2

Where can I verify how much ASIC mining power has been shipped, how much has been pre-ordered, when it will be shipped, and at what rates and time frames more units will be shipped?

  • And a related question: is there any info on how much of the ASIC supply is being bought up by mining contract companies rather than individuals? – Troy toy Mar 21 '13 at 23:15
  • 2
    This question needs a timeframe, or it will soon be out of date. If you don't like my edit, feel free to update it... – Highly Irregular Mar 21 '13 at 23:52
2

Let's analyze ButterflyLabs alone.

From the Butterfly Labs Wait list, here are the current numbers, along with the number of chips:

Orders Tracked: 325

Unit             Number tracked     Chips/unit     Chips tracked
Jalapeno SC      216                  1              216
Mini Single SC    41                  7              287
Single SC        348                 14             4872
Mini Rig SC       25                334             8350
------------------------------------------------------------------
Total                                              13725

Thus, our average chips per order is 13725/325 = 42.2 chips/order from what we know. Next, let's look at the number of orders and assume that average stays consistent. Let's say there were 18,500 ASIC orders under the first ordering system (Order numbers 1500-20000), and are currently 12,000 in the second order system (Order numbers 100014000-100026000). That brings the total orders to be filled to 30,500. Multiply that by our average number of chips per order and we have 1,287,100 4.5GHz chips on order. Thus, if these numbers hold true (and I really hope they are VERY high), we are looking at 5.8 Million GH/s added to the network from Butterfly Labs alone once they fill the orders to this point. This is about 100 times the current global hash rate of 50000GH/s.

Also note that this is based off a lot of assumptions, and they haven't shipped their first unit yet, so who really knows. When I contacted them in February, they said orders placed then would be fulfilled in May-June, but they also said they were shipping the first units by the end of February and they haven't shipped those yet.

UPDATE (4/22/2013):

The last order I see posted is 100036087, on April 8, up from 100026000 on March 22 (2 months ago to the day!), so we have an increase in order numbers of about 5,000 per month at the moment. Also, they just shipped their first developer units, and they had to up the case sizes to accomodate the extra power modules. Thus my best guess is that BFL will begin shipping to customers late May/early June.

Speaking of order numbers, I had assumed above that they increased every time an order was actually made. However, According to Sgt Spike on the wait list forum about BFL order numbers,

"It generates an order number every time someone goes to checkout, whether they actually 
 pay for it or not."

Which, if true, means that the above numbers are a definite worst-case scenario (unless of course there are a bunch of missing orders from the list for 100 mini-rigs or something crazy like that, which is very doubtful). Thus, multiply the above numbers by the percentage you think that people actually followed through to a purchase after proceeding to checkout to get a better estimate.

Also, it looks like the earlier orders were much more heavily weighted than the later orders, where the info is also much more sparse, so the estimate still may be high even after you figure the correct buy percentage in.

Last, note that going forward the mini-rig is out of stock and they have upped the price and they have lowered the hash rates available from 60GH/s and 30GH/s to 50GH/s and 25GH/s respectively, so going forward the GH/order will probably be much less on average. Even before they did that. Thus, that should lower the expected addition to the global hash rate as well going forward.

With the above considered, I'd say a reasonable upper limit is about 2 Million GH/s added, which assumes about half the people that go to checkout actually place an order and the average order is slightly smaller than the average of the orders submitted.

  • If that's is correct even if no more orders are placed that's 7200 bitcoins per day / 1,287,000 = 0.0056 bitcoins per chip per day or around 71.76 USD * 0.0056 = 0.4 USD per day per chip. So not profitable unless you ordered early. – Christopher Edwards Mar 22 '13 at 17:55
  • @ChristopherEdwards: Unfortunately, it's even worse at 3600 bitcoins per day (25 per block instead of 50). For the 30/60GHz, let's say it's about $100/chip after fees. Thus, you'd get your money back at current projected rates and that hash rate in 500 days if you bought today and Butterfly Labs was the only one in the market, but that number could get MUCH higher very quickly once the ASIC companies streamline their production process. Honestly, BFL dragging their feet is the best thing they can do, as once people see the updated Global Hash Rates, the orders are going to stop flying in. – Briguy37 Mar 22 '13 at 18:39
  • ...but again, I hope I'm wrong! – Briguy37 Mar 22 '13 at 18:40
0

Nobody has that information but some supposed ASIC companies published how many preorders they had received (BFL). This would be a good indicator of the demand or amount of machines that will saturate the market.

  • Fine, but either link to these publications or this is not an answer. – o0'. Mar 22 '13 at 16:36
  • 1
    @Lohoris I'm sorry you feel that way, the OP asked "where" he can go to get the information, and I told him who would have it, assuming that OP would then pursue knowledge in that direction. And my answer is also in line with the other answers. So I respectfully disagree with your ultimatum – CQM Mar 23 '13 at 0:54
  • I still disagree with you, but I'm sorry if my comment was harsher than intended! – o0'. Mar 23 '13 at 7:14

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.