4 ms·
This seems rather niche. Most real world applications don’t provide the luxury of knowing your desired response ahead of time. Also, consider implementation in
by p5a0u9l 7y ago
This seems rather niche. Most real world applications don’t provide the luxury of knowing your desired response ahead of time.
Also, consider implementation in an FPGA where there are decades worth of validated IP to select from vs. this.
- detaro 7y agoWhy would you add an FPGA for a simple standard algorithm like this?!
- flerchin 7y agoClock rates are higher. You can go into the multiple GHz. Mostly useful for radar and electronic warfare. (not op)
- kragen 7y agoThere are of course FPGA implementations of Goertzel, but for almost any DSP application with 10 MHz or less of bandwidth, a CPU is plenty. No need to deal with Xilinx’s shitty proprietary software and debugging with sigrok for something you can do on a CPU. In software there is nothing stopping you from changing your Goertzel coefficient, and thus the frequency detected, on the fly.
- vardump 7y ago> Also, consider implementation in an FPGA where there are decades worth of validated IP to select from vs. this. Yeah, encrypted "intellectual property" that acts like a black box. Pass.
- kragen 7y agoUnfortunately the electronics industry has adopted the propaganda term “intellectual property” to mean “designs”, so that's what the parent is referring to: decades worth of validated designs.
- jacquesm 7y agoNot really. After all if you can do something in 'just software' your market size is suddenly a very large multiple of what it would be if you did an FPGA version, the marginal costs are much lower and the delivery can be done over the internet.