Tile Memory is a specialized memory component within hierarchical Z/stencil culling units that stores depth and stencil data for tile-based rendering optimization in GPUs.
| Parameter | Typical range | Notes & selection driver |
|---|---|---|
| Latency | 2-10 clock cycles | |
| Voltage | 0.8-1.2V | |
| Capacity | 16-256 KB per tile unit | |
| Bandwidth | 512-2048 GB/s | |
| Process Node | 7-16nm FinFET | |
| Operating Temperature | -40°C to 125°C |
Ranges are indicative industry figures for RFQ preparation, not a supplier commitment. Confirm every value and standard with the legal manufacturer before ordering.
This component is used in the following industrial products
A practical evidence checklist for RFQ preparation and supplier evaluation.
Compare the published product scope and source details, then include your specifications and quantity in an enquiry.
Manufacturer profiles associated with Tile Memory.
Explore company information and manufacturing capabilities, then discuss the scope of your project with a relevant supplier.
Tile Memory primarily stores depth and stencil information for specific screen regions to enable efficient occlusion culling before full rasterization, reducing unnecessary pixel processing.
By keeping depth/stencil data locally in high-speed memory, it minimizes main memory accesses during depth testing, reducing bandwidth consumption and improving power efficiency in tile-based rendering pipelines.
When tile data exceeds local memory capacity, the system spills to higher-level cache or main memory, causing performance degradation due to increased latency and bandwidth consumption.
Editorial classification, named public sources where available, and source-reviewed manufacturer records. See the editorial policy.