What are the actual good companies out there?

[ToxiClay]

I've been hearing so much about BFL being a scam, Cointerra not meeting their timelines for product push-out, KNC being the major game in town...and then other names pop up, like ASICMiner, HashFast, Black Arrow Software...

As someone who's looking to get into dedicated ASIC rigs, because consumer hardware isn't paying the bills anymore, how do you weed out the chaff and figure out who're the major players? get the best ROI?

Looking at the Bitcoin Profit Calculator, even if the difficulty is double what it is currently, the heavy ASICS (500GH/s and up) are still profitable enough. So why the hate for the heavy miners, and where should one look to get the best ASIC? Black Arrow's Prospero X-3? KNC's Jupiter? Or one of the newer players?

[gnar1ta$]

If you're looking to buy an ASIC with BTC and make more BTC than it cost you, you're chasing unicorns.  Other than first batch Avalon, I don't think any have done this, and none look to any time soon, even if they met their deadlines - which again, none have so far.  Do your own calculations, with your own risk analysis, don't trust the profit calcs.

[Cheshyr]

There's a pretty big difference between a good company and a reliably profitable miner.

BFL made the mistake of underperforming, delivering late, and having bad PR.  I own a BFL device; it looks good at performs better than advertised.  It will never be profitable.

Cointerra actually looks like a very solid company and device.  They're the company I'm personally most inclined to trust, although I can't speak towards their schedule, pricing, or ROI.  Their executive team inspires a lot of confidence as well.

BitMain is shipping products now, and they're the other end of the spectrum from Cointerra...  small, simple ASICs that get the job done.  Communication is a bit spotty, and they're not great about updating their github with appropriate documentation, but it works and you can buy one right now and get it next week.  There's something to be said for delivering on time and on spec, even if it's nothing spectacular.

It sounds like HashFast dropped the ball on a couple things, and was a little inexperienced when it came to things like regulatory testing.

I wasn't really paying attention when KNC announced and shipped, but it sounds like they were in the right place at the right time with the right product.  I get the impression they're advanced amateurs, but money could shore them up. We'll have to see if they can capitalize on their head start.

CoinCraft just posted positive news regarding their A1 chip, and it's looking promising.  I'm hopeful that they'll be able to deliver, and perhaps become one of the first companies to provide moderate performance chips to the DIY crowd.

Black Arrow is another company I have high hopes for, but I don't have any facts to back that up other than some well established forum members vouching for them.  Maybe someone else can help there.

ROI is a tricky subject, and I don't think you're going to get any help here.  Any purchase if going to be high risk, whether it's mining equipment or market speculation.  

[ToxiClay]

There's a pretty big difference between a good company and a reliably profitable miner.

BFL made the mistake of underperforming, delivering late, and having bad PR.  I own a BFL device; it looks good at performs better than advertised.  It will never be profitable.

Cointerra actually looks like a very solid company and device.  They're the company I'm personally most inclined to trust, although I can't speak towards their schedule, pricing, or ROI.  Their executive team inspires a lot of confidence as well.

BitMain is shipping products now, and they're the other end of the spectrum from Cointerra...  small, simple ASICs that get the job done.  Communication is a bit spotty, and they're not great about updating their github with appropriate documentation, but it works and you can buy one right now and get it next week.  There's something to be said for delivering on time and on spec, even if it's nothing spectacular.

It sounds like HashFast dropped the ball on a couple things, and was a little inexperienced when it came to things like regulatory testing.

I wasn't really paying attention when KNC announced and shipped, but it sounds like they were in the right place at the right time with the right product.  I get the impression they're advanced amateurs, but money could shore them up. We'll have to see if they can capitalize on their head start.

CoinCraft just posted positive news regarding their A1 chip, and it's looking promising.  I'm hopeful that they'll be able to deliver, and perhaps become one of the first companies to provide moderate performance chips to the DIY crowd.

Black Arrow is another company I have high hopes for, but I don't have any facts to back that up other than some well established forum members vouching for them.  Maybe someone else can help there.

ROI is a tricky subject, and I don't think you're going to get any help here.  Any purchase if going to be high risk, whether it's mining equipment or market speculation.  

Why am I unlikely to get help here? And, I understand purchases are going to be high-risk, which is a large part of why I'm walking circles around the market, analyzing it about to death before investing cash in it. Fortunately, I've already cleared $thousands in pure profit from bitcoin, and I purchased a mining contract on eBay with an additional portion /beyond/ that, so I think I'm fairly decently insulated, but....money wants more of the same, and my calculations and speculations are looking pretty good -- I'm just looking for a company that won't take my money and disappear into the ether.

You've confirmed some of the things I was already thinking, w/r/t Cointerra and BlackArrow being the potential frontrunners. I can't seem to find an actual website for BitMain, though -- all I see are forum threads here and one reseller site, apparently owned by Black Arrow.

Thanks for your post and insight, and I'm sure I'll have more questions.

I hope you had a good new year's eve, and may '14 bring us all good fortune.

[U1TRA_L0RD]

Dont trust butterfly labs, they wont send you your miner until christmas next year.

[ToxiClay]

Dont trust butterfly labs, they wont send you your miner until christmas next year.

Yeah, that's what I've heard -- also that they're a hot mess. Underperformed, draw too much power, failures, etc.

[Entropy-uc]

The asic market is still very much a gamble.

I personally think Cointerra is the best organization in the running right now.  I met with their team, they do have credible technical backgrounds and they are demonstrating good performance against their schedules.  But there are plenty of pitfalls to catch even the best engineers in product development.

Bitmine appears to be good as well, and they will have smaller devices that might fit your budget better.

Wait a month, and a lot of uncertainty will be gone from the market.  But the backlog of orders from the winners might be 6 months at that stage as well.

Mining is not easy money, no matter what the lucky few might tell you.  Don't play with money you might actually need.

[Cheshyr]

Why am I unlikely to get help here?

You can call my paranoid if you like, but I find the opinions and perspective of this community to be marginal at best.  It's in their best interest to give bad advice so they can own a larger piece of the pie, uncontested.  This leads to all sorts of strange tactics, including defamation, persistent pessimism, and outright lies.

Aside from that, they (we) can't really give you good advice regarding mining.  In the current market, the winner is the person who has the lower $/GHs mining the soonest.  As soon as those products hit the network in mass, the network adjusts, and that level of efficiency is the new status quo.  If you don't make the majority of your ROI in the first month, chance are you never will.  Add to that the wild-west nature of the bitcoin ASIC manufacturing scene, and you've basically got a dice roll.

That's why we can't really help.  Even if we give you perfect advice, it's suspect, biased, and frequently on very shaky grounds.

I'm just looking for a company that won't take my money and disappear into the ether.

Bigger names are better in this case.  Cointerra is the only one that fits that description in my opinion.  We'll have to see if any of the other companies put some effort into legitimatizing themselves.  I really want to build my own miner based on CoinCraft chips.

I can't seem to find an actual website for BitMain, though.

BitMain is a bit of loose cannon.  They're a small chinese company, and they're not technically allowed to export these, I don't believe.  So they sell to an agent that is willing to play in the gray market.  Their chips appear to be solid, although I've seen some issues with their cgminer integration, especially for *nix boses.  It didn't stop me from buying their product, and I have one running just fine...  but it's not terribly cost effective right now.  I think I paid $40/GHs.  The bare chip itself comes in around $5/GHs, so there's room for another board manufacturer in that market.

[gmaxwell]

get the best ROI?

Would you count the least negative of several negative options as the best?

The "best" ROI may, sadly, simply be not to buy— just sit on your coins. Thats always an option.

I enjoy mining and hate to discourage anyone to get involved, I'd even mine with a substantial probability of only breaking even... But the current prices are crazy, and I think the hardware makers are exploiting ignorance about how mining income changes over time in order to extract more from customers than the hardware can ever produce. Substantial losses at their current prices seems likely, and this is assuming that their promises are all kept and deliveries are on time.

This depresses me: Wide distribution of hashing power is an essential element to the Bitcoin security assumptions. The COGS on these mining devices are quite low— once produced in quantity— there is more than enough profit to go around. But this is what we appear to have. There is a bunch of competition selling hardware, but apparently not enough.

Looking at the Bitcoin Profit Calculator, even if the difficulty is double what it is currently, the heavy ASICS (500GH/s and up) are still profitable enough. So why the hate for the heavy miners,

"even if the difficulty is double"  … Please pardon me while I titter and direct your attention to the first graph. Red is difficulty. The other lines are hashrate estimates— difficulty scales linearly with hashrate every 2016 blocks.

As you can see in the second chart the hashrate (thus difficulty) has been increasing at 2% per day.  This means that it doubles in 35 days, and then doubles again and again.  This behavior can't continue forever, but it has been remarkable stable for a while and appears that it will continue for at least the immediate future as more and more huge operations come online.

For an example of how this plays out on 2013-08-15 the network hashrate was 392.05 TH/s. Today it is 10392.91 TH/s. This increase of 26.5x corresponds to a compounded increase of 2.368%/day (so in case you were wondering in other threads why miners were willing to pay 55 BTC for 400-2000 GH/s miners to be delivered in October and are so angry that haven't shown up...).

If you're willing to assume the 2% model— which as I said, is clearly wrong in the sufficiently-long-term but is probably a reasonable conservative assumption in the immediate future— you can figure out the lifetime income from a device with a formula like:

0.014675 BTC * GH/s * 0.98 ^ D = Device lifetime income

where D is the number of days in 2014 until you receive the hardware. This has some minor accounting for power costs at the 1W/GH level baked in, but really the 2% assumption dominates. Under this formula you expect 2TH/s delivered 60 days into 2014 to have a lifetime income of 8.73 BTC. While the 2% assumption may turn out to be wrong— hashrate might grow faster or slower in the next few months— it's certainly a helluva lot better than just assuming a constant difficulty.

As far as "heavy miners"— really thats irrelevant. $/GH, arrival date, and W/GH are the only factors that matter much, and at the moment mostly the first two until the crazy growth stops. It easily can be more profitable to purchase more smaller miners if they're less expensive and/or more power efficient. When thinking about profit you need to think in relative terms.

[Carlton Banks]

If the OP is asking about the here and now, the answer is: none. Use the most optimistic (but still realistic) predictions of the future hashrate, and compare that with the potential profits to be had from the current coin price, and you might consider that the coins are the better investment. Or neither, but I'm personally bullish on the coin price going forward.

To answer the question directly, the answer is: Cointerra. But if you buy now, you probably won't make money. Buying months ago is, obviously, not an option. People tend to assume that being first in line, or copying successful mining gear investments from months ago will pay off. Instead, they end up with machines that will never break even. It's a tough environment to compete in.

[gmaxwell]

If the OP is asking about the here and now, the answer is: none. Use the most optimistic (but still realistic) predictions of the future hashrate, and compare that with the potential profits to be had from the current coin price, and you might consider that the coins are the better investment. Or neither, but I'm personally bullish on the coin price going forward.

In the days of GPU mining it could be argued that Mining was the less risky option... so even if the profits didn't look fantastic once you accounted for increases (and they've never looked _that_ fantastic with pessimistic assumptions) you could at least convince yourself that mining lowered the risk of taking a Bitcoin position because if Bitcoin went tits-up you could at least sell the GPUs. (If this was really realistic was anyone's guess— the market would have been flooded with used GPUs, but it wasn't a crazy assumption).

Today mining is specialized hardware that has real no other use beyond "space heater"— and space heaters are pretty darn cheap and also not noisy.  I can't think of _any_ way that a miner is less risky than just buying Bitcoin and there are quite a few ways they are more risky (risk of vendors not shipping, slower to liquidate in times of trouble, hardware failures, etc).  Of course, some risk is fine if it were profit producing…

To answer the question directly, the answer is: Cointerra. But if you buy now, you probably won't make money. Buying months ago is, obviously, not an option. People tend to assume that being first in line, or copying successful mining gear investments from months ago will pay off. Instead, they end up with machines that will never break even. It's a tough environment to compete in.

Cointerra currently appears to be a good company— and have responded admirably to their delays, but they've still yet to ship a product, and their nearest orders available are for April (and are a prices that probably won't enable a profit).

Right now, I _believe_, the best priced miner which is shipping on short notice is the prices Bitmain is offering on antminers to Chinese customers (Not, sadly, their reseller on the forum, who is charging 3BTC...), which is roughly around 1.9 BTC for 180GH/s. But to get that you'd be taking the risk of ordering through a chinese intermediary... it's also on 55nm hardware which will be very likely be power inefficient compared to the CT offerings.

[hardpick]

You are entering the market at a bad time
if you order to-day at about $800 a bitcoin - I believe you will loss money because all miners will not repay the bitcoin investment --- so best option buy bitcoins at $800 and wait till the price is over $1200 and sell .

I believe after 20 nm asics are released and the price of bitcoins goes up it may be time to buy mining hardware again  IE

about June this year


some useful links

http://decentralizedhashing.com/bitcoin-mining-equipment-table/

http://mining.thegenesisblock.com/

[Carlton Banks]

Today mining is specialized hardware that has real no other use beyond "space heater"— and space heaters are pretty darn cheap and also not noisy.  I can't think of _any_ way that a miner is less risky than just buying Bitcoin and there are quite a few ways they are more risky (risk of vendors not shipping, slower to liquidate in times of trouble, hardware failures, etc).  Of course, some risk is fine if it were profit producing…

I think we're starting to turn the corner with mining hardware companies, it's possible that some of this generation will survive long term! And that will be really important for the network long term. I'm slightly concerned we might end up in Skype-esque territory by default, where only big companies can afford to take on the risks of mining, and also reap all the rewards. I'm hoping that with hardware companies getting the opportunity to compete using higher production volume that a bit of market place longevity will afford them, combined with the propensity toward an over-saturated market, will help fend off these big corporate style mining ventures. But it's hard to see how that's really more than wishful thinking in the long term, I think it's going to take a home manufacturing revolution to keep the hobby miner in a position to compete.

Right now, I _believe_, the best priced miner which is shipping on short notice is the prices Bitmain is offering on antminers to Chinese customers (Not, sadly, their reseller on the forum, who is charging 3BTC...), which is roughly around 1.9 BTC for 180GH/s. But to get that you'd be taking the risk of ordering through a chinese intermediary... it's also on 55nm hardware which will be very likely be power inefficient compared to the CT offerings.

Agreed, although I don't think those units will break even if you buy now either. I ruled them out simply because I hadn't heard about them at the right time (and because I'm biased toward something with silicon that might survive more than, I dunno, 4 months before it's out of date? Hopefully 28nm will be good for at least 6 months). But they do seem like the best all-rounders to emerge so far.

You are entering the market at a bad time
if you order to-day at about $800 a bitcoin - I believe you will loss money because all miners will not repay the bitcoin investment --- so best option buy bitcoins at $800 and wait till the price is over $1200 and sell .

I believe after 20 mm asics are released and the price of bitcoins goes up it may be time to buy mining hardware again

This.

But judging when the price has peaked for the interim is a genuine guessing game. Should I buy today, or will there be another surge tomorrow? What if it crashes tomorrow! It's a tough choice to weigh.

[Kenshin]

Why am I unlikely to get help here?

Because we all accept a different level of risks. Some are more high risk taker, and some are low risk taker.

I personally is a high risk taker by nature. I like playing with dangerous extreme sports, and I deal with Bitcoin the same way.

I think Cointerra, Black Arrow and KNC are what I will bet on. I personally got a few KNC and already pre-order the next gen KNC product.

Cointerra and Black Arrow looks good. But due to the fact I am in the UK. All EU import doesn't have import tax. That is why I go for KNC.

So if you are in the US then Cointerra is the one I go for.

[gmaxwell]

Agreed, although I don't think those units will break even if you buy now either. I ruled them out simply because I hadn't heard about them at the right time (and because I'm biased toward something with silicon that might survive more than, I dunno, 4 months before it's out of date? Hopefully 28nm will be good for at least 6 months). But they do seem like the best all-rounders to emerge so far.

Yea, I've been a power efficiency is king guy in the past, but it looks like hashfast will not quite be 1W/GH/s at the wall (~435GH/s drawing 450w) so there is at least one of the new 28nm parties failing to make as much advance over KNC/Bitfury/Bitmain as anticipated. (Hopefully CT will— otherwise their 2TH devices may present logistical challenges).

[gmaxwell]

Because we all accept a different level of risks. Some are more high risk taker, and some are low risk taker.

There is risk, and there is just suicide.

What do you believe is the probability that a 2 TH/s miner that costs 7.5 BTC today and will be delivered at the end of April will ever mine 7.5 BTC? That it will ever mine 15 BTC?

Give numbers for probabilities on different amounts of return, we'll turn it into a bet, put funds in escrow, and you can buy lots of risk from me assuming that my probablities for positive returns are much lower than yours (which they'd have to be if you think buying these devices at current prices and pre-order lengths is wise).

My argument against the current prices has nothing to do with risk aversion. I will happily take on a risky investment (including bets over future hashrate with you, assuming the funds are blockchain-escrowed), it's a question of the distribution of risk vs profit. I can go to the stock market and get expected returns of 10% for a particular level of risk and use margin to multiply that up if its not risky enough or dillute with bonds... but there is a risk reward trade-off, and right now I think that the payoff matrix for a lot of mining products looks more like the payoff matrix for the intravenous administration of drain cleaner than a rational investment.

But hey, if you like crazy bets, I'm game to take you money.  There is no need to give up your funds to hardware companies and waste your time, noise tolerance, and power on some mining gear if you're just interested in taking a long shot bet on the future hashrate.

[U1TRA_L0RD]

Dont trust butterfly labs, they wont send you your miner until christmas next year.

Yeah, that's what I've heard -- also that they're a hot mess. Underperformed, draw too much power, failures, etc.

My buddy had one and his keeps shutting off because it overheats to quickly, BL wouldn't even accept it back.

[U1TRA_L0RD]

So IF the hardware mining makers are duping us out of our money, when the difficulty rises, there we go again buying more from them and waiting until the product arrives, and it will be too late again since the difficulty rose again.

[gmaxwell]

I've now offered to put my money where my mouth is for my claims in this thread:

"Thinking about buying an April Cointerra unit for ~8 BTC? Take my bet instead."
https://bitcointalk.org/index.php?topic=395243.msg4264552#msg4264552

[U1TRA_L0RD]

I've now offered to put my money where my mouth is for my claims in this thread:

"Thinking about buying an April Cointerra unit for ~8 BTC? Take my bet instead."
https://bitcointalk.org/index.php?topic=395243.msg4264552#msg4264552

Isnt that a risk, I dont like gambling BTC, But money yes