Topic: CoinTerra announces its first ASIC - Hash-Rate greater than 500 GH/s - page 106. (Read 231002 times)

These hail-mary hardware vendors just aren't interesting to me anymore. They have nothing to show, but ask for everything up front. Now we have Bitfury putting out cutting-edge miners at reasonable prices with minimal lead times. Bitfury has set the bar.

CoinTerra needs my money 4 month before they are able to deliver.

Okay I'll concede that the previous post was hastily written and wrong. Let's try this instead: Difficulty of 120m when you start mining, 20 GH will return 2 BTC in the first month. 0.3 of that 2 BTC = 0.6 BTC. 110 TH / 20 GH = 5500 platinum contracts ---> 5500 * 0.6 BTC per contract = 3300 BTC ~ $363,000. At retail rates, you're looking at 52 new Jupiters, for a total of 20.7 TH. Divide that among the 5500 platinum contracts and each is getting an extra 3.77 GH, which is a little less than 20%.

But CloudHashing isn't buying at retail, and BTC may be worth more than $110, and KnC is dropping their price in November. Having spoken with CloudHashing directly, I know that 20.7 TH added for month two is lower than planned by a large margin. All this to say that I do in fact expect a 30% increase in hashing power on my contract in month two, and this increase will slow month over month as the difficulty increase outpaces the GH growth, but not by as much of a margin as you're assuming, and if things do start to level off come January/February, difficulty growth could easily stop outpacing contract power growth.

Anyway, I have my bets, you have yours. We'll see how it shakes out. I'd be willing to entertain a modest wager that I make more than 10 BTC in the first year of the platinum contract, maybe 1 BTC; wouldn't want to have more than that locked up in escrow when it can be earning interest in coinenders or just-dice or via trading.

Ah man you didn't think that is what 30% revenue reinvestment means do you?  You did read the contract before sending them money.  You have 0% chance of making a positive ROI.

It means they take 30% of your REVENUE and buy hashing power with it.  Your hashing power isn't going to grow 30% a month.  Your $1000 bought you 20 GH/s of hashing power.  They take 10% as a fee, reinvest 30% towards more hardware and payout 60%.

Not sure why you think that means you hashing rate will grow 30% a month.  You got robbed and have no possible chance of a positive ROI% under any realistic scenario.  I know I won't convince you but your account stats in October will.  The good news is you won't have to wait 2 years to know you will lose money it will be obvious very good.  You can't pay 300% markup on hashing power and then somehow magically come out ahead by reinvesting.  They aren't giving you any hashing power for free they are using YOUR gross revenue to buy more.  The only way 30% revenue reinvestment = 30% more hashing power would be if your revenue for the month was 100% of your contract price.

I'm assuming you mean PH, not TH. I think that their posted projection of 25 BTC in the first year for a Platinum contract is high, but I think that 15 is probably low. They are offering 30% revenue reinvestment, so if you take as a floor number a 30% increase in hashing power month to month you're looking at 20, 26, 33.8, 43.9, 57.1... 466 GH/s in September 2014. This assumes no increase in BTC value (and thus the buying power of the BTC for new equipment), no decreases in prices, and no improvements to the tech being bought (Gen 2 KnC anyone?).  The platinum contract is also a 2 year contract, but I am looking at the second year as bonus-round.

Ok tell yourself that.   
Ok tell yourself that.  Their projection is based on 2TH to 3TH in Sept 2014.  We will likely see 5 TH/ by January.   I mean just about any rig is massively profitable if difficulty scales linarly between now and Sept 2014 with a max of 2 TH/s.   

For example Cointerra rig:
http://mining.thegenesisblock.com/a/92097362ab  Wow 112% return.
 
The host can't reinvest more revenue then it makes and with $50 per GH/s you are dead before you start.  The only way they could make a compelling projection is with totally impossible scenario of only 2 TH/s in Sept 2014.  If that happens well it is pretty much impossible not to make a fortune no matter what rig you pick.

We're derailing this thread, but you're wrong. The mining power on those contracts will grow very quickly, that $50 per GH number is extremely misleading because of how it will grow month after month; I have had private correspondence with them and am not quite sure how much of what they told me should be disclosed, but suffice it to say that they will be adding a substantial amount of power every month that will translate directly into increases in the hashing power of each individual contract.

Any miner can put money aside with any rig, but you need to be a very large scale miner to buy new hardware every month with that money. And you need priority queuing and bulk rate prices to compete with the scaling of CloudHashing.

Revenue reinvestment simply means putting aside some of gross revenue to pay for mining hardware.  Any miner can do that with any rig.  Also you almost certainly have no chance of positive return given CloudHashing price is >$50 per GH/s and hasn't started mining yet. 

I don't think the growth rate will slow down substantially until we start approaching theoretical maximums and Moore's Law; don't know what that limit is but I don't think we're anywhere near that yet. KnC has said they are launching a "Gen 2" device in March, given that their Gen 1 devices are 400 GH who knows how big and cheap those Gen 2 devices will be. Honestly it sucks to be in the mining hardware buying game, I bought CloudHashing contracts instead because of the revenue reinvestment, which based on my own due dilligence will be much more profitable over a 12 month time frame than any miner purchase because of how they are buying and reinvesting at scale.

Avalon and ASICMiner will go first but I think the network will grow slower that some project.  Here is a discussion based on probable preorders and timelines with a couple different viewpoints.

https://bitcointalksearch.org/topic/guesstimate-thread-for-total-asic-pre-order-hashing-power-278384

I wouldn't say those techs aren't missed.  GPU was ~50 TH/s.  They are effectively obsolete now @ 500 TH/s it was a 10% drag on the network growth rate.  Avalon & AsicMiner won't go "obsolete" (@ current exchange rate & $0.10 per kWh) until 17.9 PH/s.  Even with no new orders that would mean >1.5 PH/s of capacity (plus ASICMiner retails sales which are hard to pin down).  That is around the same share of the network.

Any word if a 500Gh/s (single chip) product will be offered directly from CT?

So the hardware is free, yes? No one has to factor in ROI? Okay.

I think that your point about CPU and FPGA actually folds on itself to some degree, because when CPUs and GPUs and FPGAs come offline, it's like a tree falling in the woods that no one hears. By the time they're so unprofitable that they stop getting used, they are already too small a portion of the network to be missed. I think that Avalon is relatively near succumbing to this fate based on what I know about preorders of 28nm chips.

Bitfury's developer is an amazing ASIC developer.  He also had efficiency unmatched by anyone else on FPGA and really only lacked a commercial success because of the false promises of BFL (both on FPGA side and early announcement of ASICs in a few months).   I would also point out that the rigs are essentially unavoidable underclocking.  They don't want to underclock but the chips are running slower than spec and that is going to improve the efficiency.  I am sure if you underclock or undervolt 28nm devices you will get improved efficiency (at the expense of less hashpower per $).

Still at stock clocks I don't think anyone is going to massively (<0.4 J/GH) outperform.  All the 28nm builders are taking pre-orders.  It is a competition for pre-orders and funding.  If they felt they could with high confidence say they can deliver <0.5 J/GH at the wall they say so because it would boost sales. 

Process node is only part of the equation.  KNC (28nm) is only guaranteeing 2.5 J/GH which is 4x worse than Bitfury despite being on a smaller process and BFL (65nm) is on a process very close to Bitfury but needs 5 J/GH which is 8x worse.  We can only go by what they unreleased specs and they are likely hedging their numbers to avoid an embarrassing miss but if KNC simulations were showing them 0.5 J/GH they wouldn't be building boards capable of 320W max for a 100 GH chip and only advertising better than 2.5 J/GH.  Still if Bitfury design can be shrunk to 28nm with similar efficiency it could be the most efficient chip yet.

That would be between 320 and 360 GH/s at 250W = between .7 and .8 J/GH on a 55 nm product. I seriously hope the 28nm products can beat Bitfury's 55nm by a significant margin.

To a certain extent but everyone has a pain point.  Are you still CPU mining Bitcoin?  Why not?  You already own a CPU.  It will only cost you $2,000 or so in electricity to mine 1 BTC.  If BTC exchange rate goes up to $3,000 you will profit.  So why aren't you mining on a CPU right now? (You already know the answer)

236 PH/s would mean an Avalon or Block eruptor would require $660 in just electricity to mine one BTC.   Sure at close to the break even point many miners will hang on but mining at a guaranteed 50%, 70%, 90%, 99% negative margin for week after week, month after month.  Nobody is going to do that.  Just like nobody (or at least not enough to be more than a rounding error) is CPU mining Bitcoin today. 

Yeah. People that buy mining equipment that'll  never ROI may continue to mine on the hope that BTC value will skyrocket. It may be worth the extra few hundred dollars of electricity a year for them. Big operations may shut down but I can see average joes continuing to mine, which would help decentralize hashing power anyway.

http://www.anandtech.com/show/7246/the-rush-to-bitcoin-asics-ravi-iyengar-launches-cointerra

above article has lots of info


"Q: What level of pre-orders are estimated before CoinTerra is able to ramp up production?  It has been listed in the media that a number of pre-orders are needed before final manufacturing can take place.

A: When we started looking for investment, it took two months to raise $1.5m from both Bitcoin users and the tech industry – at this point in time there is no need to ask for more investment.  In terms of pre-orders, within 24hrs of our website going live we had 150 pre-orders.  Our production is not hampered by sales in any way."

anyone been checking order numbers?  I'm assuming those are not PAID orders.  

Don't get me wrong - I somewhat believe 200 or 300 PH/s is science fiction, unless a very large percent of miners are not paying for electricity, which I doubt.

But don't forget 2 very important factors:
1 - BTC price can skyrocket. $1000 BTC price changes the landscape dramatically: with 1USD per GH/s efficiency it equals 1388.88 PH/s network speed for electricity only break even.

2 - People are greedy in nature and a lot of them will keep mining long after it is unprofitable, hoping for an increase in BTC price.

It isn't just the ATX PSU.  If a chip operates at 0.55 J/GH how are you going to power it?  Plug 12V connector from power supply to it and blow it up?  Most ASICs run at 1V.  ATX doesn't supply 1V.  So you need a DC to DC PSU to convert 12V to 1V.  Good ones are roughly 90% efficient.  So the wattage for the ASIC boards are going to be >0.55 J/GH even if the chip is 0.55 J/GH.  Now how are you going to cool up to 1400W of heat? Fans consume wattage, as does the system controller.  So the overall DC system wattage is higher than the board wattage which is higher than the chip wattage.  Now to convert AC to DC you are talking another ~10% inefficiency so the AC wattage is even higher.

Still it looks like you just assumed everything is operating at 0.7 J/GH.  That ignores the petahashes of existing hardware which is much less efficiency. Long before you get to 200 PH/s for example every 55nm or higher chip will be operating with negative gross margin.  Why would they continue to mine to turn $100 in electricity into $50 in Bitcoins?  Simple they won't and when they go idle it will slow the growth of the network.

200 PH/s isn't any more realistic.  It assumes that all existing hardware goes into the trashcan and despite hashrate already being so high that the break even point (even assumming no more hashrate growth) is years in the future people keep buying more and more rigs month after month until they are negative ROI% from day 1.