The Technomancer

# CPU Role — cores, clocks, and how they relate to this game's results

The AMD Ryzen 9 9900X3D brings a 12-core, 24-thread configuration built on the Zen 5 architecture (Granite Ridge) to The Technomancer. Its base clock of 4.40 GHz and boost clock of 5.50 GHz provide substantial single-thread headroom, which matters for an Action RPG from 2016 that typically doesn't scale perfectly across many cores. The CPU's 128 MB of L3 cache is a notable feature—this large cache pool can help keep frequently accessed game data close to the cores, potentially reducing memory latency in scenarios where the engine repeatedly queries the same assets.

The processor's benchmark profile shows strong multi-threaded capability: it scores 47,737 in Cinebench R23 multi-core and 6,739 in single-core. The 3DMark thread scaling results are particularly telling—2-thread score of 2,512, 4-thread score of 4,885, 8-thread score of 9,002, and 16-thread score of 12,490, with a max-thread score of 13,795. The relatively modest jump from 16 to max threads (12,490 to 13,795, roughly 10%) suggests that beyond 16 threads, the 9900X3D's additional cores yield diminishing returns in synthetic workloads. For The Technomancer, this implies that the game's engine likely uses a limited number of threads effectively, and the CPU's high boost clock may matter more than its core count.

The 9900X3D's average benchmark score of 54,762 places it in the 91st percentile among all CPUs. Compared to its nearest rivals, it sits nearly even with the Intel Core Ultra 5 245KF (55,093, deltaPct -0.6), meaning the 9900X3D trails by just 0.6%. It edges ahead of the Intel Core Ultra 5 245K (54,053, deltaPct 1.3) and the Intel Xeon 6505P (53,701, deltaPct 2.0), and surpasses the Intel Core i7-14700F (53,620, deltaPct 2.1) by over 2%. These small deltas suggest that in CPU-bound scenarios, the 9900X3D is competitive with but not dramatically superior to these alternatives.

Given the game's release date of June 27, 2016, The Technomancer predates the 9900X3D by nearly a decade. The hardware data shows that at 1080p Low settings, the combo achieves 334 FPS average, which strongly suggests the CPU is not a limiting factor at lower resolutions—the engine's game logic and draw call processing are being handled with plenty of headroom. The single-thread score of 6,739 in Cinebench R23 and 3,041 in Geekbench single-core indicate that even older game engines that rely on one or two primary threads will see excellent performance.

# GPU Role — VRAM, clocks, and how they relate to this game's results

The NVIDIA GeForce RTX 5090 is a Blackwell 2.0 architecture GPU built on TSMC's 5 nm process, packing 92,200 million transistors into a 750 mm² die. Its memory subsystem is massive: 32 GB of GDDR7 on a 512-bit bus, delivering 1.79 TB/s of bandwidth. The base clock runs at 2017 MHz with a boost of 2407 MHz, and memory operates at 1750 MHz (28 Gbps effective). These specifications provide enormous raw throughput, which becomes relevant for high-resolution rendering.

In terms of compute resources, the RTX 5090 has 21,760 shading units, 680 TMUs, 176 ROPs, 170 RT cores, and 680 tensor cores. Its pixel rate of 423.6 GPixel/s and texture rate of 1,636.8 GTexel/s enable it to fill frames at very high rates. The FP32 performance of 104.8 TFLOPS is substantial for a game from 2016 that was not designed with such hardware in mind—this is a case of modern GPU brute force overcoming any lack of optimization in the older title.

The RTX 5090's benchmark scores place it in the 92nd percentile among all GPUs. Its average benchmark score of 79,842 is remarkably close to several rivals: the NVIDIA Tesla P100 PCIe 16 GB (79,605, deltaPct 0.3) and Tesla P100 PCIe 12 GB (79,396, deltaPct 0.6) are nearly identical, while the AMD Radeon RX 6850M XT (78,940, deltaPct 1.1) trails slightly. The AMD Radeon Pro Vega 64X (80,959, deltaPct -1.4) actually scores higher by 1.4%. These tight deltas suggest that in synthetic GPU benchmarks, the RTX 5090 is not dramatically ahead of these older workstation-class parts, though its real-world gaming performance in modern titles would likely differ.

For The Technomancer specifically, the GPU's role becomes clear when examining resolution scaling. At 4K Ultra, the combo drops to 62 FPS average—the lowest measured result. This indicates that even with 32 GB of VRAM and 1.79 TB/s bandwidth, the game's Ultra settings impose a significant load at 4K. The 32 GB capacity is far more than The Technomancer would need, so the bottleneck is likely raw shading and rasterization throughput rather than memory size or bandwidth.

The GPU's 5 nm process node and high transistor count mean it can sustain heavy workloads, and the 575 W TDP gives it the power budget to maintain boost clocks under sustained gaming load. The dual-slot design and 1x 16-pin power connector indicate this is a flagship card built for maximum performance, which the 4K results reflect—79 FPS at High, 93 FPS at Medium, and 121 FPS at Low all demonstrate strong 4K capability in a 2016 title.

# Resolution Scaling — how FPS drops from 1080p to 4K and what it says about the limiting component

Examining the measured FPS across resolutions reveals how the balance between CPU and GPU shifts. At 1080p Low, the combo hits 334 FPS average. At 1440p Low, it drops to 220 FPS—a 34% reduction. At 4K Low, it reaches 121 FPS—another 45% drop from 1440p. This scaling pattern shows that at Low settings, the GPU becomes progressively more limiting as resolution increases, but even at 4K, 121 FPS indicates the system has significant headroom.

The High settings tell a similar story: 1080p High produces 221 FPS, 1440p High yields 144 FPS (a 35% drop), and 4K High falls to 79 FPS (a 45% drop from 1440p). The consistency of these percentage drops (roughly 34-35% from 1080p to 1440p, and 45% from 1440p to 4K) across Low and High settings suggests that the scaling is primarily resolution-driven rather than settings-driven.

Medium settings show a slightly different pattern. At 1080p Medium, the average is 259 FPS with a minimum of 220 and maximum of 298. At 1440p Medium, the average drops to 169 FPS (min 144, max 195), and at 4K Medium, it reaches 93 FPS (min 79, max 107). The minimum FPS values at Medium settings—220 at 1080p, 144 at 1440p, and 79 at 4K—are notably close to the average FPS at High settings for the same resolutions (221, 144, and 79 respectively). This suggests that Medium and High presets may share similar worst-case scenarios, with High settings pushing average performance down to those minimum thresholds.

Ultra settings produce the most dramatic scaling: 1080p Ultra averages 173 FPS, 1440p Ultra drops to 113 FPS (a 35% reduction), and 4K Ultra falls to 62 FPS (a 45% reduction from 1440p). The fact that 4K Ultra is the only preset where the average FPS dips below 100 indicates that this combination of resolution and settings finally saturates the GPU's capabilities.

The data implies that at 1080p, the CPU is the primary limiting factor—the 9900X3D's per-core performance determines how fast frames can be processed, and the RTX 5090 has abundant headroom. As resolution increases to 1440p and 4K, the GPU workload grows, and the RTX 5090 becomes the bottleneck. The consistent ~35% drop from 1080p to 1440p, and ~45% from 1440p to 4K, regardless of settings, points to a resolution-driven scaling model where the GPU's fill rate and shading power dominate at higher pixel counts.

# FAQ

Q: Does the AMD Ryzen 9 9900X3D bottleneck the RTX 5090 in The Technomancer?

A: At 1080p Low (334 FPS) and 1080p High (221 FPS), the CPU's single-thread performance likely limits frame rates, as the GPU has surplus capacity at these lower resolutions. However, at 4K Ultra (62 FPS), the GPU becomes the clear bottleneck—the resolution and settings load overwhelms the RTX 5090's capabilities.

Q: How much VRAM does The Technomancer require, and is 32 GB sufficient?

A: The RTX 5090's 32 GB of GDDR7 memory is far beyond what a 2016 Action RPG would need. The measured FPS data shows no indication of memory-related stuttering or performance cliffs, even at 4K Ultra, so the 32 GB capacity is more than adequate for this game.

Q: What is the best resolution for this combo if I want at least 144 FPS average?

A: At 1440p High, the combo achieves 144 FPS average, which meets the 144 FPS target. At 1080p, all settings from Low (334 FPS) to Ultra (173 FPS) exceed 144 FPS. At 4K, even Low (121 FPS) falls short of 144 FPS.

Q: How does the RTX 5090 compare to its nearest rivals in synthetic benchmarks?

A: The RTX 5090's average benchmark score of 79,842 is nearly tied with the NVIDIA Tesla P100 PCIe 16 GB (79,605, +0.3%) and Tesla P100 PCIe 12 GB (79,396, +0.6%). It trails the AMD Radeon Pro Vega 64X (80,959) by 1.4% but leads the AMD Radeon RX 6850M XT (78,940) by 1.1%.

Q: Does the 9900X3D's large L3 cache affect gaming performance in this title?

A: The 128 MB L3 cache could help with frequently accessed game data, but the measured FPS suggests that at 1080p, the CPU's boost clock of 5.50 GHz is more critical. The cache's impact is difficult to isolate from the data, though the high FPS across all settings at 1080p indicates strong overall CPU performance.

Q: Is there a significant difference between Medium and High settings in this game?

A: At 4K, Medium averages 93 FPS compared to High's 79 FPS—a 14 FPS difference. At 1440p, Medium averages 169 FPS versus High's 144 FPS (25 FPS difference). At 1080p, Medium hits 259 FPS while High reaches 221 FPS (38 FPS difference). The relative gap grows at lower resolutions, suggesting settings impact the GPU more at higher pixel counts.

# Measured FPS Breakdown

1920x1080 (1080p):

• Low: 334 FPS average
• Medium: 259 FPS average (minimum 220, maximum 298)
• High: 221 FPS average
• Ultra: 173 FPS average

2560x1440 (1440p):

• Low: 220 FPS average
• Medium: 169 FPS average (minimum 144, maximum 195)
• High: 144 FPS average
• Ultra: 113 FPS average

3840x2160 (4K):

• Low: 121 FPS average
• Medium: 93 FPS average (minimum 79, maximum 107)
• High: 79 FPS average
• Ultra: 62 FPS average

The pattern is clear: each settings tier costs a predictable amount of performance. At 1080p, moving from Low to Medium costs 75 FPS (334 to 259), Medium to High costs 38 FPS (259 to 221), and High to Ultra costs 48 FPS (221 to 173). At 1440p, the same transitions cost 51 FPS (220 to 169), 25 FPS (169 to 144), and 31 FPS (144 to 113). At 4K, the drops are 28 FPS (121 to 93), 14 FPS (93 to 79), and 17 FPS (79 to 62).

These diminishing absolute differences at higher resolutions reinforce that the GPU is doing more work per frame at 4K, so settings changes have a smaller relative impact. The minimum FPS values recorded at Medium settings—220 at 1080p, 144 at 1440p, and 79 at 4K—are exactly the same as the High settings averages at those resolutions, suggesting the engine has specific performance floors tied to scene complexity.

# Settings Recommendations

Based on the measured FPS data, the optimal settings depend on the target refresh rate and resolution. For players with a 1080p 240 Hz display, the Medium preset (259 FPS average) provides a comfortable margin above 240 Hz while still offering improved visual quality over Low (334 FPS). The High preset (221 FPS) falls just short of 240 Hz, so Medium is the better choice for maximum smoothness without sacrificing too much fidelity.

At 1440p with a 165 Hz or 180 Hz monitor, the Low preset (220 FPS) exceeds the refresh rate, but the Medium preset (169 FPS) delivers a better visual experience while staying above 144 Hz. The High preset (144 FPS) is acceptable for 144 Hz displays, though it sits exactly at the refresh rate boundary, making Medium the safer recommendation for maintaining a consistent frame rate above the threshold.

For 4K gaming, the choice is more constrained. The Ultra preset (62 FPS) is playable for slower-paced RPG content but may feel less smooth in combat. The High preset (79 FPS) offers a significant improvement in fluidity, and the Medium preset (93 FPS) provides the best balance of visual quality and performance—it's the highest setting that comfortably exceeds 60 FPS with headroom for demanding scenes. The Low preset (121 FPS) is the only 4K option that pushes toward high-refresh territory, but the visual compromise may not be worth it for an Action RPG where immersion matters.

The data suggests that the Medium preset is the sweet spot across all resolutions: it delivers 259 FPS at 1080p, 169 FPS at 1440p, and 93 FPS at 4K, with minimum FPS values (220, 144, and 79 respectively) that remain playable even in worst-case scenarios. The minimum FPS at Medium settings matches the average FPS at High settings for each resolution, indicating that Medium provides more stable frame pacing relative to its average performance.

# How This Combo Ranks

The AMD Ryzen 9 9900X3D + NVIDIA GeForce RTX 5090 combination ranks 39th out of 2,825 tested combos in The Technomancer, placing it in the top 1.4% of all configurations. This high ranking reflects the exceptional performance of both components, even though the game is nearly a decade old by the time this hardware was released.

The rank of 39 out of 2,825 means there are 38 other combos that achieve higher measured FPS in this specific game. Given that the 9900X3D and RTX 5090 are both flagship components from their respective manufacturers, the fact that they don't rank higher suggests that other factors—such as memory configuration, platform differences, or specific driver optimizations—can yield better results in this particular title.

The CPU's 91st percentile ranking among all CPUs and the GPU's 92nd percentile among all GPUs indicate that individually, these are top-tier parts. Yet their combined rank of 39 out of 2,825 (which is approximately the 98.6th percentile) shows that the pairing is slightly less dominant than the individual percentiles might suggest. This discrepancy could be due to the game's age and its engine's limitations—if The Technomancer doesn't fully utilize the 12 cores and 24 threads, or if it doesn't benefit from the RTX 5090's advanced features like ray tracing or DLSS, then the raw FPS numbers would be limited by the game's single-threaded or lightly-threaded workload.

The nearest rivals for the CPU show that the 9900X3D is barely distinguishable from the Intel Core Ultra 5 245KF (-0.6%), Intel Core Ultra 5 245K (+1.3%), Intel Xeon 6505P (+2.0%), and Intel Core i7-14700F (+2.1%). Similarly, the RTX 5090's nearest GPU rivals are within 1.4% of its average benchmark score. This clustering suggests that in synthetic benchmarks, these parts are close, but in The Technomancer's real-world measured FPS, the combo's specific strengths (12 cores, 128 MB L3 cache, high boost clock for the CPU; massive memory bandwidth and shading power for the GPU) may or may not align with what the game demands.

The rank of 39 places this combo among the very best configurations tested for The Technomancer, but it doesn't claim the top spot. This outcome invites the question: what hardware combination achieves the top rank in a 2016 game? The answer likely lies in a configuration that maximizes single-thread performance and high-frequency memory, characteristics that older game engines favor over modern multi-core or multi-thread scaling. The 9900X3D's strong single-thread score (6,739 in Cinebench R23, 3,041 in Geekbench) and the RTX 5090's rasterization power together produce excellent results, but the game's engine may respond even better to a CPU with an even higher per-core clock or a GPU with different driver-level optimizations for this specific title.