Microsoft Flight Simulator

Microsoft Flight Simulator is one of the most demanding titles in the benchmark database, and the combination of the Intel Core Ultra 5 135H with the AMD Radeon 890M integrated graphics produces results that are firmly in the playable-but-compromised territory. The data shows a system that is heavily constrained by its graphics solution, yet the CPU provides enough headroom to avoid being the primary bottleneck at lower settings. This analysis breaks down the measured performance across resolutions and presets, offering a practical look at what this mobile IGP setup can realistically deliver for virtual pilots.

FAQ

Q: Can this combo run Microsoft Flight Simulator at 1920x1080?

A: Yes, but only at Low settings, where the measured average FPS is 50. Medium settings drop the average to 39 FPS, High to 29 FPS, and Ultra to 28 FPS, so 1080p is viable for smooth flight only on the lowest preset.

Q: What is the best resolution for this CPU/GPU pairing?

A: At 2560x1440, the combo achieves a 40 FPS average on Low settings, which is the highest average at any resolution above 1080p. At 3840x2160, even Low settings only produce a 30 FPS average, making it the least viable option.

Q: How does the Radeon 890M compare to its closest rivals in synthetic benchmarks?

A: The Radeon 890M's average benchmark score is 9210, placing it within 0.1% of the AMD Radeon Vega 8 (score 9221) and 0.7% behind the NVIDIA GeForce GTX 960 (score 9273). The data indicates it is statistically tied with these older discrete and integrated solutions.

Q: Is the Intel Core Ultra 5 135H a strong processor relative to other CPUs?

A: The CPU holds an 81st percentile ranking among all tested processors, with an average benchmark score of 29093. This puts it within 0.1% of the AMD Ryzen 7 5800X (score 29123) and 0.6% behind the Intel Core i5-13500HX (score 29270).

Q: What kind of frame rate variance can be expected at Medium settings?

A: At 1920x1080 Medium, the measured range is 33 to 44 FPS. At 2560x1440 Medium, the range narrows to 26 to 36 FPS, and at 3840x2160 Medium, it drops to 20 to 26 FPS, showing increasing instability as resolution rises.

Q: Does the 890M have dedicated VRAM for this game?

A: No, the Radeon 890M uses System Shared memory for both its frame buffer and bandwidth, with the latter being "System Dependent." This means performance is tied to the system's main memory configuration.

GPU Role

The AMD Radeon 890M is the clear limiting factor in this pairing. Based on the RDNA 3.5 architecture and built on a 4 nm process, this integrated GPU operates with a base clock of 400 MHz and a boost clock of 2900 MHz. It features 1024 shading units, 64 TMUs, and 32 ROPs, which are modest numbers for a flight simulator that renders vast landscapes and complex cockpit instrumentation. The GPU's percentile ranking against all GPUs is 45, meaning it sits just below the median, and its average benchmark score of 9210 is nearly identical to the AMD Radeon Vega 8 (deltaPct -0.1%).

The memory situation is critical. The 890M has no dedicated VRAM; instead, it relies on System Shared memory, with bandwidth listed as "System Dependent." For a game like Microsoft Flight Simulator, which streams terrain and aircraft data continuously, this shared memory arrangement directly impacts performance. The benchmark results confirm this bottleneck: at 3840x2160, the GPU can only manage a 17 FPS average on High settings and a 16 FPS average on Ultra. The difference between Low and High at 4K is a 13 FPS swing (30 vs 17), which suggests that the GPU's rendering pipeline, not the memory bandwidth, is saturated by the increased detail.

At lower resolutions, the GPU's relative weakness becomes less pronounced but still dominates. At 1920x1080, the Low preset yields a 50 FPS average, which is the only preset at any resolution to cross the 60 FPS threshold—and it does not actually reach it. The 890M's FP32 throughput of 5.939 TFLOPS is not sufficient for high-detail flight simulation, and the data shows that users must aggressively reduce graphical fidelity to achieve playable frame rates. The pixel rate of 92.80 GPixel/s and texture rate of 185.6 GTexel/s are adequate for 1080p Low, but they are overwhelmed by the draw calls and texture loads of higher presets. The GPU's compute score of 4157 in Passmark further indicates limited headroom for the physics and AI calculations that Flight Simulator offloads to the graphics card.

CPU Role

The Intel Core Ultra 5 135H provides a much stronger foundation than the GPU. This mobile processor, part of the Core Ultra Series 1 (Meteor Lake architecture), features 14 cores and 18 threads, with a base clock of 3.60 GHz and a boost clock of 4.60 GHz. Its thermal design power is 28 W, which is typical for a mobile part, but its performance is notable: the CPU holds an 81st percentile ranking among all processors, with an average benchmark score of 29093. This places it within 0.1% of the desktop AMD Ryzen 7 5800X (score 29123) and only 0.6% behind the Intel Core i5-13500HX (score 29270), indicating that the 135H is a capable workhorse.

In multithreaded workloads, the CPU scores 11875.5 in Cinebench R23 multicore and 7834 in Cinebench R20 multicore, which are strong numbers for a 28 W part. Its single-core performance is also respectable, with a Cinebench R23 single-core score of 1700 and a Passmark single-thread score of 3521. For Microsoft Flight Simulator, which relies heavily on single-thread performance for the main simulation loop, this means the CPU is not the primary bottleneck in most scenarios. The L3 cache is 18 MB shared, and the memory bandwidth is 89.6 GB/s over a dual-channel bus, which helps feed the integrated GPU's shared memory requests.

The data shows that the CPU has enough headroom to push higher frame rates than the GPU can deliver. For instance, at 1920x1080 Low, the combo hits 50 FPS, but at 3840x2160 Low, it drops to 30 FPS—a 40% reduction that is almost entirely attributable to the GPU's fill rate limitations. The CPU's Passmark multithread score of 22422 and data compression score of 251651 suggest it can handle the game's background threading, such as scenery loading and AI traffic, without stalling. However, the 135H is not a top-tier chip; its nearest rival, the AMD Ryzen 5 5600XT, scores 28940 (deltaPct 0.5%), so it is statistically similar to mid-range desktop parts. In this pairing, the CPU is the anchor that keeps minimum frame rates from collapsing, but it cannot compensate for the GPU's lack of dedicated VRAM and raw shading power.

Settings Recommendations

The measured FPS data dictates a clear hierarchy of playable configurations. For a smooth experience, the Low preset is the only viable option across all resolutions. At 1920x1080 Low, the combo achieves a 50 FPS average, which is the highest measured value in the entire dataset. This should be the default choice for any user prioritizing frame rate over visual fidelity. At 2560x1440 Low, the average drops to 40 FPS, which is still playable for a flight sim, though the lower pixel density may be noticeable. At 3840x2160 Low, the average falls to 30 FPS, which is borderline for a genre that benefits from smooth panning; it is acceptable but not ideal.

The Medium preset is a compromise that only makes sense at 1080p. At 1920x1080 Medium, the average is 39 FPS, with a measured range of 33 to 44 FPS. This is acceptable for casual flying, but the frame time variance could cause judder during fast maneuvers or over dense cities. At 2560x1440 Medium, the average drops to 31 FPS (range 26-36), which is marginal, and at 3840x2160 Medium, it falls to 23 FPS (range 20-26), which is not recommended. The High and Ultra presets should be avoided entirely, as they produce averages below 30 FPS at every resolution. At 1080p High, the average is 29 FPS, and at 1080p Ultra, it is 28 FPS—both are effectively unplayable for a simulation that requires precise control inputs.

The practical recommendation is to run the game at 1920x1080 with the Low preset. If the user has a 1440p display, the Low preset at 2560x1440 (40 FPS) is the best balance, but the visual downgrade from 1080p Low may not be worth the resolution increase. For 4K displays, the Low preset at 30 FPS is the only option, and even then, the experience will be suboptimal. There is no measured setting that achieves a 60 FPS average, so users should set expectations accordingly: this combo is for low-fidelity flight, not high-detail sightseeing.

How This Combo Ranks

In the grand scheme of the benchmark database, this specific combination of the Intel Core Ultra 5 135H and AMD Radeon 890M ranks 2829 out of 2996 tested combos for Microsoft Flight Simulator. This places it in the bottom 6% of all systems tested, which is a stark indicator of how demanding this game is on integrated graphics. The rank reflects the GPU's limitations more than the CPU's capabilities; the 135H itself is a solid processor, but the 890M's performance is akin to a low-end discrete GPU from several generations ago.

The GPU's nearest rivals in synthetic benchmarks are the AMD Radeon Vega 8 (deltaPct -0.1%) and the NVIDIA GeForce GTX 960 (deltaPct -0.7%), which are both older parts. This means the 890M is not a significant step forward in raw compute for this game, despite its RDNA 3.5 architecture. The combo's low rank is consistent with the measured FPS data: at 4K Ultra, it produces 16 FPS, which is a slideshow. Even at 1080p Low, the 50 FPS average is barely above the threshold for acceptable playability, and it is well below what discrete GPUs achieve in this title.

The CPU's percentile rank of 81 does not boost the combo's standing because the GPU is the bottleneck. In a game like Microsoft Flight Simulator, which is heavily GPU-bound at high settings, the 135H's strong single-core score (1700 in Cinebench R23) cannot overcome the 890M's fill rate limitations. The data suggests that this combo is best suited for users who already own a laptop with these components and want to fly with minimal settings, rather than for anyone building a system specifically for this game. The rank of 2829 out of 2996 is a warning that this is a last-resort configuration for flight sim enthusiasts.

Measured FPS Breakdown

The measured FPS data provides a complete picture of performance across three resolutions and four settings presets. At 1920x1080, the Low preset delivers the best result with an average of 50 FPS. The Medium preset averages 39 FPS, with a minimum of 33 FPS and a maximum of 44 FPS, showing a 11 FPS variance. The High preset drops to an average of 29 FPS, and the Ultra preset is nearly identical at 28 FPS, indicating that the GPU is saturated beyond Medium. The lack of min/max values for Low, High, and Ultra suggests that the benchmark only tracked average frame rates for those presets, but the Medium range indicates typical frame time fluctuations.

At 2560x1440, the pattern shifts downward. The Low preset averages 40 FPS, which is a 20% reduction from 1080p Low. The Medium preset averages 31 FPS, with a range of 26 to 36 FPS, and the High preset drops to 23 FPS. The Ultra preset averages 22 FPS, making it the least playable setting at this resolution. The 1440p data shows that the GPU's memory bandwidth constraint becomes more pronounced, as the increased pixel count requires more data to be shuffled through the shared memory interface.

At 3840x2160, performance falls below playable thresholds for most presets. The Low preset averages 30 FPS, which is the only 4K setting that approaches acceptability. The Medium preset averages 23 FPS, with a range of 20 to 26 FPS, and the High preset averages 17 FPS. The Ultra preset is the worst measured result in the entire dataset, with an average of 16 FPS. The gap between Low and Ultra at 4K is 14 FPS, which is a 47% reduction in performance, highlighting the GPU's inability to handle the shader complexity and texture loads of higher presets at high resolutions. Across all resolutions, the Low preset is the only one that consistently stays above 30 FPS, and the data unequivocally shows that this combo is best utilized at 1080p with the lowest graphical settings.