Topic: What does it take to start manufacturing miners? (Read 1415 times)

My guess is inexperienced and locked into a 'well it worked before' mentality. Just gloming together more VR modules in parallel to give needed very high current and low voltage is just a small part of the battle.

Getting the desired power from the VR's cleanly to the chips is a whole 'nother ballgame and requires looking at the PDN like a radio circuit than a simple DC bus. Probing the power planes, the spiking rise/fall times and ringing you see on them even just an inch away from a different spot will be very different. To me at least, that is the biggest advantage to the string topology - currents are far lower making power plane considerations much easier to deal with.

If it was very difficult for them to design and test the regulators I would say they were inexperienced.

It's funny you mention poorly-designed power on the Monarch, because I remember talking with one of their guys on here a while ago and he was going over how difficult it was to design and test the regulators and that was one of the big holdups that delayed the product.

Agreed. The tickbox of required skillsets is extensive even if using an 'off the shelf' ASIC vs rolling your own.

Having been involved in forensics on the ill-fated Bitmine/AMT A1 boards it was clear whoever did the board design knew nothing about high current PDN's (Power Distribution Networks) much less thermal design. BFL's Monarchs are another example of failing on just those 2 points. Both Bitmine.ch and BFL apparently had no one in-house or consulting with any idea of just how different those power-hungry chips were from what they had dealt with before and what it meant to PDN and thermal considerations or if someone did raise flags they were firmly ignored.

Then yes there is the coms to work with. CGminer and using SPI of course works and is very flexible to modify/work with but... The underlying coms to/from the ASIC's leaves much to be desired. I know that chip-level IP is available to make each ASIC an address on a true local machine network so date rates to/from ASIC's and controller can now hit T1000 speeds if desired with the router/controller buffering what the mining server can handle/will allow. I'd think that ad-hoc addressing when needed vs fixed interval polling would be a far better way to handle the data today's chips can crunch.

As someone whose company is actually deep in the process of bringing to market a "from-scratch" miner, let me loudly and clearly echo the sentiments voiced in this thread. It is hard. It is expensive. It is time-consuming. It is an utterly brutal undertaking.

But huge risk sometimes occasionally hopefully yields big rewards, both for the legions of bitcoin miners and for the company selling the hardware.

This ^

An red anti-symbol stamped over a lunch being given out would be classic!

I would love to see a Suregons General Warning type thing on every miner.

"Warning:

This miner is likely to loose you money after all cost are considered. Mining may be addictive and you may have trouble quitting. Mining can cost you valuable time you will never get back. Mining will become rapidly unprofitable as the block halving nears."

I'm sure there are better ones you could come up with but you get the point.

BTW. Starting a mining manufacturing process from scratch is a pipe dream. Unless you can come up with $10,000,000 or more that is. And your willing to loose every penny, and deal with the consequences involved with that. Many have tried and failed. Best to focus your efforts and time on something realistic. Speaking of which....

You can do it through a kickstarter program and accumulate the funds needed.You also need a good ponzi scheme and a good salesman pitch to fork your idea to the people.  
But in all seriousness ,having investors is a must,unless you already got alot of money. Also the best way to get a deal done,is to go in person to look at the manufacture(s) you will be doing business with and breaking the deal with the person who can actually get "things done" .

The main reasons I refuse to work with big-die BGA chips include what you pointed out - High density high current low voltage is stupid, and high heat density is also stupid and the complex/exotic cooling it tends to require is unreliable. Also glued-on heatsinks is stupid. Starting with a chip that isn't designed badly is really conducive to boards and miners that aren't designed badly.

It's a bit more than the chips. Plenty have made chips, then failed on the implementation.

For example: Power: How are you going to move [email protected] volts an inch across a chip without significant drop? Are you going to put your dc-dc converters in a nice hot line so they commit fraticide, next to your USB control chips so they blow those up, or around the chip in an implosion style that causes the DC-DC's to miss the air cooling and... well, implode?

Heat: Heat is the ultimate bullshit detector. Where are you going to put the heat generated from the chips, from the DC-DC's, from the 12 volt input lines (oh that's always funny). from the control boards? Which way is the heat going? Up, down? Through that wonderful silicon? How are you going to move the heat somewhere else, remember the more complex the more likely it will break in shipping, handling, and general user error.

Signal noise: Nothing funnier than having your outputs corrupted by noise from those big bad buck supplies you built there. How do you handle that?

Mounting: Nothing like having the ole air cooled heat sinks falling off due to glue problems.

Users: What are you going to do when your purchaser runs it in a sealed room in the middle of a desert?

Optimism: Design for the worst case, it will probably be worse than that.

Coding: Greatest chip in the world is useless if you can't get the answers out super quick and new questions in. What is your queuing tools, language for talking to it, ways to handle errors?

Companies that have had success have been through iterations already with their earlier technology and have had time to figure out what works (ant's concept of small chips instead of big dies is a very good one, despite the repeating power and control parts) and how to keep it going (note the S2-S5 designs being very similar. Reason for that, it works). You're stepping in with a knowledge level of zero. There are probably a ton of other gotchas waiting around.

This is complex stuff. As in "Intel looks at you like you're nuts" complex. And even if you build the ultimate miner the difficulty will probably render the investment moot unless bitcoin goes to da moon. In which case wouldn't you have been better off just buying coin and hodling it?

Someone should stamp MINING IS A ZERO SUM GAME and NO FREE LUNCH on every miner. But in a way they are selling a dream. Don't sell dreams......

I suppose the biggest hurdle is getting viable ASIC's to use. That means either striking a deal with proven chip makers or going full ground-up custom.

As Bitmain proved with their revamped custom 28nm chips, using pre-packaged chip IP blocks that are cut and pasted into a mask then linked together to do the functions ala' Innosilicon and Bitmain's earlier chips is not the way to go. Most off the shelf IP blocks are to some extent fairly general purpose within their type of function and not specifically optimized just to crunch SHA.The IP blocks in question were developed by the various chip design houses like Inno and chip foundries like TSMC, GloFo, et al  to be reusable in different applications with 'acceptable' results.

That said, get licensed modification access to the IP in question so one can trim the fat and optimize the inter-block connections right at the block layout level and the results can be impressive..

It also calls for equally impressively skilled chip designers and deep pocket$$$.

Well thanks for spending time and writing things that i read twice not to leave anything out.

Now i understand the whole story much clearer. I don't have any plan yet, Im working on ideas and am trying
to understand this community. Im happy to hear, that "reputation" and "trust" both have meaning here and
it will help me to plan better.

That's why i said Im looking for opinions, and am grateful you guys give me some hints.

I am not looking for moneymaking, i do enough, believe it or not. imho profit is a byproduct, it cannot be the goal.

Well i will listen to all available opinions and maybe some day will come back with a real plan, reputation, whatever or will never come back

Thanks a lot ! Really appreciate that.

Yes, this is a great idea in theory, but it is not practical for the reasons Finsky mentioned.
Kogan, once you have been around a good while you will see this community has been burned so many times it requires people with great reputations to step up and start a project such as this, and it isn't easy.
There are a couple of different people who have offered fantastic deals on hosting in different places the community would not touch because of the location, the person did not have credibility, etc.

Say you do have that credible person on board, such as Finsky, sidehack, or other people who have done several things in working with many of us home miners, well they already have a business, have projects in the oven, and sidehack has been trying to get some chips from a manufacturer.

You'd need to show you have something or someone to have a foothold. A way to get chips, a building, what is your plan? Then, maybe then you would generate some interest. Obviously money is not the biggest issue here. It is contacts and reputation.
This community has pulled together mountains of millions. It is all written in the forum threads.

Less people will even look at anything which doesn't involve a reputable member.

Good Luck!

Ok now i understand. this happened too many times before.

But im talking not about preorders. I think of a co-ownership. Not like we ask people to fund us so we try to manufacture
miners and then give these miners to people and make some profit. Like there also can be a company that will ask people for money and promise miners, people give money, company buys ferrari and says it's sorry. Not that kind of stuff. Company is owned by all shareholders (Officially, on paper) , backed by their money, backed by some additional business that can somehow use lots of heat generated, and all those people would decide what to do with the profit, reinvest or withdraw. Otherwise, if bitcoin is going to somehow become a future currency (and there is a slight chance of it), we will end up again with big corps mining and processing and ordinary people in the same situation as it always was.

backup idea before starting to manufacture, is that some certain amount of home miners, who do not mine at the moment because of electricity price, have some nice miners and are wasting their miner's lifetime which is going to end quite soon, could ship their miners to one location where a farm could be set up and start generating profit. As far as i read there are lots of people who have their miners shut down because of many reasons.

Theoretically even if bitcoin fails we could end up with several other nice businesses set up.

There have been many manufacturers that were crowd-funded, AMT, BFL, Black Arrow I believe, and everyone else who has tried to get away with selling pre-orders but resulted in their customers being out many millions of dollars. Then there was WASP, I don't believe it ended favourably either TBH. I understand you're talking about something that differs in ethics, but in practice it is essentially the same.  I don't recommend it...

Good point. Thanks.

Maybe i dont say something new but did somebody try to make some crowdfunded project like that? Lets say finally it can turn to a big farm which is owned by 2000 people lets say and it produces miners for itself and mins bitcoins. And miners are situated in some energy efficient place with some extra facilities that can benefit from generated heat like a greenhouse or whatever else. I dont know, its on idea level and im looking for opinions.

To be serious about it, you need LOTS of money and the ability to develop a good relationship with one of the chip manufacturers (Innosilicon most likely) so that they're willing to sell you lots of chips at a reasonable cost.

 There really isn't enough market to support new manufacturers at this point, as it would probably take at least ten and more likely 20-50 million to break into the market at this point as a chip maker - given "start from scratch" you'd pretty much have to go straight to 14/16NM full custom to have a competative design by the time you had something to bring to market and TRY to compete.

 Up side, that generation should get a few YEARS to achieve RoI, as it's not likely to be available from anyone to compete WITH 'till sometime after the next halfing.

I think some of you have thought about that and maybe some of you have some good info on this topic.

So the thing is, that assembly plant is quite easy and not expensive thing to do, so lets skip this point.
Heatsinks, coolers, boxes, power supplies and so on is not a problem also. I mean supply of all this is
very easy thing to organise.

Key things to do are asic boards and their control boards + interfaces. all the rest is cgminer.

So, how do you think, where is the point to start manufacturing ASIC miners? Where to go for chip boards? chips? etc etc

I read a lot about existing manufacturers, own couple different miners, had bad experience with support, answering question etc etc, also hear about users who paid money and got nothing and so on. I believe this can be improved in some way so here is where the idea comes from.

May be am not posting this in a correct part of this forum, or maybe there is a thread on this, which i could not find. In these cases "Go to this link, there is a thread on this" or "stop posting new topics on existing stuff" etc etc will be much appreciated.