This page is moving to https://bxbsp.com/BxBFFT_Agilex7.html
The BxBFFT for Altera Agilex 7
The BxBFFT now supports Agilex-7 FPGAs. Even in this initial support, the BxBFFT is highly competive with other FFTs in Altera FPGAs. Some of the advantages of a BxBFFT in Xilinx Ultrascale/Ultrascale+ are these:
Large FFT Sizes Supports 128k and greater FFT sizes with efficiency.
High Throughput Highly parallel processing at the highest clock rates. Also the lowest latency.
Feature Support Non-power-of-2 FFTs, real-to-complex FFTs, many included options to meet design goals.
Vendor Independence Efficient support of multiple FPGA vendors, with a path to ASICs.
Productivity Meets timing easier. Easier FFT controls.
Some of these advantages apply equally to all BxBFFT implementations. These are discussed on the main BxBFFT page. Advantages specific to the Altera Agilex 7 are discussed below. The BxBFFT may acquire additional advantages as it is further optimized for Altera architectures.
Power Savings
The charts below show the BxBFFT power consumption versus other FFTs. The first graph shows power consumption of other FFTs relative to the BxBFFT in a configuration with fully-natural input and output data orders.
The second graph shows the performance of the Altera FFT versus BxBFFT. A separate chart is necessary because the fast version of the Altera FFT has no bit-reverse support, and thus it can't do fully natural data order at both input and output. Thus the second chart shows performance without a bit reverse.
These charts show that with only a few exceptions, the BxBFFT consumes less power than competitors. The second chart also points out that the Altera FFT can't do the largest FFT sizes or the highest levels of parallelism.
Throughput and Latency Advantages
Sometimes designs need to meet strict real-time requirements, either in throughput or in latency. Both of these improve when an FFT runs faster. A faster FFT can be achieved with a higher achieved FPGA clock rate (Fmax) or with increased parallelism. Parallelism is measured by the processed complex data Points Per Clock (PPC), also called SuperSample Rate (SSR). Throughput is Fmax * PPC.
One issue is that as PPC increases, more resources are used, there is more resource contention, and thus the achieved Fmax of an FFT goes down. This may make the desired throughput unachievable.
For BxBFFTs, Fmax degrades less from resource contention. BxBFFTs are thus able to achieve higher throughput, because a high Fmax and high PPC are simultaneously achievable. The graph below shows this. The BxBFFT achieves high PPC and high Fmax simultaneously, when the other FFTs do not. Thus the BxBFFT provides the best throughput and latency.
Resource Savings
The BxBFFT is highly competitive in FPGA ALM utilization, as shown in the graph below.
The BxBFFT is also a leader in FPGA DSP utilization, as shown below.
These results illustrate how the BxBFFT is a solid performer on Altera FPGAs. It is unmatched at the largest FFT sizes and speeds. It is unmatched in supported features. It is also cross-platform, supporting both Xilinx and Altera FPGAs, with a path into ASICs.