Topic: Analysis of Realistic $ Earnings From Butterfly Labs Mining Hardware - page 2. (Read 17091 times)

Difficulty *is*, to an extent, predictable.

I'm sorry but you can only claim any validity to your extrapolation if difficulty was predictable over time.
It is not and so your whole story is cool, bro.

Or otherwise please show why difficulty must progress along your projections.

Not really. Forecasting bitcoin prices is impossible to do with measurable accuracy. Forecasting the network hashrate and difficulty is possible to do with some accuracy. I do weekly forecasts here: http://organofcorti.blogspot.com.au/search?q=%22weekly+forecast%22

The four week forecast network hashrate has been quite accurate recently:

Avalon just sold a batch of 10000 individual ASICs to DIY PCB builders, with others on the way. Expect difficulty to rise hard and fast to dingledongle BFL customers in their buttocks.

VERY VERY eventually .... a lot witht the price going up had to do with the recent interest in the BTC currency and is not essentially in correlation with the difficulty increase.

jhansen858, its anyone's guess.

An 18% increase was based on current difficulty lagged from the boom market which may prove to be greatly exaggerated.  If prices continue to fall or stabilize at a low price, then the rate of increase in the difficulty levels will definitely reduce.

If prices of bitcoins drop further then miners will leave the market and the rate of difficulty levels will drop as a result, as difficulty levels are automatically adjusted depending on the total mining success in the previous period.

But this takes us back to profitability levels for mining rigs from butterfly labs which would have to be reworked taking into account reduced profitability from reduced earnings due to the lower price of bitcoins.  We may see linear rather than exponential levels in difficulty.

So forecasting difficulty levels is about as difficult as forecasting bitcoin prices!

It will be interesting to do same analysis in a months time based on the then current difficulty level changes.

A lesson here is that anyone considering buying butterfly rig that they should avoid placing an order unless they have first taken into account the fact that the butterfly labs calculator shows a snapshot in time and ignores changes in difficulty level over time.


Here is a graph of historic difficulty (courtesy of bitcoin.sipa.be).





Here is another chart of interest. In the next chart in the top graph the difficulty level is shown in red.

This puts the relationship between price and difficulty in a good perspective showing that difficulty levels do also eventually adapt to the price of bitcoins (shown in blue).

You really think a difficulty of 2 Billion is realistic?  I don't see it going that high by then.  What what do i know.  The GPU thing made it spike for a while then it kinda leveld off and even went down for a second. 

Thanks for your work, ripple.


I hope they delete this calculator asap - because otherwise no one would order their products any more 

I would like to try and simplify calculation of the breakeven point for people who might have ordered a mining rig from butterfly Labs (BFL).

I don't think that the increase in hash rate from other ASIC miners is particularly relevant to calculations of breakeven. This is because of the much bigger hash figure created collectively from every Tom Dick and Harry switching on their computer at night to make Bitcoins. Now that Bitcoin as has become so well-known there must be millions of teenagers whose parents pay for electricity that will quite happily run their computers with a couple of graphics cards in their bedroom 24/7 who collectively become the new competition.  There will even be some IT managers sneakily running their vast departments the same way at night.  

I don't think it's unreasonable to assume that with all the publicity generated about Bitcoin in the media that in the next few months there will be several million additional home or business computers switched on overnight worldwide to mine Bitcoins just using the CPU and Graphics cards.  The total computation output of 1 million home or business computers at say 50 MHs is equivalent the output of 10,000 Jalapeno rigs, or equivalent to 1,000 rigs at 50 GHs. Even if they don't make money a valuable lesson may be be learned by home computer owners is that leaving their equipment on, whether mining or idle is costing somebody money!

In the breakeven calculation, the more important figure is the difficulty level. This level is now 7,672,999 and increases every two weeks by about 18%. The latest change from 7,672,999 to 9,121,259 is 18.88745%.

This analysis shows you a method to calculate profits by adjusting the difficulty for each month. You can decide your own rate of increase in difficulty and adjust the present publicly published figure of 18% up or down.

I have compiled the following table showing the difficulty level extrapolated from present figures:

Date            Difficulty Level
---------------------------
17/04/2013    9,121,259.40
01/05/2013    10,842,874.22
14/05/2013    12,889,439.52
27/05/2013    15,322,288.88
09/06/2013    18,214,332.46
22/06/2013    21,652,242.00
05/07/2013    25,739,048.34
18/07/2013    30,597,229.13
31/07/2013    36,372,379.34
13/08/2013    43,237,574.67
26/08/2013    51,398,558.38
08/09/2013    61,099,907.29
21/09/2013    72,632,361.46
04/10/2013    86,341,537.41
17/10/2013    102,638,285.93
30/10/2013    122,011,004.83
12/11/2013    145,040,275.80
25/11/2013    172,416,263.88
08/12/2013    204,959,401.01
21/12/2013    243,644,973.60
03/01/2014    289,632,350.94
16/01/2014    344,299,730.35
29/01/2014    409,285,440.44
11/02/2014    486,537,040.22
24/02/2014    578,369,685.59
09/03/2014    687,535,512.33
22/03/2014    817,306,114.92
22/03/2014    971,570,593.08
22/03/2014    1,154,952,104.36
22/03/2014    1,372,946,415.71
22/03/2014    1,632,086,606.27
22/03/2014    1,940,138,857.49
22/03/2014    2,306,335,198.06
22/03/2014    2,741,650,178.95

If we now assume that butterfly labs will deliver in July.  Let's say 31st of July (so they won't be lying), then the average hash rate per month can be calculated for a Jalepino 5GHs rig as an example.

Month 1 is August 2013.

Month   Average difficulty    Earnings
------------------------------------------
 1      39,804,977.00      $186
 2      56,249,232.83      $131
 3      79,486,949.44      $92
 4      112,324,645.38      $65
 5      158,728,269.84      $45
 6      224,302,187.31      $31
 7      316,966,040.65      $22
 8      447,911,240.33      $15
 9      632,952,598.96      $10
10      894,438,354.00      $6
11      1,263,949,260.04      $4
12      1,786,112,731.88      $2
13      2,523,992,688.50      $1

As you can see the earnings diminish rapidly due to the difficulty level increasing, and are based on a bitcoin price of $100.

The above earnings figures are taken from the profit calculator on the butterfly labs website, but importantly I have used a separate calculation for each of the twelve months, using increasing levels of difficulty for each month which their calculator surprisingly omits.  I contacted a representative from butterfly labs about this glaring omission from their calculation page on the website and their response was that the increasing difficulty was a perfectly realistic assumption, but that it should be entered into the calculator by buyers for themselves, and it was not the responsibility of the company to do this.  In my view their calculator is misleading and misrepresents profit potential as the average consumer would use the existing difficulty level and believe the unlimited riches promised by the company.

Profitability will clearly depend on when you receive delivery. The following are three scenarios:

1. End-July Delivery.

The total earnings are $609.  The basic cost of the unit is $274, or $312 if you are shipping to the United Kingdom by standard air mail. So the profit for the year is $297 if you received at end July and start mining from 1st August.

2. End-August delivery.

Now let's say you are unlucky to receive your unit at end of August. Your estimated earnings for the first year would then be $423 less $312 = $111.

3. End-September delivery.

If you received your unit at the end of September earnings would be $292 less $312, and you would make an annual loss of $20 !!!

So the crux for miners buying from butterfly labs is when they will actually receive the unit.

The website states: "Pre-order Terms: Bitforce SC (ASIC) products are in final stage development with initial shipping scheduled for the last half of April 2013. Products are shipped according to placement in the order queue, and delivery may take 2 months or more after order. All sales are final."  Their voicemail service by contrast recently stated shipping will commence July 2013, but now has simply states "soon".

If you placed your unit early and receive end-July, you might be lucky to double your money.  If however you ordered later, or the company delays orders further, and you receive end-September or later your earnings may be squat!

I hope this helps people who are thinking of buying the unit from Butterfly labs.

It seems the trick to staying ahead is by keeping up-to-date with the latest technology, already having existing fast hardware, and knowing the right timing to incrementally upgrade, in order to keep pace with the difficulty level.