I want to clarify the above post, as there is some potential confusion. We are not shipping yet. We plan on shipping possibly by the end of next week, but I will update on a shipping schedule as soon as I have more definitive information with regards to that.
15 April 2013 Update
Short update today, more to come in the next day or two, hopefully.
We've been working on getting the software and firmware nailed down. Things are mostly stabilized at this point and with our current boards (we do not have the redesigned boards in house yet), power usage exceeds the 1w/GH, unfortunately, but it's much better than anything out by 40 - 50%, at the wall. We are gearing up to start shipping out some dev boards and a few Jalapeno's most likely this week (at least a few dev boards) and then as more chips roll in we'll be shipping out the Jalapeno's. When the new boards land in KC, we'll start shipping Little Singles and Singles at that time. I don't currently have a time frame for those, but I should have something later this week in regards to that.
Obviously, the minirig can't fit 1.5 TH/s in a case the size of what we were planning, but we have some interesting solutions with regards to that. Expect and update on there as well, but I did want to let everyone know we have several solutions to solve the minirig issue and we are moving forward on that front.
I did want to quash any rumors that there was something wrong with the chips; there isn't. The chips themselves are fine (other than using more power than we expected), the delay is strictly due to having to re-engineer the power system on the boards (which requires a larger PCB redesign to a degree) and refine/streamline our processes for getting chips from silicon to mounted on boards. Things will be rolling along soon, albeit a bit slow at first, and then picking up speed as the whole process is optimized and finalized.
6 May 2013 Update
6 May 2013 ASIC Update
It's been a few weeks since the last update and a lot has happened since then.
As many of you know, we started shipping out a quantity of Jalapeno's (around 35 or so), about half of those went to developers and the media, the other half went to customers. They have been well received so far and everyone is happy with what they have received. In the meantime, we have been diligently working on the Little Singles, Singles and Minirigs.
This past weekend, we received in the new boards for the Little Single, Single and Minirig, pictured below. As you can see, the board redesign incorporates a total of 16 possible chips, as well as some power distribution changes. One of the more prominent changes is the fact that we have moved from barrel jack power to PCIe style Minifit Jr 6 pin connectors. Each board has two 6 pin connectors to provide power, and each board is intended to provide up to 60 GH/s hashing power.
We discovered some issues that we needed to resolve with the boards and had to change a few minor things. We tested them out on a hand modified board and they worked, so we've issued the redesign to the PCB house. The new boards are being finished today, the 6th of May, and should be ready this evening. We have more chips from the packager on the way to the assembly facility, and they should arrive on Thursday, the 9th of May. In the meantime, we will be testing and making sure the new board is functioning like we intend for when the chips arrive to the assembly facility. We don't have a count on the number of chips that will be arriving on Thursday as of yet.
The current board will support the hashrates ordered for Little Single and Single in one box, so single units will be shipped out for both the Little Single and the Single. The Minirig will most likely ship as two units (where as it was just one before) and possibly some Singles to bring it up to the full hashrate ordered. 5 GH/s miner production will continue concurrently with the rest of our product line. We are incorporating some changes into the new 5 GH/s miner to bring the heat and power usage down, and that will be incorporated as we go forward. Power usage on the new boards hasn't been tested yet as we are waiting for the revised board, so I don't have power figures for it yet.
Below is a picture of the new board for the Little Single, Single and Minirig, as well as a Single running.
10 May 2013 Update
Another busy week here at the labs. We continued to refine the LS/S/MR boards while waiting for our next delivery of chips. We expected them today, Friday the 10th, but the packager has been dragging their feet in the process (and one of the reasons we initiated the switch to a new packaging facility, as mentioned previously). The first bulk shipment from the new packaging house is expected on Tuesday and should consist of several hundred chips, which we will be building into all the product lines, assuming no snags are encountered this weekend on the latest rev of the LS/S/MR boards. If there is a blocking problem, we will build them all into 5 GH/s units, but we don't expect any blocking issues at this point.
The rest of the wafers are expected at the bumping facility the following week, and once bumped, they will head to the new packaging facility which is capable of a much higher volume, so we expect them to be packaged pretty quickly and ready for use after they arrive at the facility.
13 May 2013 Update
We just got off the phone with our testing facility... they will have ~100 chips in Chicago tomorrow. What configuration of boards all those are going in, I'm not sure yet as I haven't gotten that far, but I wanted to post an update. We should have more chips in Chicago by the end of the week. The chips going through mounting tomorrow should arrive in KC on Wednesday and we'll be shipping out some units then.
We are having some trouble with the new Single board and we are working through the issues now and should hopefully have a a resolution before the end of the week. Right now, the board is having trouble powering up the second bank of chips, so we are poking and prodding to figure out where the problem is and get it corrected. When we make a change to a board, we can have a revised board the next day, so it doesn't really set things back when we make revisions to an already existing board.
That said, we'll likely be creating a few Singles boards with this set of chips and a number of Jalapeno's. The expected larger quantity of chips later in the week will probably go more towards LS/S/MR boards.
22 May 2013
Progress has not stopped here at the labs while we attended the 2013 Bitcoin Conference. We demonstrated a 5 GH/s, a 25 GH/s and a 50 GH/s miner at the conference that people could see, touch, stroke and pet, all while they were mining on an Android tablet. It was a good conference and we were glad that we could demonstrate working units and meet many of our customers and fellow bitcoiners. It was a pleasure meeting many of you.
We have resolved the issues with the long boards and are making some tweaks to the 5 GH/s boards that may reduce power consumption but will almost certainly reduce heat output. With that in mind, we should be getting about 200 5 GH/s boards on Friday, and we'll be shipping those out on Friday and Saturday.
As a term/jargon/technical note, here is the informal and/or official designations here at BFL:
Short Board = ~92mm x 92mm board that contains up to 8 chips and powers the 5 GH/s miner, typically with just 1 or 2 chips.
Long Board = The board that powers the Little Single, Single and Minirig and can contain up to 16 chips in two banks of 8.
Jalapeno is now called the 5 GH/s miner and is synonymous.
The Minirig is now a 500 GH/s unit.
Our chips consume about 3 - 3.5w per GH/s at the chip level, so there is some room for tweaking, which we will be focusing on as soon as we get the final rev out for the Long Boards. Speaking of Long Boards, we have what we are considering the last rev for shipment coming next Monday or Tuesday the 27th or 28th of May (they may even be done with manufacturing this Saturday the 25th) and we will continue to refine the firmware for that board until the final rev arrives in KC. Assuming no problems with it, we will be being shipment shortly thereafter. The long boards will be powered by one or two PCIe style 6 pin Minifit Jr connectors. Please be aware that these connectors, while using the same form factor as PCIe connectors are not PCIe connectors, though they are electrically compatible with them (Meaning you can likely use an ATX PSU at your own discretion if you chose not to use our supplied power bricks). The Little Single and Single will have a 120mm fan at each end of the case in a push/pull configuration. Each bank of 8 chips will have a heatsink, heat baffle and 92mm fan on top of it. Cooling will not be an issue with these units and they will be fairly quiet to run if you are in a cool environment. The fan will scale up and down in speed as required due to heat. The units will be stackable and have a side to side or front to back airflow, depending on how you orient them.
We are expecting several thousand chips next week from our new packaging facility and expect to start shipping all product lines next week. Beyond that, we will be receiving thousands of chips per week at an ever increasing pace between now and until all of the remaining 68 wafers are used up. We have additional wafers on order and they should be ready about the time we've used up all 68 wafers should we need them. Soon after, we hope to have product on the shelf for immediate shipment, depending on how many orders come in between now and when we catch up the backlog.
29 May 2013 - ASIC Update
29 May 2013 - ASIC Update
Almost there... I'm attaching a picture of our test board. We've almost got the firmware finalized, as you can see, we've brought power usage down quite a bit. This is a 50 GH/s unit with unbinned chips (meaning the chips were randomly selected, so some of the chips have non-functional engines. What this means is the unit can perform faster if you use binned chips with known good engines).
I think the picture speaks for itself, and I apologize for the quality... just took a quick snap while we were working. However, you can see that our power usage is under 5w/GH/s. This should apply across all product lines. I plan on testing out the short boards tomorrow. We should be getting in many of the reworked short boards tomorrow as well as some long boards. Next week, we expect a much larger shipment of both long and short boards to arrive (possibly even this Saturday, but that's still up in the air.)
In any event, we should have the firmware finalized by tomorrow night or Friday at the latest and ready to go.
22 June 2013 - ASIC Update
I haven't updated in awhile because I thought the shipping was self explanatory, but people seem to be clamoring for an update, so here we are.
We have been working our way through the backlog of orders over the past month. Production is slowly ramping up and we are easily able to manufacture 200 units a day currently - there will be no problem hitting 400 units a day once things are in full production. Right now, we are stymied by the inflow of chips. However, the inflow rate of chips is increasing and we expect it to be quite high within the next week or two. It's been a long road to smoothing out these bottlenecks here and there.
Lots of Jalapeno's have been sent out the door, as well as some singles and minirigs. Production on the longoard lines are really going to start to ramp up next week and the following weeks.
We have been refining the firmware for both the short board but mostly for the longboard. One of the primary changes has been to switch from a one job per board method to a one job per chip method. This has fixed a lot of the latency issues we were experiencing with the early firmware and what held up production on some of the 60 GH/s units. We have not yet finished the XLINK portion of the firmware to be operating at 100% yet, so the first few minirigs that are going out the door are shipping as individual USB devices (8 per box) instead of 1 big USB device. Going forward, once the XLINK stuff is finalized, minirigs will be shipped out as XLINK equipped.
We will continue to refine the firmware as we go along. With the current firmware, we are seeing between 260w to 280w per 60 GH/s unit, or 4.3w - 4.6w per GH at the wall on 120v. Actual chip wattage is just under 3w per GH.
In other news, we have released all the documentation necessary for people to begin creating their own boards to house our chips and we have begun selling our chips in bulk. The first deliveries of our chips are expected in roughly 100 days from the time of order. We still expect the current back order backlog to be cleared by the middle of September or so, although given what we can likely achieve with our production rate, I suspect it may be sooner than that. No guarantees on that either way, though.