Fivem

# FAQ

Q: What is the highest average FPS this combo achieves at 1080p?

A: At 1920x1080 with Low settings, the Intel Core Ultra 5 225 and NVIDIA GeForce RTX 3070 combination reaches an average of 214 FPS. This is the single highest measured result across all tested resolutions and settings.

Q: How does the CPU compare to its nearest rivals in synthetic benchmarks?

A: The Core Ultra 5 225 has an average benchmark score of 36,938, placing it virtually level with the AMD Ryzen 5 5600F (36,945, 0% delta) and slightly ahead of the AMD Ryzen 5 5500X3D (37,018, -0.2%), AMD Ryzen AI 7 PRO 450 (37,093, -0.4%), and Intel Core i9-12900T (37,112, -0.5%). The differences are within a fraction of a percent.

Q: What is the GPU's percentile ranking among all tested GPUs?

A: The RTX 3070 sits in the 61st percentile of all GPUs in the database. Its average benchmark score is 17,208, which places it marginally above the AMD Radeon RX 7600 XT (17,083, +0.7% delta) and slightly below the NVIDIA Tesla K40c (17,468, -1.5% delta).

Q: Which settings preset delivers the largest performance jump at each resolution?

A: Moving from Ultra to Low settings yields the following average FPS increases: at 1080p from 105 to 214 (over 2x), at 1440p from 69 to 141 (over 2x), and at 4K from 38 to 78 (just over 2x). The Medium preset consistently sits roughly midway between Low and Ultra in terms of average FPS.

Q: How does the combo rank among all tested CPU-GPU combinations?

A: This configuration ranks 1,247th out of 3,005 total combos tested in Fivem, placing it in the upper portion of the overall distribution (roughly the 59th percentile of all pairings).

Q: Does the CPU have unlocked multiplier support?

A: No. The Core Ultra 5 225 has a locked multiplier, meaning overclocking via multiplier adjustment is not supported. It runs at a 65W TDP with a base clock of 3.30 GHz and a boost clock of 4.90 GHz.

# CPU Role

The Intel Core Ultra 5 225 is a 10-core, 10-thread desktop processor built on the Arrow Lake architecture using a 3nm process from TSMC. It operates with a base clock of 3.30 GHz and a boost clock of 4.90 GHz, with a 65W TDP. The chip carries 20 MB of shared L3 cache, alongside 192 KB of L1 and 3 MB of L2 per core. The benchmark data shows strong single-thread performance — a Cinebench R23 single-core score of 3,655 and a PassMark single-thread score of 4,412 — which matters for Fivem's server-driven game logic and network-heavy workloads. Multi-threaded throughput is also respectable, with a Cinebench R23 multi-core score of 25,891 and a PassMark multithread score of 30,459.

The CPU's 85th percentile ranking among all processors indicates it sits comfortably above the median. Its nearest synthetic rivals are AMD Ryzen 5 5600F (36,945 avg score), AMD Ryzen 5 5500X3D (37,018), AMD Ryzen AI 7 PRO 450 (37,093), and Intel Core i9-12900T (37,112) — all within a 0.5% band of the Core Ultra 5 225's 36,938 average. This suggests the chip is neither a standout nor a laggard in its class; it is firmly mid-pack among comparable desktop CPUs.

In Fivem specifically, the CPU's influence is most visible at lower resolutions and settings where the GPU is not the primary limiter. At 1080p Low, the combo produces 214 average FPS, a figure that is substantially higher than the 105 FPS seen at 1080p Ultra. The gap between settings at the same resolution — from 105 to 214 FPS — indicates that the processor can feed the GPU well enough to scale with reduced graphical loads. The CPU's 20 MB of shared L3 cache and dual-channel DDR5 memory support with 102.4 GB/s bandwidth provide adequate memory throughput for the game's asset streaming, which is often a bottleneck in multiplayer environments with many players and vehicles.

The fact that this CPU shows a PassMark data compression score of 302,811 and data encryption score of 22,285 suggests it handles the kind of serialized data processing that Fivem's server communication entails. While these synthetic tests do not map directly to game performance, they indicate the processor has sufficient integer and floating-point muscle — PassMark integer math at 65,345 and floating-point math at 92,038 — to manage the game's simulation overhead. The CPU is not unlocked for overclocking, but its stock boost clock of 4.90 GHz provides ample single-thread headroom for the game's primary thread.

# GPU Role

The NVIDIA GeForce RTX 3070 is built on the Ampere architecture using an 8nm Samsung process. It features 5,888 shading units, 184 texture mapping units, and 96 ROPs, along with 46 ray tracing cores and 184 tensor cores. The GPU operates at a base clock of 1500 MHz with a boost clock of 1725 MHz, and its 8 GB of GDDR6 memory runs at 1750 MHz (14 Gbps effective) across a 256-bit bus, delivering 448.0 GB/s of bandwidth. The card's FP32 throughput is 20.31 TFLOPS, and its pixel rate is 165.6 GPixel/s with a texture rate of 317.4 GTexel/s.

The RTX 3070 sits in the 61st percentile of all GPUs, with an average benchmark score of 17,208. Its nearest rivals in the database are the AMD Radeon RX 7600 XT (17,083, +0.7% delta), NVIDIA GeForce GTX 690 (17,037, +1% delta), AMD Radeon HD 7970M (17,019, +1.1% delta), and NVIDIA Tesla K40c (17,468, -1.5% delta). The delta percentages show the RTX 3070 is essentially within 1.5% of these comparably-scored GPUs, meaning its synthetic performance class is well-defined.

In Fivem, the GPU's role becomes dominant at higher resolutions and settings. The data shows a clear scaling pattern: at Ultra settings, the combo delivers 105 FPS at 1080p, 69 FPS at 1440p, and 38 FPS at 4K. This roughly 50% drop per resolution step indicates the GPU is the principal limiter in these configurations. The 8 GB VRAM capacity is adequate for Fivem's moderate texture requirements, but the bandwidth of 448.0 GB/s must feed the frame buffer at 4K Ultra, which explains the 38 FPS result — the lowest measured average in the entire dataset.

The GPU's end-of-life production status and 220W TDP suggest it is a previous-generation part, but its benchmark scores and measured FPS values show it remains capable for this game. The 14 Gbps effective memory speed and 256-bit bus provide sufficient bandwidth for 1440p high-refresh gameplay, as evidenced by the 89 FPS average at 1440p High. The RTX 3070's DirectX 12 Ultimate support (12_2) and Vulkan 1.4 compatibility ensure it can handle Fivem's rendering demands without API-level bottlenecks.

# Settings Recommendations

The measured FPS data provides clear guidance on which preset to use at each resolution. At 1920x1080, the Low preset delivers 214 average FPS, which is far beyond what most displays can show. The Medium preset at 169 FPS with a minimum of 144 FPS and maximum of 194 FPS remains excellent, offering a substantial visual upgrade over Low while still maintaining high refresh rates. The High preset at 135 FPS is also viable, while Ultra at 105 FPS drops below the 144 FPS threshold that many high-refresh 1080p monitors target. For 1080p, Medium provides the best balance: it is 60 FPS faster than High and only 45 FPS slower than Low, with the min FPS of 144 ensuring consistent smoothness.

At 2560x1440, the Low preset produces 141 average FPS, which is ideal for competitive play on 144Hz monitors. The Medium preset at 111 FPS with a 94 FPS minimum and 128 FPS maximum remains playable, but the 94 FPS floor is close to the 90 FPS range where frame pacing becomes noticeable. High at 89 FPS is acceptable for slower-paced gameplay, while Ultra at 69 FPS is best avoided for anything that requires responsive aiming. The recommendation here is Low if competitive consistency is the priority, or Medium if visual fidelity matters more and the player can tolerate occasional dips below 100 FPS.

At 3840x2160, the situation is more constrained. Low produces 78 FPS, Medium delivers 62 FPS with a minimum of 52 FPS and maximum of 71 FPS, High yields 49 FPS, and Ultra falls to 38 FPS. Only Low achieves a smooth experience on a standard 60Hz 4K display, and even then the 78 FPS average leaves headroom for drops. Medium is borderline — the 52 FPS minimum suggests potential stutter in heavy scenes. For 4K, Low is the only reliable recommendation; Medium is only acceptable if the player prioritizes visual quality over frame stability. High and Ultra are not recommended at this resolution given the sub-50 FPS averages.

# How This Combo Ranks

The Intel Core Ultra 5 225 paired with the NVIDIA GeForce RTX 3070 ranks 1,247th out of 3,005 tested combos in Fivem. This places the configuration in the upper-middle tier of all CPU-GPU pairings in the database, roughly in the 59th percentile. The ranking reflects the balance between a strong modern CPU (85th percentile among all processors) and a mid-range previous-generation GPU (61st percentile among all GPUs). The combination is not a top-tier pairing, but it is clearly not a low-end setup either.

The combo's standing is notable because the CPU's synthetic benchmark average (36,938) is nearly identical to its nearest rivals — AMD Ryzen 5 5600F at 36,945, AMD Ryzen 5 5500X3D at 37,018, AMD Ryzen AI 7 PRO 450 at 37,093, and Intel Core i9-12900T at 37,112. This means the ranking is unlikely to shift dramatically if the CPU were swapped for a direct competitor with similar synthetic scores. Conversely, the GPU's position relative to its nearest rivals — AMD Radeon RX 7600 XT at 17,083, NVIDIA GeForce GTX 690 at 17,037, AMD Radeon HD 7970M at 17,019, and NVIDIA Tesla K40c at 17,468 — suggests the RTX 3070's performance class is well-established.

The rank of 1,247 out of 3,005 implies there are roughly 1,246 combos that perform better in Fivem and about 1,758 that perform worse. This positions the setup as a solid mid-to-upper-range option for the game, capable of high frame rates at 1080p and playable at 1440p, but not a top performer at 4K. The combination's measured FPS values at 1080p High (135) and 1440p Medium (111) align with what one would expect from the 61st percentile GPU paired with an 85th percentile CPU — the CPU provides enough headroom to avoid bottlenecking the GPU at most settings, while the GPU's limits become apparent at higher resolutions and presets.

# Measured FPS Breakdown

At 1920x1080, the measured results show a wide spread across settings. The Low preset achieves 214 average FPS, which is the highest figure in the entire dataset. Medium produces 169 average FPS with a minimum of 144 and maximum of 194. High delivers 135 average FPS, and Ultra drops to 105 average FPS. The progression from Low to Ultra represents a 109 FPS reduction, or roughly a 51% performance loss, when maxing out all graphical options.

At 2560x1440, the averages step down predictably. Low yields 141 FPS, Medium produces 111 FPS with a minimum of 94 and maximum of 128, High reaches 89 FPS, and Ultra falls to 69 FPS. The delta between Low and Ultra at 1440p is 72 FPS, approximately a 51% decrease — proportionally similar to the 1080p scaling. The minimum FPS figure for Medium at 1440p (94 FPS) is notably higher than the 4K Medium minimum (52 FPS), indicating that resolution scaling affects frame stability as well as averages.

At 3840x2160, the performance envelope tightens considerably. Low produces 78 average FPS, Medium delivers 62 FPS with a minimum of 52 and maximum of 71, High reaches 49 FPS, and Ultra bottoms out at 38 FPS. The gap between Low and Ultra at 4K is 40 FPS, a 51% reduction — the same proportional scaling observed at lower resolutions. This consistency across resolutions suggests that the GPU's rendering load scales evenly with pixel count, and that the CPU does not introduce additional bottlenecks at any resolution.

# Resolution Scaling

The FPS drop from 1080p to 4K reveals the limiting component dynamics. At Low settings, the progression is 214 FPS at 1080p, 141 FPS at 1440p, and 78 FPS at 4K. This represents a 34% reduction from 1080p to 1440p and a 45% reduction from 1440p to 4K (or 64% total from 1080p to 4K). The scaling is steeper than the pixel count increase would suggest — 1440p has 1.78x the pixels of 1080p, yet FPS drops by only 34%, indicating the GPU is not the sole limiter at 1080p Low. The CPU's ability to generate frames at 214 FPS is impressive, but the GPU cannot sustain that rate as pixel count rises.

At Ultra settings, the progression is 105 FPS at 1080p, 69 FPS at 1440p, and 38 FPS at 4K. The reduction from 1080p to 1440p is 34%, and from 1440p to 4K is 45%, identical to the Low setting percentages. This consistency suggests that the RTX 3070's rendering pipeline scales uniformly with resolution regardless of preset, meaning the GPU is the dominant constraint across all settings. If the CPU were a bottleneck at 1080p Ultra, the 1440p Ultra drop would be less severe than the 1080p-to-1440p drop at Low — but the percentages match exactly, pointing to GPU-bound scaling.

The Medium preset's minimum FPS values provide additional insight. At 1080p Medium, the minimum is 144 FPS; at 1440p Medium, it drops to 94 FPS; at 4K Medium, it falls to 52 FPS. The 1440p minimum is 35% lower than 1080p, while the 4K minimum is 45% lower than 1440p. This accelerating degradation at 4K indicates that the GPU's memory bandwidth (448.0 GB/s) and 8 GB VRAM become increasingly taxed as pixel count rises, producing wider frame time variance. The data suggests the CPU is capable of sustaining high frame rates at 1080p and 1440p, but the GPU is the clear limiting factor at 4K, particularly at higher settings where the 38 FPS Ultra result demonstrates the practical ceiling of this combination.