Mining Economics

[willsterling]

A month ago, I ordered a triple 6990 mining rig. At the time, the payout on 2100 Mh/s looked to be about $1900 per month, so the cost of over $4,000 (including taxes) looked liked it would recoup itself in just over two months.

Shipping was delayed for weeks because of scarcity of 6990s. I finally received the unit and got it set up yesterday. In the intervening time, the economics of mining have changed dramatically. Difficulty has increased and prices dropped. One would expect a machine running 2100 Mh/s to generate about $600 per month. Of course, it doesn't run 2100Mh/s 24/7, so a more realistic number is $500 per month.

At these returns, electricity becomes a major factor. The rig draws about 1.2kW, but also included in the electricity is the air conditioning costs at perhaps another 500W -- so 1.7kW per hour, or 1,300kW hours per month. California has tiered electricity pricing for people who use above the "baseline" average household use, so one could expect to pay at least $260 per month for that electric usage. (If you just paid the average cost for electricity in the US, you'd be at around $190 per month in electric costs.)

Net payoff for Bitcoin rig is now $240 per month. Break even in... about 17 months. Of course, a lot can happen in 17 months. During that time, difficulty will only increase, and my 2100Mh/s in processing power will be static. My equipment can get fried by running it 24/7. The only way it continues to make $240 per month is if I have no breakdowns and bitcoins increase in value faster than the difficulty increases. These are two pretty shaky propositions to invest in.

What do you all think? I'm sure many people thought mining looked like easy money a few weeks and months ago... but do you still think it? Who out there is placing new orders for 6990s? If you thought bitcoins were going to increase in value, wouldn't it be better just to buy bitcoins?

[bitebitebite]

wouldn't it be better just to buy bitcoins?

 Has been like that for a while, I cancelled a build ~ a month ago after running some very rough numbers while keeping in mind increasing difficulty. In fact Im going to cancel mining on my already bought 6990 soon, as imo a few $ a day just isnt worth the time/heat/noise. Bought some and am trading (which dosent really seem like its worth my time so far either, unless I put in a serious amount of cash, $4000.00+ like you mention). My opinion may be different if I was a kid and living on pocket money etc.

[nmat]

Currently, mining hardly compensates. You are better off buying bitcoins directly. Regarding your setup, you can always sell it when you don't want it anymore. Hardware also has its value.

[The_JMiner]

Currently, mining hardly compensates. You are better off buying bitcoins directly. Regarding your setup, you can always sell it when you don't want it anymore. Hardware also has its value.

This is why I think mining still has it's value because you can always sell the hardware later to recoup some costs.

[myjack]

Well if everyone sell their graphic cards, the value you will get for it will be low as hell.

[gen. specific]

The idea was to generate bitcoins from "thin air" without incurring overhead costs

Everyone who gets a machine would have thought about the overhead costs like electricity......... you would assume.

The value of bitcoins is relative to their availability and the strength of a basket of currencies. But from what I can tell, the liquidity of bitcoins is not thick enough to make this an efficient market for an exact science.

Can you tell me the exact specs of your machine? I was previously considering one. PM me if you must. What power supply and mobo are you using?

[99Percent]

Electricity and weather conditions are a big factor in mining cost effectiveness.

Those who live in cold weather or can either get free aircon or electricity (in an office or school for example) have a huge advantage.

[Forte]

College kid.  I don't need to pay for power or AC because it is covered in the dorm fees.  Milking it for all it's worth right now.

 

[coinvestor]

difficulty will only increase

False...I expect it will go down in the next couple of days

[willsterling]

difficulty will only increase

False...I expect it will go down in the next couple of days

Where would you get that idea? http://bitcoinwatch.com/ shows that the next change in difficulty in about 10 days will be a modest increase. As long as the ac and electricity costs don't yield a negative return, I doubt that many miners will go offline.

And if difficulty did decrease temporarily, it would just encourage retired miners back into the market... Any drop would reverse pretty quickly.

[Einewton]

No way the Diff level will drop. For every retired minter, there are 4 to replace him. I know at least 3 people that are spinning up mining operations and one of them that has substatial funding behind them.

[99Percent]

College kid.  I don't need to pay for power or AC because it is covered in the dorm fees.  Milking it for all it's worth right now.

 

That is great. I wonder though if it gets to be abused, campuses are going to start banning mining rigs...

[rnicoll]

That is great. I wonder though if it gets to be abused, campuses are going to start banning mining rigs...

I believe per-room or per-block of room metering is on the to-do list where I work.

[st4rdust]

My equipment can get fried by running it 24/7.

Correct me if I'm wrong, but I thought it was less stressful to keep a computer running constantly instead of turning it on and off. Something like the constant, steady voltage is less stressful than constant fluctuations.

[coinvestor]

Didn't mean a couple of days. Allow me to familiarize you guys with a term I coined. I believe that the next bitnight will be longer than 14 days. A bitnight is the amount of time it takes to solve 2016 blocks. There's no proof that for every miner that drops, three come on. The numbers aren't showing that at all right now.

The network hash is just barely greater than the target network hash. Every person entering the mining operation right now is probably doing it based on long-term investments. Now is conceivably the best time to enter since before June.

Admittedly, a small increase in difficulty is better than none at all in my opinion. Increases prove the speculative value. I just think the overall health of the economy isn't going to support an increase during this bitnight phase.

Coinvestor (Ryan)

[willsterling]

My equipment can get fried by running it 24/7.

Correct me if I'm wrong, but I thought it was less stressful to keep a computer running constantly instead of turning it on and off. Something like the constant, steady voltage is less stressful than constant fluctuations.

That may or may not be the case with a typical CPU (i've heard reasonable arguments from either side) but when you're running 3 6990s at a time, they generate a lot of heat, with a lot of moving parts and there are a lot more things that can go wrong. Typical desktop computers are reasonably low-stress environments. When I've got this thing cranking, it sounds a bit like a jet engine. My guess is that it won't work flawlessly, continuously (what does?)... I'd love to be proven wrong, though.

[Pentium100]

Correct me if I'm wrong, but I thought it was less stressful to keep a computer running constantly instead of turning it on and off. Something like the constant, steady voltage is less stressful than constant fluctuations.

It all depends.

When you turn the computer on and off:
1. It does trough a temperature cycle, which can result in cracked boards etc (because they expand then contract).
2. Hard drives have to spin up, which causes more strain on the motor than just spinning; On the drives that park their heads on the platters, the heads and the platters wear while the speed is not enough for the heads to be lifted away.
3. Incoming current surge (to charge the capacitors) can in time damage other components.

When you leave the computer on 24/7:
1. The temperature inside the case is higher, which can result in failure of capacitors (they live longer at lower temps) and other components.
2. Hard drives and fans are spinning and will wear out eventually.
3. A surge (if you do not have protection devices) can damage the power supply and the rest of the components.


Same thing happens with, say, light bulbs. Turning it on causes a lot of stress on the filament, leaving it on causes lower, but continuous stress. I wonder what the duty cycle (>0) for the longest life would be...

[Oldminer]

Currently, mining hardly compensates. You are better off buying bitcoins directly.

Yea well trading is currently more appealing IMO. Lets say with good trading tactics you make an average profit of 0.5BTC or (at current price) $7/day. Thats equivalent of running a 630 MHash/s rig 24/7...

[RandyFolds]

Currently, mining hardly compensates. You are better off buying bitcoins directly. Regarding your setup, you can always sell it when you don't want it anymore. Hardware also has its value.

This is why I think mining still has it's value because you can always sell the hardware later to recoup some costs.

Who actually thinks that they can sell their used out-of-date video card for any significant return on their investment? If you are selling your top of the line card, anyone computer savvy in the least will know that card has been beat to shit and isn't worth the gamble. Heck, search the forums here and see how many people were boned on their eBay GPU purchase...

[patvarilly]

Currently, mining hardly compensates. You are better off buying bitcoins directly.

Yea well trading is currently more appealing IMO. Lets say with good trading tactics you make an average profit of 0.5BTC or (at current price) $7/day. Thats equivalent of running a 630 MHash/s rig 24/7...

People, don't be fooled, this is crazy: trading in bitcoins is sheer speculation, and you'll just as easily loose a good amount of money doing this: it's impossible for every single one of you to make $7/day, even if you all read and follow "Oldminer's Good Tactics for Trading" manual.  Bitcoins, as they currently stand, aren't a viable vehicle for trade, just for speculation.  This has been discussed at length before by level-headed people since the project's inception, and keeps being discussed again and again nowadays, e.g.

http://forum.bitcoin.org/index.php?topic=57.0
http://forum.bitcoin.org/index.php?topic=24821.0

Bitcoin's technical underpinnings are wonderful, and many of the problems it tries to solve are worth solving.  I used to be enchanted by these aspects, and still am.  But don't be fooled: it's also stupidly tied to a fringe economic model that will prevent it from ever taking off as a mainstream virtual currency.  Bitcoin's value is sustained by lots of new patsies coming in, people thinking that they're going to make money out of thin air, when all they're doing is allowing the older ones to slowly cash out.  It will crash when not enough people keep rolling it. Do yourself a favor: don't do it.