what is the difference between ASIC and FPGA

Question

Recently I started to experiment with FPGAs and while doing this I found out about ASICs.

Can anyone explain me the main differences between them.

From what I have read so far, I understand the following:

• FPGA can be reprogrammed, while ASIC can not (so sometimes people are testing the idea on FPGA and then move to ASIC)
• the cost of one FPGA is lower than ASIC, but the cost of many ASICs is lower than many FPGA

There is a question raised here, but 2 answers there are actually from bitcoin prospective (which is not surprising, given that this is bitcoin-related site). I am more interested in a common answer.

Does anyone has a better way to explain the difference?

P.S. I know that the tag I am using is not really relevant, but because there are no tags ASIC, FPGA, there is the best I was able to find

Answer 1

Well, alone from the wording:

• an ASIC is an Application specific IC (integrated circuit)
• an FPGA is a Field-programmable gate array

which means that ASICs execute hard-coded operations (they will never be able to be used for any other operations that they're not designed to do), while FPGAs are reprogrammable "in the field", i.e. by the user. So one implication is that you could design your algorithm on an FPGA (because it's basically a logic whiteboard) and tweak it until it does exactly what you want, then you can have an ASIC made which implements that same logic, and never any other. You can then use the same FPGA to implement other algorithms as often as you like.