3.5 GH/sec ASIC Mining Noob Question

[WarkSheek]

Hi all,

Long time lurker - first time poster.  I'm still new to the bitcoin community, and am a total noob when it comes to mining though it has certainly piqued my interest.  I've done some reading and research on my own and am hoping someone can take a look at some estimating I've done and let me know if I'm on the right track, or if I'm not considering something, or if I'm entirely off altogether.

I'm particularly interested in the BFL "Jalapeno" unit which is advertised to do 3.5 GH/sec for $150... source: http://www.butterflylabs.com/order-form-bitforce-sc-jalapeno/ - It interests a noob enthusiast such as myself as it provides a much much cheaper barrier to entry at $150 and still has what seems to be a great upside.  Here's where my math comes in - I've taken a look at past difficulty increases and the length between and by how much on average the difficulty increases and am trying to find out how profitable mining with a $150 ASIC is going to be in the long run. 

For the attached image I made the following assumptions:

• The price of BTC/USD would be $10.00, I have no idea how realistic an estimation/assumption that is but that's what it looked like it's been going out at.
• The days between difficulty increases would be static
• BFL claims they will ship the units in October - I assumed the 15th
• The increase in difficulty would be static. 
• The increase in difficulty won't change after ASICS hit the market - I know I'm wrong here, but wasn't sure by how much the difficulties would increase

https://i.imgur.com/L8xtp.png

Am I right to assume at the current conditions I could make ~$91.47 per week mining 24/7 at 3.5GH/sec?
Am I right to assume come October 15th, if I'm mining 24/7 at 3.5GH/sec I would be earning ~$72.61 per week?

In that case I just don't see the downside in pre-ordering one of these if they will literally pay for themselves in 2-3 weeks and then anything else made is pure profit?

Thanks in advance for your guys' help, criticisms, and pointers.

[1714947537]

1714947537

[DeathAndTaxes]

No.

Difficulty adjusts every 2016 blocks which is ~14 days when difficulty is relatively static.

If BFL numbers are to be believed difficulty will more than double just the first batch of pre-orders go out.  So even those who get a Jalapeno on day 0 won't be getting ~$84.69 for the first week.  More like (assuming rest of your # are right) ~$12 per day for the first 2 or 3 days and then difficulty will double. 

However if you buy now you aren't going to get your BFL product on day 0.  Take a look at the BFL Single's delivery thread for an idea of what kind of delays to expect.  If you get your Jalapeno 12 weeks later difficulty will have already tripled or possibly quadrupled by then, also factor in the 25 BTC drop and likely some future difficulty increases much more significant than 8% per adjustment.  If we compare the MH/W and MH/$ to GPUs difficulty could easily be 5x to 10x as high as it currently is within a few months if BFL can keep deliveries going out in bulk.

I am not saying "don't get a BFL Jalapeno" I am just saying your numbers are pie in the sky fantasy land.  Even those who get theirs on day #0 won't generate that kind of return.

[WarkSheek]

Wow - great insight.  I figured I must be off somewhere as it almost sounded too good to be true, and I've learned anything that is too good to be true is just that.

I figured difficulty would increase quite a bit but I had no idea just how much - if you think it should literally more than double then that certainly throws a wrench into my calculations.

Thanks again

[adamstgBit]

i think, when its all said and done, buying a 3.5 GH/sec ASIC rig will pay for its self in 6-12 months. at best

could very well be that it will never pay for itself, and actually cost you 10$ / month.

IMO buying is cheaper / less risky then mining

[Monkey1]

No.

Difficulty adjusts every 2016 blocks which is ~14 days when difficulty is relatively static.

If BFL numbers are to be believed difficulty will more than double just the first batch of pre-orders go out.  So even those who get a Jalapeno on day 0 won't be getting ~$84.69 for the first week.  More like (assuming rest of your # are right) ~$12 per day for the first 2 or 3 days and then difficulty will double. 

However if you buy now you aren't going to get your BFL product on day 0.  Take a look at the BFL Single's delivery thread for an idea of what kind of delays to expect.  If you get your Jalapeno 12 weeks later difficulty will have already tripled or possibly quadrupled by then, also factor in the 25 BTC drop and likely some future difficulty increases much more significant than 8% per adjustment.  If we compare the MH/W and MH/$ to GPUs difficulty could easily be 5x to 10x as high as it currently is within a few months if BFL can keep deliveries going out in bulk.

I am not saying "don't get a BFL Jalapeno" I am just saying your numbers are pie in the sky fantasy land.  Even those who get theirs on day #0 won't generate that kind of return.

This. Main points:

1.  Reward will halve in early Dec (earlier if ASIC hits before then).

2.  BFL are now saying pre orders will start to be shipped in early NOv and they expect to have them all out by Xmas.  If you order now it could be next year before you see anything.

3.  Difficulty will go through the roof when these start to properly hit the network.  Difficulty could well go up by over 10x very quickly.

Therefore, Jalepenos are still worth buying (I have a pre order in) but you cant use todays figures with tomorrows equipment.  ROI is likely to be in the order of months, depending on how quickly they ship the kit and where you are in the order.  Still worth it but not a ROI in 2 weeks!

However, electricity is not cheap where I am so these will be a god send!

[spugyek]

Guys seriously, think a bit...

To forge these ASICs, are about 2M$ . Let's assume they did it. With a jalapeno a breakeven comes in 10 days. They would be nuts to give away such an equipment even the information about such an equipment. They would simply just mine in silence. I see three possibilities behind this:

1. They have the equipment now for a long while and stocked up huge ammounts of bitcoins. They convert it to $ and then they sell the hardware to cancel or even profit from the hardware they used for their mining.

2. They made the singles for profit, and then some governments (who thought the whole bitcoin idea is dangerous) came up with a devilisly good plan: they will disrupt the market by pouring asics that make the bitcoin market collapse (as they for sure will when the 40000 MH/s equipment ships). What is 2M$ for a government: nothing, and the chips still can be good for code braking, so they can build some defense even from the created asics.

3. The whole thing is a scam...

Eitherway, who believes that a company will sell something that can bring its value back in 10 days (7 days in the case of the mini rig) is worth to be fooled...

[QuantumKiwi]

Guys seriously, think a bit...

To forge these ASICs, are about 2M$ . Let's assume they did it. With a jalapeno a breakeven comes in 10 days. They would be nuts to give away such an equipment even the information about such an equipment. They would simply just mine in silence. I see three possibilities behind this:

1. They have the equipment now for a long while and stocked up huge ammounts of bitcoins. They convert it to $ and then they sell the hardware to cancel or even profit from the hardware they used for their mining.

2. They made the singles for profit, and then some governments (who thought the whole bitcoin idea is dangerous) came up with a devilisly good plan: they will disrupt the market by pouring asics that make the bitcoin market collapse (as they for sure will when the 40000 MH/s equipment ships). What is 2M$ for a government: nothing, and the chips still can be good for code braking, so they can build some defense even from the created asics.

3. The whole thing is a scam...

Eitherway, who believes that a company will sell something that can bring its value back in 10 days (7 days in the case of the mini rig) is worth to be fooled...

Seriously, i think you have been watching a few too many movies.

Govt's dont buy up bitcoins.

BFL have been selling hardware for a while now - They know what they are doing and the staff are skilled. If its only a team of 10 staff, do you really expect them to run their own rig bank? Of course, but not a huge one - one that pays the bills.

Get realistic, no one is holding gear back from anyone - There is huge demand for this hardware, and there is already hardware near like it already available....

[mrb]

Guys seriously, think a bit...

To forge these ASICs, are about 2M$ . Let's assume they did it. With a jalapeno a breakeven comes in 10 days. They would be nuts to give away such an equipment even the information about such an equipment.

If you believe that BFL's efficient ASICs are a big lie, you should bet "disagree" on http://betsofbitco.in/item?id=665
You will win a nice amount of money, as the odds are at 1:18 right now

[RATM69]

What I did was go to a Bitcoin Calculator like this one https://bitclockers.com/calc. Then use 150 Th/s as the minimun number to find out what dificulty it takes to get 3600 bitcoins in 24 huors. 3600 per 24 hours is what it will be after the halving. Then I plugged in the hash power that i'm buying to see approximatley what it will be. Then I did the same process with 300 Th/s. 300 Th/s is what some are saying the hash rate will be by april? May? June possibly. No one really knows what it's going to be until the ASIC's are released and pluged inmining away.
By doing this you will soon realize a 3.5 Gh/s Jalapeno won't earn you very much unless you have a 100 of them.

[dooferorg]

I tried this:

http://bitcoinx.com/profit/index.php?difficulty=26940470.952955&hashrate=3500&cost=150&electricity=0.10&power=40&months=1.0000&decline=0.50&blockcoins=0

I put difficulty at 10x the current one, for added pessimism

Looks pretty dismal for a Jalapeno on its own (nearly a year for ROI), but it is certainly worse for GPU owners. I got one myself simply for the fun of it. It won't let you retire