Hi, creativex!
I think, the main question that is interesting for everybody is what equipment are you going to by in the future. So what and when?
Perhaps, the situation with share price decreasing is the result of lack of information.
Hi AbendWind. I could go back months, but the bottom line is that there's nothing available for immediate shipment that makes much sense and pre-ordering gear with uncertain or lengthy lead times is a recipe for disaster in this environment. I believe bitfury Europe got some 55nm products shipped prior to the end of the month. Bitfury US missed their projected timeline, but I believe they're hoping to ship early this week. Hopefully we'll have more information about their finished products very soon and we can plan to move toward G2 equipment in the near future. I even saw an unsubstantiated rumor that bitfury is working on a 28nm die shrink of their amazingly efficient 55nm chip design that may be available as early as November. Network expansion such as we're currently seeing means we must be very careful how we deploy funds.
I realize there are those that would like me to simply say we're pre-ordering x,xxx Gh/s from vendor x, but doing so would be unwise.
Cheers.
1) I've worked toward bulk pricing/preferential treatment with vendors in the past and will do so in the future.
2) We do not have any pre-orders with anyone at this time. Pre-orders have been fairly toxic for mining companies. I'm keeping abreast of knc's progress, but with so many vendors looking to ship in Q4 I can't say who's gear we'll end up with at this point. As bitfury's chips are substantially more efficient than knc's, all else being equal then we'd surely be better off with bitfury gear as it should remain profitable longer.
Hi guitarplinker. The thing is it's not a question of resources at this time. We're sitting on over 775BTC precisely 776.5493 in reserves right now. This is enough to obtain nearly 14Th/s in pre-ordered gear from cointerra for December delivery or 4.8Th/s from bitfury for October delivery at current exchange rates(bitstamp). The problem is the same as it's been since spring/summer of 2012, and that is that locking in our funds for unproven vaporware with lead times measured in months(or years) is unwise. It's even worse than it appears because we'd also be locking in an exchange rate. An exchange rate which btw is rapidly climbing in our favor. We also cannot run out and load up on currently available hardware that's soon to be obsolete because we may or may not ever see a return on that investment. A couple such mistakes could cripple us going forward. It's my belief that when things are moving so quickly that the future is nearly impossible to predict with any level of confidence then the correct play is to shorten your scope substantially. That is, reduce risk by sticking to decisions with outcomes that can be predicted with some measure of confidence.
Hi weezerfan. I'd very much like to simply calculate how many avalons we'd need to add on to keep up with difficulty each re-target and go out and source them, but that won't work as we'd eventually go broke. Safeguarding shareholder value is much more than just maintaining a consistent dividend payout. I have to run each and every purchase candidate through my own common sense detector to determine if we're likely to at the very least get our money back out of this gear before I have to unplug it to stop losses from expenses. We're at a crossroad here where those on the cutting edge are transitioning to 2nd gen ASIC equipment and just like the previous major equipment shifts(CPU --> GPU --> FPGA --> ASIC) there's a lot to gain or lose. I believe we have to keep sufficient powder dry so that we can make significant purchases of gen2 gear when it's actually available.
What FMG points out in the post following yours is accurate, but, if we play it smart we don't have to be victimized by the laws of big numbers. Sure while it would take 17 avalons to double our hashrate now, a single cointerra unit would more than double our hashrate in Dec or January. We just have to not shoot ourselves in the foot before we can get there. In between I see potential in bitfury's products for October delivery and possibly knc's products for November delivery to possibly tide us over.
Also I think it's important to keep in mind that while the law of big numbers applies to our hashrate, so too it applies to everyone else's and the global network itself. It's not hard to believe the global network can expand by 3%/day for a few quarters after new technology revolutionizes mining, but the amount of capital required to sustain that growth long term eventually makes it unsustainable unless the exchange rate grows along with it, which in turn would make older less efficient gear profitable longer.
Thank you FloatesMcG, kind of you to say. It saddens me somewhat to have to disagree with you after such a nice lead-in.
While obviously nobody can keep up with difficulty increases step for step right now(not even FC), nor should they beat themselves to death trying, we all know this kind of expansion cannot go on forever. It will go on until the market is saturated, pre-order queues dry up, and mining margins thin significantly. If you take the pulse of the forum just now it's easy to see attitudes changing from "ASIC ANYTHING = TAKE MY MONEY NOW!" to "2Th/s for $16k? pffft...overpriced". Remarkably that vendor quickly lowered their pricetag in response. For now I'm just holding on and trying to make decisions that ensure we at least break even on every purchase.
It was never my intent to build a company based on scavenging gear, it's just my mechanism for adapting to the current market forces. There just isn't a reliable supplier of in hand original equipment that's priced in such a way that there's still enough meat on the bone to justify the risks. This will change in the near future IMO, and the people with dry powder will benefit. Love what's going on over at bitfury for instance.
When I speak of efficiency, I'm referring to power consumption. BFL's 65nm based products are somewhere between 50-60% more efficient than ASICMiner's current 130nm offerings. While this hasn't mattered much in the past as operating costs are currently negligible, this will change as operating margins for miners continues to contract. Eventually all but the most efficient ASIC gear will be forced offline as they'll cost more in electricity than they're generating in revenue. Bitfury's 55nm gear shipping is significant in that it's the first gear to take the efficiency lead away from BFL.
As you said, the auction that you linked is for Oct/Nov delivery. Dave at bitfury has suggested that bitfury chip orders would ship early in October and finished products would likely ship toward the end of October. This is one more link in the chain we'd have to rely on and the potential savings we'd get by going with Barntech's auction over a direct pre-order purchase at megabigpower would seem to offer a poor risk/reward profile.