Archeage

The Intel Core Ultra 5 135H paired with the AMD Radeon 890M delivers playable frame rates only at the lowest settings and resolution, with the data showing a clear performance ceiling that makes 4K Ultra essentially unplayable. The benchmark results indicate a system heavily limited by its integrated graphics, as the CPU's strong multi-threaded scores cannot compensate for the GPU's modest throughput in Archeage.

Resolution Scaling

The measured FPS data reveals a steep and consistent penalty as resolution increases, which is the primary indicator that the GPU is the bottleneck. At 1080p Low, the system achieves 63 FPS, which drops to 43 FPS at 1440p and further to 24 FPS at 4K. This represents a 32% performance loss from 1080p to 1440p and a 44% loss from 1440p to 4K, demonstrating that the rendering load scales almost linearly with pixel count.

The scaling pattern becomes even more telling when examining the High preset. At 1080p High, the average is 38 FPS; at 1440p High, it falls to 26 FPS; and at 4K High, it collapses to 15 FPS. The percentage drops here — 32% from 1080p to 1440p, then 42% from 1440p to 4K — mirror the Low preset's trajectory almost exactly. This consistency across settings strongly suggests that resolution, not graphical complexity, is the dominant factor limiting performance.

The Ultra preset shows the most severe scaling, with 26 FPS at 1080p, 18 FPS at 1440p, and just 10 FPS at 4K. The jump from 1440p to 4K costs 44% of the already-low frame rate, pushing the experience into single-digit territory that is untenable for any real-time gameplay. The Medium preset follows the same pattern: 57 FPS at 1080p, 38 FPS at 1440p, and 22 FPS at 4K, with the 4K result representing a 42% drop from 1440p.

What these numbers collectively indicate is that the limiting component is unequivocally the GPU. A CPU-bound scenario would show smaller performance gaps between resolutions, particularly if the CPU were insufficient for the game's logic at lower resolutions. Instead, the near-constant percentage losses across all four presets point to a graphics pipeline that is saturated by pixel fill and shading work rather than by processor computation. The 4K Ultra result of 10 FPS, compared to the 1080p Low result of 63 FPS, shows that the system loses 84% of its frame rate when both resolution and settings are maximized, a ratio that can only be explained by GPU exhaustion.

GPU Role

The AMD Radeon 890M is an integrated graphics solution built on the RDNA 3.5 architecture, and its benchmark scores place it in the 45th percentile of all GPUs. Its average benchmark score of 9210 sits nearly exactly at parity with the AMD Radeon Vega 8, which scores 9221, a delta of -0.1%. The data shows the 890M is also within 1% of the NVIDIA GeForce GTX 960 and GTX 465, and it trails the GTX 850M by exactly 1%. This cluster of rivals confirms that the 890M operates in the lower-midrange of historical discrete GPUs, which explains its difficulty with Archeage at higher resolutions.

The GPU's clock behavior provides context for these results. The base clock is 400 MHz, with a boost clock of 2900 MHz, and the memory is system-shared with bandwidth dependent on the platform. The 890M has 1024 shading units, 64 texture mapping units, and 32 render output units, with a pixel rate of 92.80 GPixel/s and a texture rate of 185.6 GTexel/s. The FP32 throughput is 5.939 TFLOPS, with FP16 running at the same rate in a 1:1 ratio. DirectX 12 Ultimate support is present, along with Vulkan 1.4 and OpenGL 4.6, so API compatibility is not the limiting factor.

The VRAM situation is critical. With system-shared memory, the GPU has no dedicated video memory, and the bandwidth is system-dependent. Archeage at 4K Ultra demands substantial memory bandwidth and capacity, and the shared-memory architecture cannot deliver it. The measured 10 FPS at 4K Ultra versus 24 FPS at 4K Low demonstrates that even when texture and shading demands are reduced, the GPU still struggles at 4K, suggesting that raw pixel throughput, not just memory, is a bottleneck. The 890M's 32 ROPs are simply insufficient for 3840x2160 output, regardless of preset.

The GPU's PassMark scores reinforce this interpretation. The G3D score is 8076, while DirectX 11 scores 73 and DirectX 12 scores 37. The DirectX 12 score being roughly half the DirectX 11 score is notable, as Archeage is an older title from 2014 that may not utilize modern APIs efficiently, but the lower DX12 score also indicates the hardware's compute capabilities are modest. The Geekbench OpenCL score of 37254 and Vulkan score of 40808 show the GPU can handle compute workloads, but gaming performance remains constrained by the rasterization pipeline.

CPU Role

The Intel Core Ultra 5 135H presents a stark contrast to the GPU, with benchmark data showing a far more capable processor. Its average benchmark score of 29093 places it in the 81st percentile of all CPUs, and its nearest rivals include the AMD Ryzen 7 5800X at 29123 (-0.1%), the Intel Xeon Phi 7210 at 29245 (-0.5%), the AMD Ryzen 5 5600XT at 28940 (+0.5%), and the Intel Core i5-13500HX at 29270 (-0.6%). The deltas here are all within 0.6%, indicating the 135H performs virtually identically to these desktop-class processors in synthetic benchmarks.

The CPU specs explain this strong showing. It has 14 cores and 18 threads, with a base clock of 3.60 GHz and a boost clock of 4.60 GHz. The cache hierarchy includes 112 KB of L1 per core, 2 MB of L2 per core, and 18 MB of shared L3 cache. The processor is built on Intel's 7 nm Meteor Lake architecture and supports DDR5 memory with dual-channel configuration and 89.6 GB/s bandwidth. Its Cinebench scores are substantial: 1718 in R15 multi-core, 244 in R15 single-core, 7834 in R20 multi-core, 1105 in R20 single-core, 11875.5 in R23 multi-core, and 1700 in R23 single-core.

The PassMark multi-thread score of 22422 and single-thread score of 3521 confirm that the CPU is not the weak link. In Archeage, which is an MMORPG known for CPU-heavy scenarios in crowded areas, the 135H's 14 cores and 18 threads provide ample headroom. The data shows that even at 1080p Low, where the CPU would be most likely to bottleneck due to lower GPU load, the system achieves 63 FPS, which is a respectable result. If the CPU were the limiting factor, we would expect to see frame rates plateau regardless of resolution, but the measured FPS drops consistently with resolution, confirming the GPU is the constraint.

The launch MSRP for the 135H is $342, and it was released on 2023-12-13, making it a relatively recent mobile processor. The fact that this modern CPU with 18 MB of L3 cache cannot push Archeage beyond 63 FPS at 1080p Low highlights how severely the integrated GPU restricts the platform. The CPU's 89.6 GB/s memory bandwidth is shared with the GPU, and this contention may contribute to the performance ceiling, but the CPU's raw compute scores suggest it has the capacity to feed a far more powerful discrete GPU if one were present.

FAQ

Q: What is the best resolution for this system in Archeage?

A: The data shows 1920x1080 is the only resolution where the system achieves playable frame rates. At 1080p, the Low preset hits 63 FPS, and Medium reaches 57 FPS, while 1440p tops out at 43 FPS on Low and 4K maxes at 24 FPS on Low.

Q: Can this system run Archeage at 4K?

A: Technically yes, but not comfortably. The 4K Medium preset averages 22 FPS with a minimum of 19 FPS, and 4K Low averages 24 FPS. The 4K High preset drops to 15 FPS, and 4K Ultra falls to 10 FPS, making those settings unplayable by any reasonable standard.

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

A: The Core Ultra 5 135H scores 29093 on average, which is 0.1% below the AMD Ryzen 7 5800X, 0.5% below the Intel Xeon Phi 7210, 0.5% above the AMD Ryzen 5 5600XT, and 0.6% below the Intel Core i5-13500HX. These are all within a fraction of a percent, indicating near-identical synthetic performance.

Q: How does the GPU compare to its nearest rivals?

A: The Radeon 890M scores 9210 on average, which is 0.1% below the AMD Radeon Vega 8, 0.7% below the NVIDIA GeForce GTX 960, 0.9% below the GTX 465, and 1% below the GTX 850M. The 890M sits in the 45th percentile of all GPUs.

Q: What is the frame time consistency at 1440p Medium?

A: The 1440p Medium preset averages 38 FPS with a minimum of 32 FPS and a maximum of 44 FPS, giving a 12 FPS range. This is the only 1440p preset with min/max data available, and it shows moderate variance.

Q: Is the CPU or GPU the bottleneck in this configuration?

A: The data clearly indicates the GPU is the bottleneck. The CPU scores in the 81st percentile, while the GPU sits in the 45th percentile. Resolution scaling shows consistent percentage drops across presets, which is characteristic of GPU limitation rather than CPU constraint.

Settings Recommendations

Based on the measured FPS data, the optimal experience for this system in Archeage is the 1080p Low preset, which achieves 63 FPS average. This is the only configuration that reaches the 60 FPS threshold, and it does so with a comfortable margin. The 1080p Medium preset is a close second at 57 FPS, and given the modest 6 FPS penalty, it may be preferable for users who prioritize visual quality over the last few frames per second.

The 1080p High preset at 38 FPS is a significant step down, losing 25 FPS compared to Low and 19 FPS compared to Medium. This makes High a poor value in terms of frame rate for the settings increase. The 1080p Ultra preset at 26 FPS is not recommended, as it drops below 30 FPS, which is generally considered the minimum for playable action in an MMO.

Moving to 1440p, the Low preset at 43 FPS is the only option that remains above the 30 FPS playability threshold. The Medium preset at 38 FPS is borderline, and the High preset at 26 FPS falls below acceptable levels. The Ultra preset at 18 FPS is unplayable. At 4K, no preset achieves even 30 FPS; the best result is Low at 24 FPS, which is too low for smooth gameplay.

The best experience per the measured rows is unequivocally 1080p Low, with 1080p Medium as a viable alternative for users who accept a slight frame rate reduction for improved visuals. The data shows that the system's sweet spot is low settings at 1080p, and any attempt to increase resolution or settings beyond this results in diminishing returns that quickly become unplayable.

Measured FPS Breakdown

At 1920x1080, the system delivers its best results. The Low preset averages 63 FPS, which is the highest measured value across all configurations. The Medium preset averages 57 FPS, with a minimum of 48 FPS and a maximum of 65 FPS, showing a 17 FPS variance. The High preset averages 38 FPS, and the Ultra preset averages 26 FPS. The progression from Low to Ultra shows a 37 FPS drop, representing a 59% reduction in frame rate.

At 2560x1440, performance drops noticeably. The Low preset averages 43 FPS, which is 20 FPS below the 1080p Low result. The Medium preset averages 38 FPS, with a minimum of 32 FPS and a maximum of 44 FPS, representing a 12 FPS variance. The High preset averages 26 FPS, and the Ultra preset averages 18 FPS. The drop from 1080p to 1440p at Low settings costs 20 FPS, a 32% reduction.

At 3840x2160, the system struggles across all presets. The Low preset averages 24 FPS, the Medium preset averages 22 FPS with a minimum of 19 FPS and a maximum of 25 FPS, the High preset averages 15 FPS, and the Ultra preset averages 10 FPS. The Medium preset's min/max range of 6 FPS is the tightest of any configuration, indicating relatively stable frame pacing even at this demanding resolution.

Comparing presets across resolutions, the Medium preset shows the most consistent relative performance: 57 FPS at 1080p, 38 FPS at 1440p, and 22 FPS at 4K. The Low preset shows a similar pattern: 63 FPS, 43 FPS, and 24 FPS. The High preset drops from 38 FPS to 26 FPS to 15 FPS, and the Ultra preset goes from 26 FPS to 18 FPS to 10 FPS. The data shows that Low and Medium presets are always within 6 FPS of each other at any given resolution, while High and Ultra are consistently far behind, suggesting the graphical settings beyond Medium impose a steep cost that the integrated GPU cannot absorb.