Performance Guide
Last updated
Last updated
Through the Performance Guide feature, key performance indicators such as FPS, Draw Call, and memory usage can be monitored in real-time within the Studio. This allows creators to identify performance degradation factors such as frame drops and rendering delays in advance, enabling effective optimization. Ultimately, this ensures overall stability, providing users with a smoother and more enjoyable gameplay experience.
To use the Performance Guide, click the Stat button that appears when you select the Play tab in the top-most tab area of OVERDARE Studio.
The Performance Guide is displayed on the Studio’s viewport, allowing you to monitor the real-time changes in performance metrics during test play. This helps you intuitively assess performance changes in various game scenarios, such as player movement, object creation, and effect activation, and use this information for situational optimization.
FPS (Frames Per Second)
The number of frames rendered per second. Higher values provide smoother visuals.
30 or higher
CPU Usage
The percentage of CPU usage. High usage can affect other processes
70% or lower
GPU Usage
The percentage of GPU usage. Affects rendering and graphics processing.
85% or lower
Memory Usage
Total memory usage. Exceeding memory can cause game crashes or performance degradation.
3GB or lower
Texture Memory
The amount of memory used for texture data. Too much can overload the GPU.
150MB or lower
Texture Count
The number of textures in use. More textures increase memory and rendering load.
200개 or lower
Mesh Tri Count
The number of triangles in meshes displayed on screen. Too many can decrease rendering speed.
300,000 or fewer
Draw Calls
The number of commands sent from the CPU to the GPU for rendering. Higher values increase the risk of performance degradation.
200 or fewe
Network
Network traffic and latency. Greatly affects communication responsiveness
Less than 20KB/s, under 80ms
Resources used by the Studio itself are included in the measurements, which may cause a difference in performance compared to actual mobile devices. (In particular, memory usage and rendering performance metrics may appear higher than in mobile environments.)
When using the multi-test play feature, performance metrics such as CPU, memory, and network usage may be higher compared to single-player. (While this is useful for simulating a multiplayer environment, it should be interpreted separately from single-client performance.)
GPU usage and FPS values may be affected depending on the viewport resolution settings.
Performance metrics may also vary due to other background programs running on the PC where the Studio is executed, so it is recommended to keep the testing environment as controlled as possible.
Check to ensure stable FPS, and if frame drops occur in specific areas, examine the calculations, effects, scripts, etc., at those locations to optimize them.
Monitor CPU and GPU usage to ensure they remain within a certain level. If there is a sudden spike in specific situations, identify the computational load and bottlenecks at those points and improve the processing logic.
Check if memory usage consistently increases during long test plays. Verify if there are memory leaks or unnecessary objects being retained, and implement cleanup routines (e.g., Destroy(), setting reference variables to nil) to stabilize performance.
Excessive texture usage can lead to GPU overload or loading delays, so use high-resolution textures only within the necessary range and adjust asset resolutions for optimization.
If the mesh triangle count for characters or environments is high, the GPU rendering load increases. Simplify meshes that are unnecessarily complex to improve performance.
If there are too many draw calls, the cost of calls between the CPU and GPU increases. Minimize draw calls by combining object placements or standardizing materials and shaders.
If network traffic spikes within a short period, it can cause server processing delays or increased latency. Control events that cause excessive packet transmission (e.g., repeated updates, high-frequency position sending) and adjust transmission intervals as needed to optimize traffic.
