Bitcoin mining discussion

Hey everyone, I’m new to this kind of info search, so excuse me if it’s a bit off-topic (not sure where else to go). I’m looking to create an ASIC and have figured out some basics like how computers operate (adding binary numbers, boolean logic, RAM functions). I also get how hashing works (taking a float and turning it into random outputs) and have some knowledge about chip-making (photolithography). But I need to dig deeper and I’m feeling a bit stuck, so I’m reaching out for help: First, the easy stuff: 1) Is this a good place to ask these kinds of questions? 2) Where would be a better spot to ask them? Now, the more complex things: 3) Any recommendations for books or resources on:   3.1) Bitcoin ASICs in general (their history and current tech)   3.2) Complex processor functions   3.3) An in-depth look at how SHA256 works   3.4) How ASICs get their tasks   3.5) The workings of mining pools   3.6) How USB miners function   3.7) The infrastructure behind ASIC chips   3.8) Steps to design an ASIC chip Thanks a lot for any help, and sorry if my English isn’t perfect.

I'm not sure anyone here has experience in designing and making chips it's dominated by only a few companies. My answers below are my best guesses based on what I think will be a useful starting point. I don't think you'll get a good profitibility out of chips if you're just doing this hobbyistically - there's quite a bit of competition in this space imo and they probably have to compete quite well with each other already (some have stopped producing entirely too). Kinda but you might want to ask it on more forums too or in communities where chips are designed. The sha256 algorithm is the main function asic chips implement as weel as (probably) a comparison to check if the header is below the target - operations are normally done in parallel as much as possible afaik to increase the speed a block is found (this might be less necessary though as series processing over the same amount of time will produce a similar number of outputs for ~10 minute intervals). Pools send jobs to computer workers that mine the blocks based on the information the pools give them. These "workers" are pieces of software on a "normal" machine (ie one running Windows or Linux with the asics connected as peripherals) that deligate tasks to asics. I think the APIs for these are quite well implemented though so you'd just have to look at integrating the asic chips with preexisting software to start with. This might be easier to work out by buying a cheap USB miner (or making one yourself out of a cheap device). You could also look at software like cpuminer or check other crypto mining software that's open source. You'll likely want to start off with the basis of a chip that can do sha256 algorithms and then work from there to add things.

Is you goal to manufacture an ASIC or just design one? Either way, if you have only a rudimentary understanding of everything, then this is not something that you can accomplish in your spare time over the summer.

Designing a new ASIC chip, let alone an ASIC miner, will take a team of electrical/electronic/computer/software engineers. I'm thinking at least 5 people in total. It will cost you hundreds of thousands at the lowest. If you want to build something new for mining, why not write GPU mining software for a new coin or for a CPU coin? Or a stratum profit-switching proxy? There are so many ideas that you can bring quicker to market by yourself, as long as you know programming.

From the discussions about the new intel miner & chips it was in the 8+ figure USD range. So at least $10 million from conception to prototype. And that is from a company that HAD the staff and knowledge on how to do it. Would be interesting to know what it cost some of the original makers of miners in the 2011-12 time-frame vs today. -Dave

According to this source, Butterfly Labs sold pre-orders for 60k units at first. I vaguely remember that the average price of the 50 Gh model was $500 or so. That comes out to $30m of funding in total. However, many of these funds went to Butterfly's own warehouse in Missouri. They later failed to pay the power company, then folded in 2013 or so for whatever reason.