Butterfly Labs Announces Bulk ASIC Chip Sales for Bitcoin Mining


Considering a bitcoin mining hardware purchase? Make a more informed decision with our Bitcoin Mining Dashboard.

Following Avalon’s announcement of bulk ASIC chip sales in April, Butterfly Labs (BFL) has announced they will be doing the same. ASIC chips are the customized component that performs hashes in bitcoin mining. Previously, all BFL preorders were ASIC chips already installed on boards and ready to be used immediately. This is a new product category in addition to their previous miners and will require designing boards to set the ASIC chips in. BFL estimates a 100 day lead time, with pricing is as follows:

Grade A chips have 16 engines and will do 250 MHz comfortably, equating to 4 GH/s per chip.

Chips will be graded and priced as follows:

Grade A: 16 Engines – $97/chip
Grade B: 15 Engines – $93/chip
Grade C: 14 Engines – $89/chip

Grade D: 13 Engines or less – $83/chip

They will be testing and grouping the ASICs into performance grades A-D, and pricing appropriately. Grade A will run at 4 GH/s and cost $24.25 per GH/s, while grade C will run at 3.5 GH/s and cost $25.42 per GH/s. There is a minimum order size of 100 chips, and the costs themselves can be further reduced up to 20% by ordering in bulk. Most likely there will be group buys organized through the bitcointalk.org forums to take advantage of these discounts.

100 – 999 chips – Regular Price
1000 – 9999 chips – 5% discount
10,000 – 99,999 chips – 10% discount
100,000 or more – 20% discount

Chip orders will require 50% down and 50% upon order delivery. Payment can be made via BTC (preferred) or bank wire.

Avalon released open-source reference designs about a month after they began their bulk sales, so it would be expected that BFL will do the same. These ASIC chips use a more advanced manufacturing process than Avalon, and will  provide a better power and cost efficiency than Avalon’s 282 MH/s chips costing $33.20 per GH/s. All other things being equal, this should allow for significantly smaller units that run far cooler for the same hashing speed. The real question is whether BFL’s continuous delays will be worth those efficiencies.


For the latest analysis and updates