Configuration Comparison

For a comparison of the various configuration schemes, refer to Table 1.

Table 1. General Comparison for the Various Configuration Schemes

Notes:

1. Different devices using the same configuration scheme may support a different external controller and/or configuration device. Refer to the configuration chapter of the respective Intel® FPGA device in the Configuration Handbook for more information.
2. Configuration time is presented as a relative comparison and serves only as a general guideline. Configuration time varies for different configuration schemes and depends on the configuration file size, configuration data width, frequency of the driving clock, and flash access time.

Active and Passive Configuration Schemes

In general, Intel® FPGA configuration schemes are categorized into active configuration schemes or passive configuration schemes. In the active configuration schemes, the device controls the configuration process and gets the configuration data from an external memory device. Active serial (AS) and active parallel (AP) are active configuration schemes. The memory device is a serial configuration (EPCQ) device for AS configuration and a supported parallel flash memory for AP configuration.

In the passive configuration schemes, the configuration device controls the configuration process and supplies the configuration data. The configuration device can be an external intelligent host, such as a PC, a microprocessor, or a MAX series CPLD. Passive serial (PS), fast passive parallel (FPP), and JTAG are passive configuration schemes.

External Memory and/or Configuration Device

All configuration schemes require either an external memory or a configuration device. These external devices are necessary to store configuration data and/or configure the Intel® FPGA when using a particular configuration scheme. For example, an external memory device can be a serial configuration (EPCQ) device or a supported parallel flash memory device. A configuration controller can be a microprocessor, or any MAX series CPLD. Note that different configuration schemes are supported by different external memories and/or configuration devices. The MAX series CPLD supports Parallel Flash Loader intellectual property (IP) to program common flash interface (CFI) flash memory devices through the JTAG interface and provides the logic to control configuration (Passive Serial and Fast Passive Parallel) from the flash memory device to the Intel® FPGA.

• Parallel Flash Loader Intel® FPGA IP User Guide
  
• MAX Series Configuration Controller Using Flash Memory white paper

Relative Configuration Time

The configuration cycle consists of three stages: reset, configuration, and initialization. The relative configuration times here refer only to the configuration stage. The time it takes for the device to enter user mode is actually longer.

Configuration time varies for different configuration schemes and depends on the configuration file size, configuration data width, frequency of the driving clock, and flash access time. You can estimate the relative configuration time between various configuration schemes of the same device family and density.

AS configuration time is dominated by the time it takes to transfer data from the EPCQ to the FPGA device. The AS interface is clocked by the FPGA DCLK output generated from an internal oscillator. The DCLK minimum frequency when using the 40 MHz oscillator is 20 MHz (50 ns). For example, the maximum AS configuration time estimate for an EP3C10 device is (2.5 Mb of uncompressed data) = RBF size x (maximum DCLK period / 1 bit per DCLK cycle) = 2.5 Mb x (50 ns / 1 bit) = 125 ms.

In general, the FPP configuration schemes have the shortest configuration times. For all the FPP schemes, the configuration frequency is controlled by the external device. The AS, PS, and JTAG configuration schemes have a relatively slower configuration time. However, the relative configuration time is just an estimate. The actual configuration time depends heavily on the configuration data width, the configuration frequency at which the device is clocked, the configuration file size, and the flash access time.

Support for CLKUSR Feature

In some devices, the CLKUSR pin is an optional pin that inputs a user-supplied clock to synchronize the initialization of one or more devices after configuration. This feature allows one or more devices to enter user mode at the same time. This pin is enabled by turning on the Enable user-supplied start-up clock (CLKUSR) option in the Quartus® Prime or Quartus II software.

For more information, refer to the configuration chapter of the respective Intel® FPGA device in the Configuration Handbook.

Scalability

The Intel® FPGA EPCS devices (PDF) and EPCQ devices (PDF) support a single-device configuration solution for Agilex series, Stratix® series (except for Stratix and Stratix® GX), Arria® series, and Cyclone® series FPGAs.

To choose the appropriate configuration device, you must determine the total configuration space required for your target FPGA or chain of FPGAs. If you are configuring a chain of FPGAs, you must add the configuration file size for each FPGA to determine the total configuration space needed.