Topic: BFL ASIC products, worth it? (Read 1674 times)

Just thought you guys would like to read this from parts of Josh's post.

"Another delay we've had to endure is the fact that we have effectively tied the ASIC teams payment to the success of the chip. If the chip were to be a failure they don't get paid... so they have incentive to get it right but that has made them very cautious and slow to approve final masks (This is why we can refund all pre-orders we want and why we have the capital to do what we need to do without a failure putting us in bankruptcy)."

If my boss said as a mechanic if this engine does not work good you don't get a paycheck I would roll my box out!  This is the kind of employer I refuse to work with. Are they partnerships here? or what? Just tell me what type of person would work for free atm and possibly risk their pay on a new chip.

Who has the responsibility here? The employees? the Company? the sales team? Something is a miss here imo..... Am I the only one here that finds this very wrong?

https://forums.butterflylabs.com/bfl-forum-miscellaneous/690-13-jan-2013-asic-update-discussion-thread-5.html#post10399

I myself preordered a BFL jalapeno and did some calculations.... at the current network hashrate it would theoretically make me 10x the investment in 4 months! BUT the problem is that only BFL (there are other companies) are planning to release 20k chips (4.5 or 7.5GH/s processing power per chip) in the upcoming months... this will increase the hashrate DRAMATICALLY and increase difficulty by atleast 10x the current 25TH/s... it's quite possible to get a rough estimate on the future hashrate presuming everyone uses the devices they buy... as most of the ASIC companies order data can be found on their respective forums.

What do you think about the future BitCoin network hashrate? Place your bets here: http://betsofbitco.in/item?id=1139

Wait till they ship then watch the people who receives them complain about the problems then decide if they're profitable. There is a chance they will be less in price after to reflect the ROI.

Learn from others and you will be better off.

In a perfect world the BFL ASIC products are a no-brainer.

Consider the mining pool "ozcoin" (I have no connection with it)

As of the time of writing their site https://ozcoin.net/ shows pool metrics of:
Pool: [ 1,511.46 GHash/s | 273 Users | 616 Workers ]     Server Time: [ 2013-01-13 04:17:16 CET ]

Their 1511 GH/s is almost the same as one BFL ASIC mini-rig which costs US$29.9k
http://www.butterflylabs.com/products/

Checking out http://blockorigin.pfoe.be/top.php shows ozcoin hashing 120 of the last 2016 blocks (basically during the first fortnight of 2013).

So: 120 x 25 BTC x 13.50 (avg fx rate) = US$40.5k in two weeks!  Assuming no downtime, but fees would probably cover electricity.

So a BFL mini-rig is a money-printer - in a perfect world. The challenge is obtaining one first, before they become "widespread". The decay in profitability is proportionate to the rate at which similar machines enter the market and become actively mining. The landscape for bitcoin mining will be completely different at the end of 2013 than the beginning.

> You certainly won't be making $134/day unless you're one of the first ones to get an ASIC miner, and even then it won't be that profitable for long.

Seems to me any advantage you might have mining at a higher rate with current difficulty will last anywhere from 1 minute to 14 days. 1 minute if you are unlucky to start mining at 2015 blocks and the difficulty goes up. 14 days if you start your ASIC at block 1 and mine at a lower difficulty.

https://en.bitcoin.it/wiki/Difficulty#How_often_does_the_difficulty_change.3F
https://docs.google.com/spreadsheet/ccc?key=0AmcTCtjBoRWUdHVRMHpqWUJValI1RlZiaEtCT1RrQmc#gid=0

1465 days since first mining divided by 108 difficulty changes = 13.5 days.

2) Any guesses as to how long a purchase of a Jalapeno ASIC would take to pay for itself? Would you say a 20% increase in difficulty with each 2016 blocks is a fair increase? (I'm taking the average increase in difficulty from the Google Docs).

Cool, makes sense. Most of the work done isn't network data intensive, it's processing on the local machine...

Right now the amount of network bandwidth required to mine bitcoins is negligible, especially with specialized mining protocols like Stratum and Getblocktemplate.

When ASICs come out you will probably want to use a pool and miner program that supports Stratum and/or Getblocktemplate, both to decrease stales and mining pool server load. Poclbm, cgminer/bfgminer, and a few others support this, and there is a Stratum proxy available for miners that don't natively support Stratum.

Although mining at ASIC speeds will use more bandwidth than current devices require, any basic broadband connection (even the cheapest ones offered by the major cable companies) should be sufficient to mine.  I haven't actually measured it but I'd be surprised if mining at ASIC speeds will use more than 100MB/day.

True  So for someone like me (who uses electricity for heating) I can still state I am not earning money from mining.
But -> my ovens are earning the money back now 

there are still people making money with thier GPUs or FPGA setups, but not enough to justify the risk of buying additional hardware with the looming ASIC releases.

If ASICs are not delivered by any company, then a new GPU/FPGA investment could pay for itself in about a 1-1.5 year period (give or take electrical cost and hardware price changes). However, if ASICs are released, that could change to 10+ years or not at all (although some small profit may be made over the cost of power/cooling it wouldnt pay for the hardware)

Likewise, once ASICs hit the market and take over the majority of computing, they will raise the difficulty and slowly saturate the difficulty/profit/hardware cost equation such that it will approach the point that FPGA mining is at now, with 1+ year hardware payoff times. The big debate is how long this saturation period will take, with the huge number of *supposed* preorders and some philisophical debates over how quickly people adopt new technology

Yes more expensive-than-average. Just out of curiosity, if all the parameters were the same for every miner (costs, processing power, equal pools, etc..), would the algorithms in place essentially determine that profitability rests entirely on free market forces, such as fee's (and say for realities sake, MINUTE cost differences between miner efficiencies, resulting in some to go out of buisiness, decreasing the difficulty, and making it more profitable, then essentially balancing out again, etc.
.)?

Also, if someone could answer my question on internet connectivty speeds for mining I would really appreciate it.

Thanks guys! I'm really enjoying these responses!

The OP question is non sequitur because the products are vapor ware.

That said, if these things prove to be as purported then I either expect to see a marked increase in the mining of alt-cryptcurrencies and/or the appearance of a bevy of cheap used GPUs on the market.

I believe it will use a standard USB-serial driver that's probably included with/automatically available to your OS.  With BFL's current (FPGA) products, Windows will automatically install the necessary drivers when you connect it and no drivers are required on Linux.

You will need a compatible mining program; BFL's ASIC protocol was just released a few days ago (and still seems to be under development, based on the topic in their forums) but I remember reading somewhere that bfgminer will definitely be supported.  I'm going to assume most of the popular miners will include support for BFL's and other ASICs, but that's just my assumption.


I assume you meant someone with a "greater-than-average" cost of electricity?  That's not necessarily true.  Although people with the cheapest electric cost will probably have the lowest operating costs and therefore highest net returns, mining can still be profitable as long as your electric costs are lower than your mining income.  Cooling is another consideration; if you can cool your units wihout using a lot of additional power (i.e., conventional air conditioning), you may have higher returns than someone who can't but pays less for electricity.  You also should consider if the heat generated from mining can be used, such as to heat your home in the winter.  In that case your electric costs are essentially just the difference between the cost of purely resistive electric heat and the cost to run your normal heating system.

So basically, the difficulty increases due to the built-in algorithms as more hashing power enters the network. Thus making it harder to do the work, requiring more hashing power... thus making it pretty much pointless for someone with a less-than-average cost of electricity.

Does it pretty much come down to that then? How cheaply can I get power? Since everyone can effectively process at the same hash rate with these ASIC chips?

Do one need a special driver? will that be included or can one download that driver yet to be ready at best when the device will arrive?

^well stated. An ASIC device is a *theoretical* good investment, and i intend to buy from the first demonstrated manufacturer.

however, ASIC mining will quickly become a big thing, and the difficulty will scale up VERY fast.

Most likely, but only time will tell. (and yeah, the idea is you connect to an Internet-connected computer via USB)
You certainly won't be making $134/day unless you're one of the first ones to get an ASIC miner, and even then it won't be that profitable for long.

The only things known are:
- Lots of people have pre-ordered ASIC devices
- The only people who will definitely be profitable are the companies selling ASICs. (that being said, I've preordered BFL devices myself)

Not known:
- When any ASIC miner will actually be released.  Avalon ASIC says 16 days but BFL has no definite date. (and I don't know about bASIC)
- How high the difficulty will jump, which will determine the profitability of any mining device.  There are estimates that it will jump anywhere from 10x current to 75x current.  It really depends on how quickly ASICs are released, and nobody knows that. (and how many people buy them)

The Mining/Hardware/Custom Hardware forum is full of ASIC speculation... good luck!

So Butterfly Labs has been collecting pre-orders for a while now on their new line of ASIC Products (which im sure has been widely talked about all over this forum):

BitForce Jalapeno: $149
Performance: 4.5 GH/s

BitForce 'Little' Single SC: $649
Performance: 30 GH/s

BitForce Single 'SC': $1,299
Performance: 60 GH/s

BitForce Mini Rig 'SC': $29,899
Performance: 1,500 GH/s

So how does it work? Do I just connect this mining rig box via USB to a computer with an internet connection? Should I have a minimum up/down connection with my ISP? If so, what should that be? Is it profitable at .13 cents per kWh?

More importantly, with such a huge demand and strong sales, will a lot more people with this hashing capability effect the network, and potentially differ from the estimated profitability used on most websites like http://www.bitcoinx.com/profit/ , which calculate based on current estimates?

So, Im thinking to myself for the 60 GH/s, "ooo man! I'll make $134.58 USD a day at .13c per/kWh, and $13.32/BTC! It's totally worth it!" Only to later flip the switch when I get it and see that nothing has really changed since all other miners are using ASIC hardware as well...

Does it work like that?