Implementing custom DDR and DDR2 SDRAM external memory interfaces in FPGAs (part 1)
By Danesh Jothiprahasam, Altera Corp.
pldesignline.com (March 17, 2010)
Note
This is a two-part article that focuses on the design guidelines and describes how to implement DDR or DDR2 external memory interfaces (EMIFs) using FPGAs via ALTDLL and ALTDQ_DQS megafunctions. This article does not include information about simulation, timing analysis, and board-level constraints that are used to demonstrate and validate these interfaces. For more information, refer to Application Note AN565.
ALTDLL and ALTDQ_DQS Megafunction Overview
FPGAs referenced in this article have complex dedicated I/O circuitries that are primarily designed to support EMIF. The ALTMEMPHY megafunction is designed to support the most common memory standards, such as the DDR , DDR2 SDRAM, and QDR II+/QDR II SRAM (in a burst length of 4) interfaces. Other external memory standards such as Mobile DDR, QDR II+/QDR II SRAM (in burst length of 2), or customized DDR and DDR 2 SDRAM external memory standards are not supported. Instead, the ALTDLL and ALTDQ_DQS megafunctions are used to access the FPGA architecture and build a custom EMIF.
DDR and DDR2 SDRAM in an FPGA Design Flow
These design guidelines provide the best practices for DDR and DDR2 SDRAM custom memory interface implementation in Stratix III and Stratix IV FPGAs. Figure 1 shows the design flow that is required for the EMIFs for these FPGAs.
To read the full article, click here
Related Semiconductor IP
Related Articles
- Memory Design Considerations When Migrating to DDR3 Interfaces from DDR2
- Which DDR SDRAM Memory to Use and When
- Building high-speed FPGA memory interfaces
- DDR3 memory interface controller IP speeds data processing applications
Latest Articles
- Automated Multi-Line Bug Repair Using Model Checker Aided LLMs: A Case Study on RISC-V Processors
- Cryogenic Recharacterization of a Commercial 22 nm FD-SOI Foundry Standard-Cell Library
- A 105 MS/s 4.18 nJ/inference 0.765 mm² Mixed-Signal Streaming ML Engine for Wireless Baseband in 12LP+ FinFET
- Agentic AI for Heterogeneous System Design Across Chip, Scale-Up, and Scale-Out Domains
- Minimum RISC-V: Maximum FPGA Efficiency