When I first found out about Bitcoin I was drawn to three main aspects of it.  The mining, the economics, and the investment / trading potential.  The premise that you could earn money passively by running your computer 24/7 is very intriguing but there are a lot of things you have to consider before you jump in and attempt to mine these coins.  

Your Motivation

Some people are just doing it as a challenge or as a non-profitable hobby.  They possibly even want to just be apart of securing the network.  I see no problem with approaching it this way.  When I first started I was mainly mining to be apart of the process and to learn from it.  I started mining Quark because at the time I was doing the math on Bitcoin ASICS and it seemed like I would have to put a decent investment into it for it to be worth it.  I was constantly looking at videos on YouTube of people showing their mining farms.  The general thing I would come up with in my head was "If they are mining on this level it has to be very lucrative." 

In the process of figuring out what I was doing I was having a lot of fun watching Quark go into my wallet.  This was so cool to me.  I was mining on 3 laptops at the time. Quark was still cpu mine-able and there was no GPU miner publicly available. Around Thanksgiving of 2013 the price of Quark went "Super Nova!"  My $40 of coins turned into over $700.  I was hooked and really felt like I need to step up my game.    

When mining turns into something you are going to make a serious investment into and your motivation is to be profitable that is when you need to be very careful.  

Scaling Up

The difficult thing with cryptocurrency mining is that you can't really find out about Bitcoin one week and then have a mining farm the next week or even the next month.  The whole thing takes intense studying and research to understand what to do.  Around this time I had a friend who ended up buying a Bitcoin ASIC.  He wasn't even paying power because he was living with his parents.  Even despite that the ASIC still ended up being a bad investment.  The difficulty was shooting up so fast that the payoff date was getting pushed further and further out all the time.  This wouldn't be such a big deal if the price would continue to make huge increases but nothing that good lasts forever.  

I was still hung up on the Quark situation and it seemed like there were all kinds of good things going on with that coin.  I was able to find some cheap lots of computers on www.PublicSurplus.com.  They were older machines but I researched enough to know that if I scaled up to running GPU rigs that those machines should work even though they were Core 2 Duo machines with PCI-e 1.0 slots.  I got these machines for about $45 a piece.  I also got 6 workstations that had Dual Opterons in them.  I really scored big when my girlfriend at the time was able to get a couple end of life machines from her work for $50.  I got a 8 core Xeon machine with 12 GB of ram and a nice flat panel display.  For $50 it was a steal.  I actually am writing this post on that machine.  It is one of the only ones I still have.  

Having this many computers as a computer nerd was pretty awesome even if it wasn't the best route to take.  Here is a picture of about half the machines I had at my ex-girlfriends house.  

(Credit: Original picture of several of the computers I was running in late 2013.) 

One of the main reasons I got these machines was because I looked at the market for them and I got them cheap so if suddenly crypto was worth $0 in a week I could turn around and sell them as personal computers for more than I paid.  

It was around this time I decided to start messing around with GPU mining.  With some of the machines I bought from www.PublicSurplus.com there were some video cards.  One was an old Nvidia card that was CUDA capable.  I started experimenting with it by mining Scrypt coins.  I would also utilize one of the cores of the CPU to mine Quark.  I was trying to figure out if that would increase the efficiency of the machine since the GPU mining only needed 1 core to operate.  

Here is a picture of the old Nvidia card mining DOGE and one core of the CPU mining Quark as an experiment.  

(Credit: Original picture of me mining Quark and Doge on old equipment in 2013.)

This was like Nerd Porn to me.  It was so exciting to see coins going into my wallets even though it wasn't worth much.  It was proof of concept and made me dream of having a large scale mining farm. 

My friend was doing intense research as well and crunching the numbers in the mining calculators.  We determined that GPU mining was what we needed to be doing. The Gigabyte R9 270's had a great price / hash ratio.  He got all new equipment and 6 of those cards.  I set the rig up but was having all kinds of bullshit issues trying to get the machine to even turn on with card 5 and 6 hooked up.  It was a mystery of why this wasn't working.  That is one of the things you will run into with mining.  It is all a hack and things won't work and it is hard to figure out why.  

Here is a picture of that rig with 4 of the cards setup.  

(Credit: This is an original picture of my friends GPU rig I setup.)

This was a great experiment and and the rig was profitable.  The only downside was about $1600 was invested into that machine since the board, the power supplies, the processor, and the ram were all new.  

I had gotten all those older machines for so cheap that had a good enough processors and 4GB of ram that I didn't have much invested in the machines themselves.  I had to get 2nd power supplies, video cards, adapters to add second power supplies, riser cables, and some other spitter cables to get everything set up correctly.  

I didn't spend any money on ascetics like the rig casing or anything like that.  I even was using an old CRT monitor I would push around on a cart to plug in and configure the machines.  

My machines were NO FLASH ALL HASH!  

Here is a picture of two of my rigs running in 2014.  

(Credit: Original picture of two of my GPU rigs running)

If you notice the rig which is closest in the picture.  There was a MSI R9 280x and two Diamond R9 270x cards on that rig. I had a lot of different cards not by choice.  I would have rather had a much more consistent workflow but it was so hard to get cards in late 2013 and early 2014.  There was a world wide shortage for GPUs at the time and I had so many orders get canceled on me because re-sellers were out of stock.  

I had a total of 24 GPUs cranking away at the peak of the mining farm.  The entire breaker system in the house was maxed out.  During that time we were in a flux period because the Scrypt ASICS were hitting the market.  

Boy Am I Glad I Didn't Buy Scrypt ASICS

All my rigs were running on Linux installed on USB thumb drives at the time.  Suddenly the numbers weren't looking so hot for Scrypt coins so I started looking at other GPU mineable coins.  I went through a period of time where I was sorting through complete shit coins.  Some didn't seem like shit but some were 100% pump and dump scam coins.  It was really hard to tell what was what at the time.  I mined Maxcoin, Groestlcoin, Mammothcoin and several other completely garbage coins.  Then I found Monero.  At the time the GPU miner was only for Windows so I had to hook up hard drives to all the rigs and install Windows and get them running on that.  

When I first started mining the coin was worth around $3 and it was costing me about $1.50 to mine.  This was pretty solid because it would basically be like getting a Bitcoin for half price.  That is the way I looked at it but then I really saw a ton of potential in Monero and didn't trade it for BTC as it came in.  The price dropped to around $1.50 so it was suddenly a break even to mine vs just buy off the exchange.  That is when you are in trouble because the machines are heating up the house and causing you to try to cool the house so your AC is working harder. 

The summer got hotter and the coins continued to decrease in value. One month in summer 2014 the power bill was over $750  I shut the machines down all except a few machines to do testing on. I was hoping the values would increase but they weren't.  I decided my best bet was to create contacts at the University of Arizona and move rigs to free power situations and then split the coins with the person housing the rigs.  This was a really hard situation to find.  I found one guy who took on 3 rigs but we had so many connection issues in the dorm that they never hardly mined any coins.  It was so frustrating.  

The majority of the rigs went to a free power situation up at Arizona State University.  This was a very bad setup.  The guy who was housing those machines was an absolute idiot which I didn't really find out until later.  

Here is a picture of the two rigs from earlier installed at their new temporary home.  

(Credit: Original image of two of my rigs at Arizona State University)

If you notice the rig on the pink crate was changed slightly.  Instead of having the MSI R9 280x on it I put an R9 290x into the equation.  It is the one laying on its side on the far left.  As I mentioned before the guy who was housing the machines was beyond idiot status.  That first night I went out into the living room to configure one of the other rigs and he somehow broke one of the fans on one of the R9 270x.  He then left the house and I walked back into that room to hear the fan completely destabilized. 

This was me after seeing the broken blade.  

I called the idiot up and asked him what happened and he pretended he didn't know what I was talking about.  I knew 100% that he broke the fan blade.  There was no other explanation.  He eventually did admit to it and acted like he would pay for it.  That never happened.  Ultimately one of the R9 290x's got broken too. That was $650 down the drain. Also one of the other R9 270x's suddenly was only VGA capable and couldn't accelerate.  It was FUBAR.  

It was a really bad situation that could have been completely avoided.  I then learned later that this same idiot keyed his name into his friends dad's Suburban, had been in about 6 car wrecks because he was an idiot, and also was shop lifting from Home Depot.  

I WAS LIKE WHY IN THEEEEEEEEEE FUCK WASN'T I TOLD ABOUT THIS BEFORE I MOVED MY PRECIOUS COMPUTERS UP THERE????? 

Here is a picture of one of the other rigs that went up there.  This one was called the "Green Machine" because I put it in a Waste Management trash container.  Again....No Flash All HASH!  

(Credit: Original picture of my mining rig in the closet.)

Specific instructions were given to not close the door on the closet.  That probably didn't happen.  Luckily these cards didn't get burnt up.  

Here was another one of my rigs sitting in the corner of the living room up at ASU.  

(Credit: Original picture of one of my rigs running at ASU.)

Ultimately the price of crypto continued to not perform well and I was under so much financial pressure that I was forced to sell out of all my equipment.  It would have been nice to keep one rig to mine with if the prices surged again but I couldn't afford to do this. I had planned on depreciating the equipment over a 3 year span of time but I couldn't withstand the financial pressure to keep it going for that long or longer.  It was sad but I had taken the risk and took a beating.   

I got to a point where I couldn't even bring myself to look at Coindesk or CoinTelegraph.  It was too depressing to look at the crypto news.  I would allow myself to look at CoinMarketCap once a week and that is when I saw STEEM.  Boy was I glad I saw that!  Now I'm back Baby!  

The roughest part of the mining experience was that I dedicated my entire farm of 24 GPUs total to Monero and held on for a long time but eventually had to sell it a year ago at $0.50 - $0.55. A lot of stuff can be a waiting game and if you can't afford to wait then you might never benefit from the mining.  

Things to Consider

1. Are you technical enough to mine? I saw a couple friends try to mine that weren't technical enough and they were banking on me helping for free.  That can become an issue.  
2. How much does your power cost? If you can't get power for $0.06 / kWh or less I would seriously consider not doing it. Places in the United States where this would be possible would be Universities, Fargo North Dakota, Eastern Washington, Eastern Oregon. My power was over $0.10 / kWh. That will only be profitable in times of extreme price surge.  
3. Can you do a consistent work flow?  You really need the same setup replicated many times which will cut down on individual tweaking.  All cards, all risers cables, and all entire setups are different.  This can be an administration nightmare.  
4. Is Bitcoin or the coin you are mining in a bubble? This is really hard to tell but assume that Bitcoin will drop to 1/4th of whatever it is when you are doing your mining calculations and ask yourself if you can endure a price drop like that for years. 
5. Can your equipment be resold if Bitcoin goes to $0? If you are reading this then it is unlikely you should select to mine with ASICS unless you have 100% free power.  Even then it is tough because you are competing with Chinese miners taking the equipment that was just built and they start cranking away immediately with very low power cost.  If prices dip at all you may never get the money out of your equipment because difficulty could still rise and render your ASICS useless.  That is why I feel you should stick with GPU mining.  Everyone has a CPU.  The barrier to high end GPUs isn't accessible to everyone and they can be re-purposed to mine other coins or resold for gaming or other work.  
6. Can you endure 3 years of not seeing a return? Even though 6 months is an eternity in crypto the very large gains happen over years typically.  We saw this with Monero.  We have been waiting for two years for the GUI wallet.  It still isn't here but then the price has surged clear up to $10.  The miners who were able to hold on finally can take some profits off the table.  
7. Can you afford to lose 100% of your investment and your time?

Things to Note

1. If the coin you are mining is CPU mineable there is likely a secret GPU miner out there.  If it is GPU mineable there is likely FPGAs or secret ASICS out there mining.  At the very least there might be a more advanced GPU miner out there people are secretly using.  
2. The market will naturally push all profit margins toward $0.  Business is brutal and it is tough to make a buck.  It is not different in crypto mining.  If the profits look fat it won't last forever and you have to be ready to move to the next coin or make the right strategic moves.  
3. You are up against the worlds brightest people when you are mining crypto.  
4. You might get rich or you might go broke.  Maybe somewhere in between.  It is unpredictable.  
5. You are speculating on computer hardware and the price of crypto when you mine.  Computer equipment is almost always a depreciating asset.  You know that is becoming worth less and you are hoping the crypto currency becomes worth more.  That is a tough game to play.  

Final Thoughts

The decision to mine and for what reason and on what scale is going to be determined by each individual and their risk tolerance. I don't regret mining.  I took the risk and it didn't pan out.  That's life.  Some miners hit it big right away but others were able to make their money on a 2nd or 3rd wave of profitability.  For instance some of the Bitcoin miners who were mining on GPUs didn't score big on Bitcoin but were able to utilize their hardware to mine the Scrypt coins like Litecoin and DOGE.  Some of the GPU Scrypt miners didn't hit it big until Etherium.  I would love to get back into mining on a large scale but for it to happen it certainly won't be in Arizona and I would need to have investors who can withstand the dips for years.  Hopefully this helps and gives insight into whether you should mine or not.  Your best bet might be to just speculate on coins off the exchange or blog on a platform like Steemit for your crypto fix!  

Thank you for reading my post and please consider following me @brianphobos