Graph acceleration optimization method based on domestic platform

Abstract

The invention provides a graph acceleration optimization method based on a domestic platform, and the method employs a double-card interconnection display scheme, achieves the cooperative work of twoindependent display cards under the control of the same CPU, and enables the two independent display cards to bear different graph drawing tasks, thereby greatly improving the overall graph acceleration performance of the platform. And high-speed data transmission and synchronous rendering of the double graphics cards are realized. High-speed data transmission of the double graphics cards is realized through a special high-speed interconnection channel, and the bottleneck of the data transmission speed is solved; and meanwhile, synchronous rendering of the double graphics cards is realized byusing an interlaced frame mode or a separated frame mode through bottom-layer drive scheduling.

Description

Technical field[0001]The invention belongs to the field of computer graphics, and specifically relates to a graphics acceleration optimization method based on a domestic platform.Background technique[0002]In the 4K, 4-screen output display state, when the graphics output by the graphics application program is relatively complex, the existing domestic GPUs have insufficient computing and processing capabilities, and cannot meet the display performance requirements. There are many graphics rendering applications in special application scenarios, which have higher requirements for graphics card performance, especially for GPU graphics rendering capabilities. Domestic GPU single graphics cards have low performance and cannot meet the high-performance display requirements of complex application systems, and cannot solve the problem of insufficient single-card display performance in 4K and 4-screen scenarios. Graphics card overclocking can temporarily improve the performance of the graphi...

AI Technical Summary

Problems solved by technology

The performance of domestic GPU single graphics card is low, which cannot meet the high-performance display requirements of complex application systems, and cannot solve the problem of insufficient single-card display performance in 4K and 4-screen scenarios

Overclocking the graphics card can temporarily improve the performance of the graphics card, but it cannot run for a long time, because long-term overclocking not only affects system stability, but also reduces the life of the hardware

Images