- Channels
- Select the number of channels from 1 to 12. Multichannel operation is available for the three Burst I/O architectures. When multichannel is used, control logic is shared over the datapaths, saving resource, though the additional fanout and routing might reduce the achievable clock rate. For Floating-point formats (pseudo and native), the channels must be 1.
- Transform Length
- Select the desired point size. All powers of two from 8 to 65536 are available.
- Architecture Choice
- Select an implementation option, as described in Architecture Options.
- The Pipelined Streaming I/O, Radix-2 Burst I/O, and Radix-2 Lite Burst I/O architectures support point sizes 8 to 65536.
- The Radix-4 Burst I/O architecture supports point sizes 64 to 65536.
- The native Floating-point SSR=2 supports point sizes 16 to 65536, while SSR=4 supports point sizes 64 to 65536.
- Check Automatically Select to choose the smallest implementation that meets the specified Target Data Throughput, provided the specified Target Clock Frequency is achieved when the FFT core is implemented on an FPGA.
- Target Clock Frequency and Target Data Throughput are only used to automatically select an implementation and to calculate latency. The core is not guaranteed to run at the specified target clock frequency or target data throughput.
- Only the Pipelined Streaming I/O supports the native Floating-point format. Automatically Select is applicable, which configures the core to the Pipelined Streaming I/O eventually.
- Transform Length Options
- Select the transform length to be runtime configurable or not. The core uses fewer logic resources and has a faster maximum clock speed when the transform length is not runtime configurable. This option is disabled when the native Floating-point format is selected.