SKIP TO CONTENT

Samsung Semiconductor no longer supports Internet Explorer. Please try using a different browser

Samsung Launches Sokatoa to Enhance GPU Performance Analysis on Android

Bringing multi-frame GPU profiling, Sokatoa enables developers to identify and resolve complex graphics performance issues with greater speed and precision

Various screenshots of Sokatoa

Samsung Austin Research and Development Center (SARC) and Advanced Computing Lab (ACL), Samsung Semiconductor’s U.S.-based design center focused on GPU, System IP and SoC architecture, today announced the launch of Sokatoa, a new GPU profiler for Android. 

While Sokatoa was optimized specifically for Android game developers, it can be used by any developer working on graphics-intensive mobile applications built on the Android platform. Sokatoa supports Samsung’s Xclipse GPU and other industry leading GPUs on Android devices.

“We understand the challenges Android developers face when optimizing the graphical performance of their applications,” said Taekhyun Kim, Vice President of GPU Software Development at SARC/ACL. “Our new GPU profiler for Android with multi-frame support gives graphics developers the tools they have been lacking to identify and resolve rendering bottlenecks more efficiently, leading to smoother, more stable and visually stunning Android experiences on Samsung, Qualcomm and ARM GPUs.”

Sokatoa’s multi-frame profiling elevates productivity for graphics developers building graphics-intensive applications for Android.  By stepping through multiple frames of GPU rendering simultaneously, developers gain deeper insight into complex graphical issues that have traditionally been difficult and time-consuming to diagnose.  Locating and diagnosing performance or rendering problems is easier and faster in Sokatoa due to its rich data visualizations and modern user experience.

  • Multi-Frame GPU Profiling
    Developers can analyze rendering behavior across multiple frames to uncover intermittent, frame-to-frame performance issues that are difficult to detect with single-frame capture tools.
Screenshot of the system profiling data view
  • Fast Iteration Through Shader Editing and On-Device Replay
    Developers can edit shaders, replay workloads directly on device and compare results instantly—dramatically shortening the optimize-test-iterate loop for graphics and game development.
Screenshot of the shader editor
  • Built in Collaboration with Industry Leaders
    Developed in collaboration with Samsung Semiconductor, Google and LunarG, Sokatoa aligns closely with modern Android and Vulkan workflows, ensuring accuracy, performance relevance and forward compatibility.
  • Extension-First Architecture for Custom GPU Insights
    Sokatoa is designed to be extensible from day one, allowing GPU vendors, third-party tool developers and studios to build custom extensions—similar to VS Code plugins—for metrics, analysis panels and workflow automation.
Screenshot of the pipeline debugging view

Early access partners across the Android graphics ecosystem report that Sokatoa provides deeper GPU visibility, supports faster iteration and improves confidence when optimizing performance for graphics-intensive Android applications.

We’ve been testing Sokatoa since the first beta release and one capability we like is that we can see two traces side-by-side. Comparing differences visually has proven helpful for quickly isolating problem areas. We like that we can use Sokatoa to correlate the data from a draw call to a frame.

Vinh Truong, Graphics Engineer, Supercell

We evaluated Sokatoa across several of our titles and found it to work reliably without requiring additional setup or special debugging effort. Its intuitive workflow and stability stood out compared to other GPU profiling tools we’ve used.

Edwin Smith, Senior Graphics Engineer, Feral Interactive

Teams in our ecosystem have been testing Sokatoa and have shared positive feedback on the quality and usefulness of the GPU data it provides. We look forward to seeing Sokatoa continue to evolve and add new capabilities over time.

Oliver Schnabel, Graphics Product Manager, Unity

About Samsung SARC/ACL:

Samsung Austin Research and Development Center (SARC) and Advanced Computing Lab (ACL) is a U.S.-based R&D organization within Samsung Semiconductor focused on advancing next-generation semiconductor and computing technologies. With teams across California and Texas, SARC/ACL drives GPU, System IP and SoC architecture innovation powering premium consumer devices worldwide. As the first to bring hardware-accelerated ray tracing to mobile platforms, SARC/ACL continues to push the boundaries of high-performing, power-efficient computing—shaping the future of immersive graphics experiences and beyond.

List