The CPU's role is to be a reliable, mid-range processor that does not hold back the RTX 5070, and it fulfills that role admirably.