Doom

The pairing of the Intel Core Ultra 5 225 CPU with the NVIDIA GeForce GTX 1660 SUPER is a study in asymmetry: the CPU sits at the 85th percentile among all processors with an average benchmark score of 36,938, while the GPU lands at the 53rd percentile with an average score of 12,986. That gap shows up plainly in the game's measured frames. In Doom, the combination ranks 3,168 out of 3,712 tested combos, placing it ahead of only a small slice of the field. The measured frame data is complete for four presets at three resolutions, and it consistently tells one story: this is a system where the graphics card dictates performance, and the processor is never the limiting factor.

Resolution Scaling

The measured average FPS across resolution transitions is remarkably uniform. Moving from 1920x1080 to 2560x1440, the Low preset drops from 148 to 96 Fps—a 35.1% reduction. Medium goes from 129 to 83, a 35.7% fall; High from 109 to 70, which is 35.8%; Ultra from 89 to 58, a 34.8% dip. This first step, a 77% increase in pixel count, produces a nearly identical loss of about 35% regardless of the graphics preset.

The second step, from 1440p to 4K, is far more punishing. Low plunges from 96 to 52, a 45.8% drop. Medium falls from 83 to 45, also 45.8%. High goes 70 to 38 (45.7%), and Ultra 58 to 31 (46.6%). The step from 1440p to 4K adds about 2.25x as many pixels as the step from 1080p to 1440p, yet the FPS loss is disproportionately large. This is not a linear pixel-to-FPS relationship; it is a sign of the GPU being saturated.

Taking the entire 1080p-to-4K journey quickly: Low drops 148→52 (64.9% loss), Medium 129→45 (65.1%), High 109→38 (65.1%), Ultra 89→31 (65.2%). The loss is roughly uniform across presets, hovering at 65 either way. Consistency across all presets is the strongest tell. If the CPU were the limiting factor, lowering resolution would flatten the curve because the CPU would be able to feed more frames at lower resolution but would plateau. Instead, the FPS die at a steep rate, and that pattern holds for Low and Ultra alike—suggesting the GPU is the primary constraint in every measured scenario. The fact that the Medium preset yields a min/max window that shrinks with resolution (from 110–148 at 1080p, to 71–96 at 1440p, to 38–52 at 4K) also supports a GPU-bound condition: the frame-time consistency degrades as pixel work grows.

FAQ

Q: What is the highest average FPS this combo can produce in Doom?

A: At 1920×1080 with the Low preset, it reaches 148 FPS. That is the highest average recorded in the dataset.

Q: Can this system maintain 60 FPS at 4K?

A: No. The best 4K result is 52 FPS on Low; Medium gives 45, High 38, and Ultra 31. None meet the 60 FPS target.

Q: Which settings deliver the largest single jump in performance?

A: Comparing presets at each resolution, the step from Ultra to High is consistently the biggest relative gain. At 1080p, Ultra yields 89 FPS and High yields 109 FPS, a 22.5% improvement; at 1440p, 58 versus 70, a 20.7% gain; at 4K, 31 vs 38, a 22.6% gain.

Q: How far ahead is the Core Ultra 5 225 over its closest CPU rivals in synthetic benchmarks?

A: The CPU’s average benchmark score of 36,938 is statistically indistinguishable from its top rivals: it trails the AMD Ryzen 5 5500X3D by 0.2% and the Ryzen AI 7 PRO 450 by 0.4%, while sitting 0.5% below the Intel Core i9-12900T.

Q: Where does the GTX 1660 SUPER rank relative to its closest GPU neighbors?

A: With a GPU score of 12,986 and 53rd percentile, it is 0.4% above the RTX 3050 Ti Mobile and 0.4% above the AMD Radeon RX 580, but 0.7% below the Tesla M2090 and 0.9% above the Radeon 740M. In short, it is bracketed by a tight cluster of mid-range parts.

Q: Which resolution shows the least performance difference between presets?

A: At 1440p the spread from Low to Ultra is 107 down to 70, a span of 37% or 58-58 difference, while at 4K the same range is 52 down to 31, a 40% gap. Prefer 1440p if you want more consistent quality scaling.

How This Combo Ranks

In the global list of 3,712 tested combos, this pairing has a rank of 3,168. Only 544 combos rank lower, so this sits within the 14.6% of all combinations as measured against the game's full pool. The CPU’s 85th-percentile standing clearly places it well above the typical machine, but the combination overall sits far down the leaderboard. This indicates that the GPU’s mid-pack class (55th percentile) drives the system to the bottom quadrant of game-specific results. The FPS numbers by themselves are solid for 1080p play, but in the context of every other possible configured match-up, this system is near the lower fifth.

Measured FPS Breakdown

Here is the total set of measured data from the database.

At 1920×1080:

• Low: 148 FPS average.
• Medium: 129 FPS average, with a documented min of 110 FPS and max of 148 FPS.
• High: 109 FPS average.
• Ultra: 89 FPS average.

At 2560×1440:

• Low: 96 FPS.
• Medium: 83 FPS, min 71, max 96.
• High: 70 FPS.
• Ultra: 58 FPS.

At 3840×2160:

• Low: 52 FPS.
• Medium: 45 FPS, min 38, max 52.
• High: 38 FPS.
• Ultra: 31 FPS.

These numbers are all from measured runs, not estimates or inferred values. The FPS deltas between presets are small relative to the resolution deltas; for example, at 1440p, the difference between Medium and High is only 13 FPS on average, while the difference between High at 1440p and High at 4K is 32 FPS. The scale of the resolution effect entirely swamps the preset effect. Another interesting detail: the min/max bounds shown for the Medium preset grow narrower as resolution rises—the spread at 1080p is 38 FPS (110–148), at 1440p it is 25 FPS (71–96), at 4Kp it is 14 FPS (38–52). Simpler pixel workload also leads to more consistent frame times, which points again to the GPU.

Settings Recommendations

Based solely on the measured rows, the sweet spot is at 1440p. The High preset at 2560×1440 reaches 70 FPS on average, which is comfortably above the 60 FPS threshold while still leaving headroom for dips. Medium at the same resolution is 83 FPS, nearly as fast, and even the extra headroom of Low (96 FPS) could be nice for players with 144 Hz or faster displays. Ultra at 1440p at 58 FPS is just below the 60 FPS line, so unless the game is run with adaptive vsync or tolerance for a slight stutter, it is better to step down to High.

At 1080p, every preset exceeds 60 FPS, so the user can freely pick. Ultra gives 89 FPS, which is enough for a premium experience, and High steps up to 109. The margin between presets is significant—going from Ultra to High gains 29% FPS—so if any dips are a concern, High is a safe starting point.

The 4K resolution is not practical for this combo. Even with Low, the average is 52 FPS, and the min/max data for the Medium preset shows the frame rate can be as low as 38. A 60 FPS cap is outside reach at this resolution. If the user wants 4K, they should expect to accept variable frame rates in the 30s-40s range, or consider swapping the GPU for one that ranks substantially higher in this database.

CPU Role

The Intel Core Ultra 5 225 uses 10 cores and 10 threads, seated in the Core Ultra 2 series (Arrow Lake). Its base clock is 3.30 GHz, and the boost clock is 4.90 GHz; the chip draws 65 W TDP. The memory bus is dual-channel DDR5 with a total bandwidth of 102.4 GB/s. Synthetic data places it at the 85th percentile, with an average score of 36,938. Its Cinebench R23 multicore result of 25,891 is a solid segment score; the single-core R23 of 3,655 is useful for largely single-threaded engines.

In Doom, the CPU’s influence is not directly measured in the captured FPS, but the pattern is clear. At 1080p Low, the system reaches 148 FPS, which is high enough that a weaker CPU could start to bottleneck. Yet the system does not show any unusual dips during the row—the Medium preset that has an actual min/max shows a floor of 110 FPS. That floor suggests the CPU is feeding 148 FPS without trouble. When resolution is raised to 4K, the FPS falls to 30-50 even on the same preset, a massive depletion. A CPU-limited system would not show such a dramatic FPS drop from 1080p to 4K, because the CPU would not double the frame count at 1080p. Therefore, the strong 10-core CPU is delivering headroom, and the game is clearly limited elsewhere.

GPU Role

The GTX 1660 SUPER uses the TU116 chip on a 12 nm process with 1,408 shading units, boost clock 1785 MHz, and a TDP of 125 W. It has 6 GB of GDDR6 on a 192-bit bus, giving 336.0 GB/s of bandwidth. Its synthetic scores place it at the 53rd percentile, and it sits near the RX 580 and RTX 3050 Ti Mobile in performance. With a memory of only 6 GB, the 4K Ultra preset is likely to be memory-constrained, which explains the low 31 FPS. At 1080p, the card has plenty of memory and bandwidth for its work.

The measured FPS show a GPU that turns in capable results at low resolution, not so much at high pixel density. The 1080p numbers—109 to 148 FPS—are strong for a 1,408-shader part. But once 4K resolution is applied, the card’s pixel fill rate and raw FP32 throughput cannot keep up; the FPS slows to 30-50 as the preset goes from Low to Ultra. The 6GB VRAM is adequate for 1080p even on Ultra, and the mid-50s percentile is honest. The card’s role as the key limiter is unambiguous.