A quick update on mining, the datacenter, bots, and a few other happenings over the past two weeks.
I found another GPU!
I went through my tech graveyard and happened to find another graphics card destined for recycling. I am excited to see if it would run on my mega miner, I haven’t ever seen 13 gpus on the same motherboard. So, I painfully shut down the miner and wired in the new GPU, then powered up – ugh… only 8 cards show…
It was the end of the day and I wasn’t in the mood for a battle. I figured it should rest anyhow and I left it off for the night. It makes me so sad when my miners aren’t running, but this seems like something that will be solved easily in the morning. For some reason, I don’t know why, but whenever I hit my head against a technical wall one day, it seems I’m able to quickly resolve the issue the next without too much pain.
I did in fact resolve the issue – I had to re-do some of the wiring for the new GPU addition as I had 3 GPUs plugged into the same power plug. Apparently that was too much juice for one power supply circuit to handle. I then balanced the GPUS having no more than 2 plugged into the same power cable all 13 gpus came online!!!
Check it out:
I changed the configurations a bit over the next few days and only the last gpu started acting up, eh, it’s no big deal. The temperature was still 67 degrees celsius inside the GPU but the fan started spinning at 84%. So it’s using a tiny bit more energy, but all 11 other GPUs are humming along. Still has an internal wattage of 777, which is honestly amazing considering the hash I’m receiving at 235.8 MH. Lets see if I can get over .2 ETC per day!
OK, I have to demystify a few things here. You’re probably thinking something along the lines of “what in the heck is this monstrosity hanging on a basement wall really doing?”
Put as simply as I can, it’s recording financial transactions on a super large ”excel-like” spreadsheet called a Blockchain. Great, so what’s so special about that?
Excel can be changed after that first record insertion. That’s a feature of Excel so it’s not a problem and that’s why banks DON’T use excel to store their client’s transactions. Spreadsheets are just an easy way to relate the concept of a really long list of bank balances and updating them over time to reflect withdrawals, loan payments, and interest…
Databases are what banks use to store all the information on their clients and databases can be changed as easily as Excel if you know how. Of course, this creates issues around trust and security. Think of a bank having a datacenter of servers holding all the balances, deposits, withdrawals, interest, and loan payments of their customers. That one datacenter is susceptible to a security breach or even environmental disasters that could alter or destroy the information.
Banks have the money to run redundantly out of several data centers and pay a near fortune to protect their systems from hackers and inside threats. This is where a Blockchain thrives as it is immutable. Meaning that once a transaction is recorded by a miner it can never be changed. Therefore a Blockchain doesn’t need a large budget to solve the security, environmental, and redundancy issues that traditional finance institutions face.
Blockchains have many miners that ALL have the exact same copy of the blockchain stored on their devices. IF ANY one miner has the wrong information, it’s automatically kicked off the network so it can’t poison the blockchain. Additionally, if any one miner doesn’t follow the well documented procedures, its hash will be ignored. Even a single GPU on a miner can get booted and the rest of the GPUs run normally – I saw this a lot when this miner was upstairs in the heat. I had to restart the miner several times a day to reset the GPUs and start their mining again as they all were spending electricity but NOT participating in hash generation.
Hash is how crypto miner owners measure things. The more hash you get out of your equipment, the more crypto the computer is producing.
I like to think about it as 1 hash represents one full cycle through the miner process to record a transaction. There is a dance that each miner has with the network and whichever miner is quicker to get the correct result wins that “hash.” This dance involves several mathematical equations that a human being can’t solve. These cryptographic equations can only be solved by extremely powerful processors and are baked into the mechanisms to record transactions. This makes it impossible for a human being to alter the transaction. Altering a blockchain manually is off the table.
For each successfully completed transaction, the blockchain equates to a “hash value” – so the inverse is used to predict the profitability of a miner. So the more hash I get out of my miner, the more ETC I will receive.
For comparison, I was seeing about 160 MHz hash upstairs and downstairs setup in the temporary space about 190 MHz and now a whopping 235 MHz! This means I’m making almost double the amount of crypto since my wall miner is processing almost double the number of transactions.
I’m still waiting for the electrician, I think at this point it’s my fault – I gotta get calling him and not expect him to suddenly remember a random conversation before he left for vacation… but I did make progress on the nodes. I worked with the techs and we customized a 3-node cluster with NUC hardware. NUC (Next Unit of Computing) is a product from Intel that changed the face of PCs forever. These NUCs are small but powerful computers.
Sorry to geek out, but I got 3 x 1TB hard drives and 64GBs of RAM for each and they came with a massive 6 core 12 thread CPU. Once everything arrives I’ll assemble the three servers and install the OS. This OS is an open sourced project that allows many virtual machines to be set up inside each of these physical machines. It’s kind of like an apartment building – all the virtual machines live inside the same physical machine. This is at least 2x what I need currently as it’s important that I can expand my nodes over time as this is a pillar in my plan. Nodes, Mining, Solar, Staking, and Bots, oh my…
Each of these servers are joined together via a network so I can set up virtual machines on any of the devices in one interface. This allows for several virtual machines to be running on each server and each uses a tiny bit of the overall cpu, ram, and space. They are going to be configured in a high-availability cluster which means that if one of these NUCs stops working the other two will take over without dropping any virtual servers. This is to be built for performance with a low power footprint. I’m used to servers that weigh upwards of 80 lbs and to me this is a dream! But I’m still probably going to hang them on my wall LOL.
I have expanded out my bot army onto a few new platforms and gone through nearly all my assets to find a proper place to stake them. With some crypto assets I’m getting 250% APR while others only 2% APR. On average, the staking APR is probably in the 20% range, but my bag size is pretty small.
Staking is really neat, it allows a person who holds crypto to lock their crypto in a contract that delivers in-kind rewards for the length of the lock up. Think of staking as a savings account. If your savings account is a USD account it will gain interest monthly in USD. If you had a EURO savings account, it would earn interest in EUROs. All of my staking assets earn more of the asset I stake, so it’s a way to increase the amount of crypto I hold passively and is key to my success.
Originally, I bought a small amount of many different crypto assets when things were starting and each of those have become a “bag of crypto” – I have bags of ADA (Cardano) ATOM (Cosmos) LINK (ChainLink) etc. Each is separated into many exchanges, investment platforms, staking services, and wallets, meaning that each of these little savings accounts are currently small in value if I were to sell to USD – some only worth a few dollars total. I have hundreds of them set up and staking which means that the quantity of digital assets is increasing over time. I track them all and know the US Dollar value at any given time as well as the BTC value of any asset that trades with BTC, so if I see an opportunity for a decent enough gain, I unstake and sell.
In a crypto trader mindset, each unique place with unique crypto is called “a bag”. For example, I currently have 2 bags of ADA: one in each of the leading software wallets, but BOTH are staking for the same percentage (10%). I split this across two platforms for security of my assets just in case I lose the password to one, I still have the other. Though to date I have not yet lost a password, knock wood…
For me, I pulled out my US Dollars that I invested a long time ago, so whatever crypto remains is without a true USD cost basis… Though I consider all the pain and suffering I went through as a cost as well as all the money that was stolen from me, so even though I don’t have a US Dollar cost, I track a USDT cost basis on all of my assets that I have yet to de-risk.
I’m a HODLer (Hold On for Dear Life) meaning that I wont sell my crypto until it hits my price. It’s like someone who is vested in a company; they don’t want to sell the company stock when it goes down, they want to hold onto that and wait for years as it appreciates. I don’t need all my US Dollars right now, nor do I need to know how many US Dollars I will have access to in my lifetime at all moments of my life.
It’s a hard thing to convey, but the crypto market has taught me that there is no need for 100% liquidity and oftentimes many US citizens assume they need to measure their life by what USD liquidity they have access to at any given moment. This leads to people wanting to see the dollar value of their investments go up everytime they check their balances. This is an unrealistic aspiration. In adopting the crypto mentality, certain of my hundreds of bags are matured and de-risked, so the value in USD can be left alone to grow or be sold anytime.
De-risking is taking back my initial USDT investment and no matter who you are in crypto it’s a very good idea to do this.
Let’s suppose I bought 20 UNI (Uniswap) for 100 USDT total giving each UNI a cost of 5 USDT. Let’s assume that over time that “bag” appreciates and is now worth 200 USDT. If I sell half my UNI (10) for 100 USDT, I’ll still own 10 UNI worth 100 USDT and get back the original 100 USDT I spent to purchase all the UNI. That becomes a “de-risked bag”, meaning that I can sell it for anything and it’s 100% profit. It may go down to 10 USDT and if I sell then, I make 10 USDT. It may go up to 1000 USDT and if I sell then, I make 1000 USDT.
To make a gambling analogy, this is “house money” and for the cpa’s out there, I’m “net-neutral” for a de-ricked bag. (I think that’s the term.)
I plan on holding most of my bags at least another year until we get to the next bull cycle, though some of the bags may die this bull cycle and I’ll sell those off as they hit their peaks.
So as I branched out on a new service (Pionex) I was delighted to see that there is an “infinity bot.” This claims to make profit on HODLing assets. This is precisely my investment strategy – so to start, I transferred a de-risked bag of UNI and some ETC which I mined.
From what I see, this “infinity bot” makes trades that net each other out on one side of the trade over the course of general trading. It’s NOT like a grid bot because there are no predefined buy and sell points – those are established dynamically by the price and momentum of the market.
I have turned to some deep analytics and sifted through quite a bit of my portfolio organizing it even farther to indicate what bags have no cost basis. My goals are to get all my bags de-risked and to start adding to them with some cleverly created Base Building Bots. I just haven’t found any yet and I’m beginning to think that I may need to make one myself. Pionex Infinity bots look promising, but I still want my own.
I am a systems architect who specializes in API (Application Programer Interface) usage. I dreamt up many business systems and oversaw developers to create them. Each exchange has an API which allows authorized commands to be sent directly to the exchange to place trades. Once I have all of my miners running nicely and all my nodes set up on my new server cluster, I’ll be ready and able to dive into creating some bots of my own.
I believe the best way to realize my future is to focus on creating an autonomous bot that will go on indefinitely and win on each trade. This is something that’s in my brain now, but very difficult to explain. Once I have the virtual server built for this need I’ll be able to dive in for real. Until then, I have to use other people’s bots. But, those bots aren’t bad at all, they are just annoying to restart all the time…
Once my system takes over in the basement I want to be hands off trading as computers do it far better than any human. For context, there have been computer-only trading applications since the 1980s (using APIs), which were called “Black Boxes” so this is nothing new. In fact, the market crash in 2010 was a direct result of several automated trading computers being put online so these can be powerful! It’s just so exciting that these things are available now for a Joe Schmoe like me who has enough time, knowledge, and interest to invest in figuring them out.