A Reality Check

[AngelusWebDesign]

5830 -- the favorite card for price/performance -- makes 0.26 BTC per day according to Deepbit's estimator (pay-per-share). Proportional would be about 7% higher.

This isn't an unrealistic estimate, as Deepbit is currently almost 1/2 the Bitcoin network hashing power, so many people are obviously using them.

The current price of BTC as I write this is $13.58 (not $20, and certainly not $30!)

Of course, this assumes you can *get* a 5830. They are out of stock in most places, and when they come back in stock it's like Black Friday trying to get your hands on one. They last about 10 minutes once word gets out. And if you can't get a 5830, you won't get the same price/performance.

.26 BTC/day is exactly $3.53 per day -- without subtracting electricity, which is $0.42 for the 160W card alone (assuming an electric rate of $0.11/KwH). The amount your PC uses will vary, but it's going to be at least another 60W (or $0.16/day). So that leaves $2.95 after electricity costs are subtracted.

Now overclocking might result in SLIGHTLY higher profit per day (though you risk shortening the life of the card & fan), and some youngsters have free electricity -- but that gives you a rough idea of "how much can you make today". Not exactly a cash bonanza.

How long would it take to pay back the card itself, much less a card plus mobo, CPU, ram, etc.?

I think the manic phase of "I *gotta* expand unless I'm stupid" is over. At least for now.

Sure, some people would do anything for a few (as in, not even 3) bucks. But it's not like decent amounts of money are materializing out of thin air with no effort. You *do* need to tie up a machine for 24 hours to get that $2.95 ($3.53 for those who still live with Mom), pay for A/C to counter the heating effects, endure the noise/heat the machine produces, give up gaming/outdoors/family, mess around with Linux/Windows/overclocking/dummy plug headaches, etc.

And then you have to worry about downtime/DDoS/outages -- you have to make sure your miners are always mining.

And then there's taking that $2.95 a day and turning it into cash in your wallet. Once you decide to sell, you need to get a good price. When you do find a price you're willing to live with, Mt. Gox charges 0.65% for the trade into dollars. Then you have to get it into your hands somehow. Either pay someone a % to a middleman to get gift cards, precious metals, etc. or you can just transfer it to Dwolla (indeed that costs just 0.25 per transfer -- at least something is easy!)

And, last but not least, there's the periodic Difficulty Level resets every 8 days or so. Each time it happens, you lose a hefty chunk of what you were making before. So you can't multiply $2.95 by 30 to see what you'd make in a month. "$2.95 a day? That's $88.50 a month. Almost a free video card!" Yes, if difficulty held steady, which it HASN'T and WON'T in the future. Don't talk about how "difficulty dropped once" because that was a long, long time ago well before most people found out about Bitcoin.

Yes, you ONCE could buy a video card and a cheap rig and pay it off reasonably soon. But that was then. I think it's too late to go out and buy new hardware today -- unless you're already sitting on some partially paid-off rigs (as in, 2 GH/s worth at least). Even then it's risky. The Big Bad Drop to 25 BTC per block is closer than many people think. When that happens, anyone who hasn't paid off their hardware NEVER WILL. So be careful.

Mark my words: difficulty will continue to increase, even if all the small-time miners drop out. As long as you make *anything* over the cost of electricity, the big boys will find it worth their while to keep adding capacity -- their existing hardware is already paid for. When you have 100 paid-off video cards, you can buy a few new cards and pay them off within hours using your existing hardware. It's the "snowball rolling down a hill" effect.

Just like Wal-mart can drop prices to the basement and drive all the small Mom & Pop shops out of business -- then take all the customers, and the profits that go along with it. Call it fierce competition, the law of the jungle, unbridled capitalism, or whatever you want.

Just a bit of a reality check.

Matthew

[1715628453]

1715628453

[Mousepotato]

Except a lot of us (myself included) don't even do this for the money, although it's a nice secondary benefit.  Some of us are gadget freaks and just enjoy building systems and doing something other than playing WoW 24/7.

[_s3v3n_]

Yes, you're right about even these small miners stopped, it won't prevent the rise of the difficulty because they weren't even the reason it went up in the first place.

But honestly, all we can do right now is *hope* that this difficulty stays on 800k+ for at least.... a month or two. At least on my part, I can break-even on the hardware cost that I just spent.  

[Adam]

No one can really tell how rational the market will be when difficulty is approaching the point where GPU mining is breakeven.  It's possibly that big miners will keep adding capacity, but it doesn't make any more sense for someone with 100 paid off computers mining to add 1 more than it doesn't for someone with zero to buy 1.  Either way that specific computer earns the same rate of payback.  If people are rational they will do a cash flow analysis on every piece of equipment before purchasing it, factoring in all direct and indirect costs, and never take a negative cash flow investment.  

However you could also get in the situation where people continuing purchasing even though they are losing money, or have the difficulty propped up by a bunch of 14 year olds who find out late about bitcoin and run their computers 24/7 at a loss because their parents paid for their electricity.

Conversely you could get in a situation where people are too risk adverse after the first shock wave to add on any more hardware (the stuff that came online in the past week or so would have been ordered during the run up to $34/BTC) and actually have GPU mining profitable for a long time to come.

[game4life]

Bitcoins is a hobby, i bought 1 6970 that i use to mine and play games. I am very satisfied at the .5 bitcoins i make per day. i can buy a video game once a week if i feel like it, thats good enough for me.

[AngelusWebDesign]

No one can really tell how rational the market will be when difficulty is approaching the point where GPU mining is breakeven.  It's possibly that big miners will keep adding capacity, but it doesn't make any more sense for someone with 100 paid off computers mining to add 1 more than it doesn't for someone with zero to buy 1.  Either way that specific computer earns the same rate of payback.  If people are rational they will do a cash flow analysis on every piece of equipment before purchasing it, factoring in all direct and indirect costs, and never take a negative cash flow investment.  

However you could also get in the situation where people continuing purchasing even though they are losing money, or have the difficulty propped up by a bunch of 14 year olds who find out late about bitcoin and run their computers 24/7 at a loss because their parents paid for their electricity.

Conversely you could get in a situation where people are too risk adverse after the first shock wave to add on any more hardware (the stuff that came online in the past week or so would have been ordered during the run up to $34/BTC) and actually have GPU mining profitable for a long time to come.

See my Wal Mart reference. You think Wal-mart wouldn't operate at a loss for a while, to drive the Mom & Pop stores out of business (who CAN'T compete in the same way; they don't have the money or resources -- they're not "big enough")

And a big operator adding a new machine to increase his bottom line by 1% would certainly do so -- while an individual might not be "set up" to have that many machines yet (space, cooling, finances, etc.)

Speaking of which, HOW did you get to 5 GH/s? How many addresses are you mining from? And how long have you been mining? That's a lot of $ in hardware/power draw/cooling need/space taken/etc.  As they say, that's a lot of iron.

[shakaru]

Here is the REALITY of the situation.

I have 1600 in hardware purchases for the past month. Ive made over 3k in btc. It works.

[Adam]

1. See my Wal Mart reference. You think Wal-mart wouldn't operate at a loss for a while, to drive the Mom & Pop stores out of business (who CAN'T compete in the same way; they don't have the money or resources -- they're not "big enough")

2. And a big operator adding a new machine to increase his bottom line by 1% would certainly do so -- while an individual might not be "set up" to have that many machines yet (space, cooling, finances, etc.)

3. Speaking of which, HOW did you get to 5 GH/s? How many addresses are you mining from? And how long have you been mining? That's a lot of $ in hardware/power draw/cooling need/space taken/etc.  As they say, that's a lot of iron.

1. I don't think Wal-Mart is analogous because people can simply shut off their rigs if they're losing money, they don't have to sell them.  Sabotaging yourself just to put someone else out of business couldn't possibly be a viable plan with BTC unless you could somehow end up with over 50% of the hashing power and control the network.  Other than that being a big player doesn't really have any advantage like economies of scale, so it wouldn't be worth losing money to get there.

2. Your bottom line is net income.  If you do things that are negative expectation it hurts your bottom line.  Just because someone could afford to burn money doesn't mean they would.

3. I've never heard anyone say "that's a lot of iron" but right now I think I have 17 5830s online.  It would be a lot more, but GD70s are pieces of shit and I've had 3 arrive DOA so I have a lot of idle capacity.

[beeph]

Truth is i went gung-ho an set up a buncha rigs myself.. and i'm regretting it already.  I did a spreadsheet with minor 20% difficulty increases every 2 weeks (which is very optimistic) at 20$ a bit coin, and its a loss.  And if bitcoins get hot again, and prices shoot up, we'll just get more 30%+ difficulty increases.

The fact is there's enough kids with graphics cards who will run the miners part time at no electrictiy cost (college dorms) to overwhelm the current network of about 30,000 GPUs.  I'm in chicago and paying 15 cents per kilowatt hour.  What ends up killing the profability is electricity, depreciation of equipment, and difficulty increases.. not so much the value of bitcoins themselves.  If you do believe in bitcoin appreciate then just buy the bitcoins themselves.

My partner has dropped out and if anyone is in the chicago area and wants to buy some 6870's please private message me.  It's not a realistic business plan to compete against a miner who's 'happy to pay off part of his graphics card to play the next cool game'.

On the plus side i got all rigs running and fully operational, very sweet set up, i may take pics.

[mjsbuddha]

$2.95 a day means you paid for the card in 1 month. everything it makes after that is pure profit. and most people building mining rigs have at least 4 cards so thats almost $12 a day or $360 a month for sitting on your ass.

[Cluster2k]

$2.95 a day means you paid for the card in 1 month. everything it makes after that is pure profit. and most people building mining rigs have at least 4 cards so thats almost $12 a day or $360 a month for sitting on your ass.

$2.95 a day, but as the OP stated that gets reduced by a third every 10 days or so.  By the end of the month you're getting $1, not $3.

[Adam]

That's one possibility of what will happen, but we've yet to see for sure whether difficulty will continue to scale that quickly.

[beeph]

common sense would dictate that the price will quickly move to the lowest common denominator which is someone living where electricity is 'free' (dorm room, IT techs, kids in basement) etc, with people who dont factor in depreciation at all (gamer who would buy the graphics card anyway), and someone who values his time at 0$/hr (an enthusiast, someone who enjoys working on gaming rigs).

Factor in these 3 costs, 0$, 0$, and 0$.. and it takes very little to keep some sizeable hashing power online.  Consider 'folding@home' which reached 8 Petaflops with NO monetary incentive.  And I think you can see the future of bitcoin mining...

And people dont realize how quickly exponential decay destroys margins.

Please, if you live in the chicago area I have plenty of high end mobos, power supplies, radeon 6870's, i will sell at reasonable prices. 

[AngelusWebDesign]

$2.95 a day means you paid for the card in 1 month. everything it makes after that is pure profit. and most people building mining rigs have at least 4 cards so thats almost $12 a day or $360 a month for sitting on your ass.

Even if you bought your 5830 for $109.99, you would not get there in a month @ only $2.95 a day. Check the math again.

I specifically said you CAN'T multiply by 30 to get your "monthly income" because that fails to take into account DIFFICULTY INCREASES which have been a fact of life in "this era" which is the one we live in today. We can't go back to April or May 2011, just like we can't go back to 2010 and mine dozens of bitcoins a day with a CPU

The fact is, the genie is out of the bottle, the cat is out of the bag, and Bitcoin isn't a secret anymore. If difficulty went down just a bit, people would hurry up and start mining -- that wouldn't last for long.

Unfortunately for the miners (including myself), I think competition and growth will bring the profit/electricity cost ratio all the way down to 1:1 eventually. If there's money to be made, even only a little, people will take it because, as you say, it's easy money (after you factor in all the headaches!)

Just this evening I had one card that was overheating (=84 degrees or higher) which I had to solve, then while I was working on that my wife's machine (2 5870's) crapped out and "VPU recover" had to recover it -- it took the speed down to 850. I had it set at 875. I had some real cool temps on that rig, too -- don't know why Windows XP had a problem with it. Anyhow, there IS some work involved keeping the operation running. It's hardly "free money". I could spend the same hours doing something else which would probably make me far more. If you're talented enough to build, configure, set up and maintain your mining rigs (including Linux and installing various software) I'm sure you could make more than $50 for a couple hours' work...

Matthew

[perfectchoice4u]

I decided to get into the bitcoin mining business recently, and have other bitcoin projects going.
I should be coming online to about 3.2ghash's in about a week, this cost me $4400

The only problem I see is with scale, and your cost of electricity. But really, the electric only matters as to when u get to your breaking even point. Maybe I'm just lucky my electrics low, 8c a kw hr at peak

The way I see it, its 8 weeks to break even on equipment if bitcoins stay at $15-20 a piece.

And really, when it comes to "breaking even" you also have to factor in being able to sell the machines too. I'm sure i could get $500 per pc with a ATI 6970 in it (yeah i know i should have gotten a different cheaper card live and learn), but thats 8x500=$4,000

So, really I only have to make $400 before I "break even"

Now I got kinda lucky because i had 8 pc's laying around. But u can get a cheap $200 dollar old computer and pay the $550 I paid per cheap Pc (power supply, video card, I got some other stuff put in too that was unneeded). So say total you pay $750 per machine, that you can realistically resell for $500. But lets say you want to be conservative, and say $450 a machine resale on CL, eBay, etc

So your looking at $750x8 (in my case), and your getting 3.2ghash/s. Total you paid $6000. Selling them off fire sale prices you can get $3600 back meaning you need to make only $2400 to "break even", not $6k

3.2ghash's @ $15 a bitcoin @ 877226.666667 is $1,673.09 per http://www.alloscomp.com/bitcoin/calculator.php
$1,673.09 divided by 30 = $55.75
$2400 divided by $55.75 = 43.wholebunchofnumbers days

So if u wanted to step into the market today with $6,000 after 43 days u would be pure profit to the tune of $390 a week

Mind you, I've overstated expenses, and understated earnings and prices u can receive too make WORST CASE SCENARIOS

The matter is scale. 400mhash's is a piss in the pot. I only got 3.2g as a starter to see how it goes, but if it goes well I will be scaling.

[smell]

At the rate that difficulty is growing, once you've broken even in eight weeks, you don't really stand to profit much more... by the end of August you'll have mined 82.4% of the perpetuity.

[skyhigh]

perfectcoice4u, what about the fact that difficulty will change few days after you will get your rig online and running and your daily profit will immediately drop from $55 to $46, plus you didn't include your cheap electricity in your formula, not to mention that you won't be able to run this 100% of the time. So your 43 days just turned into 52 days not to mention the difficulty will increase 3 more times in the meantime, making it hardly possible to ever get all this costs back. All this while assuming bitcoin price stays at $15 while majority is eying $8-10 area in the near future.

[perfectchoice4u]

Its really hard to be able to look into the future with a crystal ball. But, at least with bitcoins, it seems like we have a clearer picture than most "economic" phenomenon

At the end of the day, the price per bitcoin is going to end up pairing with how difficult it is to get them, and how much people want them.
Its simple supply and demand.

So right now, as we speak, there's MORE supply than there will be later. That's a fact.

Assuming the demand stays the same, that alone would increase the price.
The tightening of supply increase, with stable demand = bitcoins +

What will happen is those that give up mining will sell their rigs
Some will sell their rigs for bitcoins + demand

Some will sell their rigs for cash. X amount of those people will convert some of that money into bitcoins to

Continue as part of the community + demand
to simply speculate prices and still try to make money +demand

What number is that X? I don't know. But its more than ZERO. And can only be beneficial for the bitcoin.

[skyhigh]

RIGHT.

[bitcoin0918]

3.2ghash's @ $15 a bitcoin @ 877226.666667 is $1,673.09 per http://www.alloscomp.com/bitcoin/calculator.php
$1,673.09 divided by 30 = $55.75
$2400 divided by $55.75 = 43.wholebunchofnumbers days

This is oversimplified. The difficulty is not going to stay at 877k for the next 43 days. It is likely still going to shoot up by a large percentage every 6-10 days for the foreseeable future.

Also, given the continued deteriorating economies of the western world, it is seeming less and less likely that, in a few months time you would make as much back if you had to sell these as gaming PCs (although would these work as good gaming rigs if only the video card and PSU were upgraded?)