We moved from GPU mining to FPGA (Field Programmable Gate Array) mining more than a year ago. Why did we go for FPGA instead of other hardware? What are the advantages of using FPGAs for bitcoin mining?
Why should I be interested in FPGAs?
Let’s move ahead with our cryptocurrency mining adventure, starting with GPUs for mining Monero and Ethereum. Everything was going swimmingly until the catastrophic market meltdown of early 2018. Everything started to go wrong, and the GPU mining enterprise as a whole became unprofitable. At that point, we had no choice but to look for another option.
In June 2021, we came upon a bitcoin talk article. The promised figures were far too wonderful to be true, and it would be a waste not to give them a shot. We decided to try FPGA mining via Amazon Web Services (AWS) in order to test them out without having to spend a lot of money and spent some time studying verilog.
It proved to be a key step after successfully performing FPGA mining on AWS.
The results were amazing. We immediately recognized there was untapped potential when we saw the VCU-1525 only uses 80–90 Watts with a 300Mhz clock speed. The VCU-1525 FPGA mining card can perform some algorithms 6–20 times quicker than a GPU while using the same amount of electricity!
As a result, we decided to join the FPGA mining bandwagon.
FPGA may earn around $12 per day in today’s market. — 15th Oct 2021
What is an FPGA (Field Programmable Gate Array)?
The full form of FPGA is Field Programmable Gate Array, and it quite resembles a set of LEGO bricks (the building blocks). You can create nearly anything with them, and the choice is entirely yours. Going through SystemVerilog before is better to understand and access the functioning. The construction blocks in this scenario are digital circuits.
We generally develop software to operate on the CPU, which is a digital circuit with the overall goal of performing many tasks/things. The FPGA, unlike the CPU, is a blank digital circuit. All you need to do is design the digital circuit, and the FPGA will handle the rest. To put it another way, you have complete control over the hardware.
GPU vs. FPGA
Let’s make a comparison!
This similarity refers to the speed with which cryptographic algorithms may be computed. Let’s assume you’re tasked with mowing the grass and want to use the most efficient equipment possible.
First, let’s take a look at a graphics processing unit (GPU). GPU is a toolbox containing a variety of tools such as a screwdriver, wrench, hammer, and so on. Although not all of the goods within are practical, the box does contain a few tools that may be used for lawn mowings, such as scissors, a machete, and a hand-held grass cutter. It gives you a lot of alternatives, but they aren’t always the most effective.
You also have an FPGA. FPGA stands for Field Programmable Gate Array, and it is a LEGO engineering kit. You can make a lawnmower using this kit, but it will take more time and work. FPGA, on the other hand, can mow the grass quicker than GPU. After you’ve finished mowing the grass, you may use this kit to make another tool.
GPU (Graphics Processing Unit)
The fundamental function of the GPU is to render graphics. It can generate hashes quicker than a CPU, but it is still slower than FPGA and ASIC. GPUs are flexible because their algorithms can be changed. They’re also simple to obtain, user-friendly, and reasonably priced when compared to the alternatives.
FPGA (Field Programmable Gate Array)
FPGAs, like GPUs, have the ability to alter algorithms, making them flexible. Unlike GPU, though, you must create both the digital circuit design and the software. It’s not user-friendly, since learning how to use it may take weeks to months, and even learning how to programme the FPGA could take months to years. FPGAs used to be difficult to come by, but they are now widely available online. FPGA chips come in a variety of types and sizes, ranging from low-cost ($179) to high-cost ($5750), making them more accessible.
When compared to GPU mining, FPGA mining efficiency (hashing speed/power consumption) is extremely high, and it much beats CPU mining. However, ASIC still outperforms FPGA in terms of speed and efficiency.
FPGA offers numerous benefits over GPU, including being quicker and more versatile than ASIC, which means FPGA may be highly profitable when mining currencies that are resistant to ASIC. As a result, FPGA should compete with GPU rather than ASIC.
We chose FPGAs to mine crypto because of their greater hash rates and algorithm flexibility, which result in larger profits. Miners will eventually understand that they will never receive a return on their investment, and only lower-cost ASICs will survive (ROI). GPU, on the other hand, is up against a device that can run at 5–20 times its own speed, and they’ll soon be out of business. As a result, we opted to begin our experience with FPGA Mining.