Topic: ASIC HOMEMADE (Read 276 times)

ASIC is short for Application Specific Integrated Circuit.  It is basically a microchip that is custom-designed to do one specific job super efficiently and  unlike something general-purpose like an FPGA that you can reprogram for different stuff after manufacturing, an ASIC chip can only do that one thing it was originally designed for.  It is way faster and lower power and cheaper at scale since every transistor and connection is optimized for its exact purpose.  But on the flipside, the hardware is set in stone.  You would have to design a whole new custom ASIC chip from scratch if you wanted it to do something else.

What you are looking for is the Bitaxe project: https://github.com/skot/bitaxe
Skot, the owner of the project, is active on Bitcointalk through these topics:
Bitaxe ESP32 miner firmware released!
The bitaxeUltra: Open source Bitcoin miner based on the BM1366 ASIC

Their Discord ( https://discord.gg/JPNQtsDQ ) is active, and full of people which are taking old chips to build miners as a hobby and as small solo miners.

Surely you could use those turbo asics (with old chips) on different coins.

Some where there is a thread with the s7 it could do 4 or six boards on the controller .

But like you said it would not be efficient.

The best you can do get an s19xp  and run a board not three.  use the epic controller and set fans real slow.

Aside from the readily-available in the market,
he may be able to fulfil his "Homemade ASIC" objective by modifying the control board schematics available in the internet.
Of course, he need a qualified (ASIC) computer engineer to do this.

Some of those who still own those old Bitcoin ASICs do exactly that: they use it to mine low-hashrate Altcoins with compatible algo to earn $0+.

ASIC's full name could be the answer to that question: "Application Specific Integrated Circuit".

The difference between using a GPU and an ASIC to mine might be 0 to 100, but in the end if your profit due to the ASIC's energy consumption is 0... even x100 that's still 0.



Maybe it worked, but it still doesn't mean that it was profitable.



An ASIC can literally only run the algorithm they have been designed to do. The only way you can make it run more efficiently is by making it run slower or maybe by having it run at a lower voltage; though the latter would result in a slowdown as well due to increased error rates.

you can get six board controllers.

let me look for a link

here is a board that can do six or 8 hashboards

https://www.ebay.com/itm/196040868236?


good luck with it.

So you are going to slave and get ASIC chips from the s9 hashboard and put to anywhere you want?
You can do that just like other developers do but it requires developing skill and knowledge look at the USB stick miner, they also use the same ASIC chip type BM1387.

However, I don't know if the chip that came from s19 would work and make it a USB stick miner never heard that. The chips from the s19 I think it draws more power that a USB stick can't handle.
Would you mind sharing the link where you saw the Russian who made a pen drive into ASIC USB miner? I just want to see how did they make it work.

I understand and thank you for the response.

When I spoke about the S9, it was an example to imagine. I'm looking for a way to try something different in relation to mining.

We use S19 and we think we are at the top of the chain, and then, they launch a new machine, which mines an algorithm that has not yet been mined and "POFF", this machine starts yielding 2 thousand dollars a day.

I think about innovating, thinking ahead. No chip can be used to mine another currency, using the Chip which is sometimes even inefficient, but compared to machines with cards they earn from 0 to 100.

I saw a Russian who put an S19 chip on a pendrive and the thing worked.

I'm trying to think outside the box, analyze the options and cryptocurrencies that exist and look for something that doesn't need to be 0 to 2 thousand dollars a day, but 0 to 10 or 100... maybe.

Does an ASIC designed for ETC or KAS or SHA256 only do this? At the hardware level, can it not do anything less than less efficient?

And?
You would have a machine made out of old used chips, highly energy inefficient that will probably never make a penny even if you got them for free.The s9 is making 84 cents while burning 32kwh, anyhwere above 3 cents per kWh with those chips you're at a loss. What would be the point?


Everything costs as much as you're willing to pay for it!
If anyone would be able to make the machine for $2000 he would have done so and sell it for $14 000, making a profit and outsmarting bitmain. But nobody was able to! Second thing, would you sell all the chips on a miner you have paid 15k for, and which would make you $100 a day for 500? Again, no!


You can build a Pachinko game in a game because it's just code, here you need a special physical tool in order for that, the chip, in Minecraft you could build an ASIC miner or a Bitcoin mining farm, it won't mine a single coin, just as playing Farmville will not give you a single apple.

Hello,

I see antique machines, like the ASIC S9 miners and I wonder if we could create an ASIC machine using these chips, without thinking about the commercial format.

For example, imagine a central processor (control board) that can simultaneously control 20 Hasboard S9 boards, it would be 120TH. Maybe in immersion.

The question is the following: Wouldn't it be possible to create an ASIC card controller device or even the board, using commercial chips that are sold on the market?

The KS3 antminer costs 15 thousand dollars. It cost 50 thousand dollars. Therefore, it should have a manufacturing cost of 500 dollars, like the current S19. However, when launching companies, they charge a lot because they need to 1. make money, 2. make a lot of money and 3. pay off the investment.

We could use common market processors at the hardware level to be used to mine some crypto, I'm not talking about creating a computer, installing an OS and then installing mining software. I mean using processors as slaves just to mine.

Would this somehow be possible?

People created a virtual computer within Minecraft that runs Minecraft... the sky is no longer the limit.

I hope someone has this knowledge to comment here, talking about all the aspects and possibilities. Analyzing the processors and chips that we have available on the market today.

For example, a KS3 could turn into 10 machines of 800MH each, couldn't it? The chips are inside and sold for maintenance and repair.

Best Regards!

:-)