Seems a bit deceptive to use a rate of $20/gh. Considering bitmain is shipping devices now for $4/gh. And how many units are hashing now? The statement says feb 3 so it has been a month now. Another thing, why is icedrill trading btc on bitstamp? I thought this was a mining operation, not a gambling on bitcoin exchanges fund. |
|
|
|
|
It would take a massive amount of electricity for a company to use even a small percentage of the amount if hardware they could sell.
They can sell money printers for more than they will print in their life. And as someone said once you have the money upfront from selling the hardware you can make/sell another batch while the customers are left slowly hashing away.
Also nobody is saying that they don't use their own hardware because they all most likely do. Just that selling hardware is much more profitable than mining alone.
|
|
|
|
Have they announced who is going to distribute their products for the US?
Will likely be many places. AM said they would only sell small batches of working miners and mostly focus on chip sales. So we will probably see a bunch of miners using their chips. For the small batches I have no idea how they will be distributed. They seem to just randomly pop up. The cube was first seen on wtcr.ca and taobao so who knows. |
|
|
|
If power cost are at .04 and the mining rig is operating at 1-2 petahash it seems profitable even at the lower BTC value.
If you have an immersion cooled datacenter with next gen chips (0.2w/gh) then maybe. Current gen hardware will not be profitable and a datacenter with 1ph would require massive amounts of electricity. Also please buy a cheap miner to start off with. Not sure what is up with the amount of noobs who want to dump loads of cash on something they barely understand. Just to be clear, don't invest. You will lose. And calling it an investment is even a bit of a stretch. Gambling is more accurate. |
|
|
|
If I'm right most of them works on same algorythm. between them there is not much difference
This. Calculating mining profitability is rather easy for anyone not afraid of a bit of high school math. The thing that matters the most is what parameters you use, especially for difficulty increase. Many online calculators may use some pre-set value that, while user-changeable, most people don't adjust. Often this value is rather inaccurate, giving bad results. Before you search for a calculator, make sure you have a good idea of what you expect the difficulty to do in the future. Calculators like bitcoinwisdom will be better than using your own equation because it calculates based on difficulty change(some use monthly %). You could write your own equation to do the same but I think bitcoinwisdom basically has it down. With that said I would like to remind OP that historically every single calculator has been wrong and not by a small amount. Bitcoin difficulty is impossible to predict so it is important to remember that it is a gamble not an investment. |
|
|
|
Sold the GH/s at a low price, +-0 basically but whatever  Think the BTC will rise during the comming months anyway. Good choice. A minor loss is small compared to the loss if you would have held cex.io mining shares. |
|
|
|
I think you have 28nm confused with 20nm which just became available in the past few weeks.
Not in China. TSMC has a 20nm fab in taiwan. |
|
|
|
Get your crap spam out of here
Cointerra's ASIC hashrate is "greater than 500 GH/s?" False.HashFast's ASIC hashrate is "greater than 500 GH/s?" True.The objective truth is not "spam" regardless of your subjective personal emotional reaction. Amount of pissed off cointerra customers: ~1% Amount of satisfied hashfast customers: ~1% |
|
|
|
Sorry to hear more problems and thank you for sharing. I was actually about to order from Hashfast but ordered instead from another company last October, not sure yet if it was any better...
My experiences so far... Avalon Batch #3... they delivered finally (only) three months late.
Now waiting for Bitmine delivery which was promised last week of Dec, but nothing yet. Anyway already two months late.
LA3M unit from asiabtc / lightingasic came actually fast a bit over week after payment!
Generally buyingpreordering ASICs seems to be really risky.
FTFY |
|
|
|
The specs are what is called 'below the fold' - you need to scroll down and depending on your monitor size many customers will not even realize it is there. This is what the average customer sees:  Deceptive. Just because you have been able to push it to 800 for a minute in a lab using liquid nitrogen cooling or something else exotic does not mean you should give customers the impression they will be able to run at that specification. Just be honest with your customers. Don't act like a slimy car salesman. Thanks for the feedback. The "average customer" isn't in the market for $19,000 worth of raw boards intended for assembly into DIY miners, but is still capable of understanding the commonly used common phrase "up to." And no one (not even your theoretical "average customer") is going to spend $19,000 without bothering to scroll down slightly to finish reading a very short product description. The "improved cooling" HashFast refers to isn't necessarily immersion, much less liquid nitrogen. Maybe you still think liquid cooling is "exotic" but it's actually been mainstream for several years, and many off-the-shelf kits/components are widely available from several manufacturers at a variety of price/performance points. The chip is running at 750GH/s. That's a world record! Having a chip that runs 50% faster than the nearest competitor is a big deal, especially given the rapidly rising rate of mining difficulty. Wow. That is bullshit if I've ever seen it. Since when did having a larger chip mean anything? Shit efficiency means your chips will NOT be able to compensate for the rapidly rising difficulty. Hell, even bitfury 2.5gh/s chips are more efficient. A 28nm precision mining ASIC that hashes at four times the speed of its nearest rival
Even this statement is clearly a lie since cointerra has 400gh/s chips. Unless HF has a 1600gh/s chip I don't know about. |
|
|
|
Does AM frequently disappear from blockchain.info's mining charts? I don't see it on the piechart..
Yes because it has just enough hashrate to show up (1%). Even less than 1% recently. Will likely jump back to 10% with the launch of gen3. 10% is not quite possible, 5% is more possible. Cause Friedcat willing to sell most of total hashrate (400p-1600p) . We could see 10% solomining hashrate for a while right when gen3 is released. Considering 1 immersion cooled datacenter can hold 1ph or more I think it is possible. But I agree that they have no chance of maintaining 10% in a network with over 1000ph. 100ph would require too much electricity for one operation alone (unless its cex.io or someone else swimming in cash). But they don't need to mine with their own hardware. By selling it they would get more btc than it will earn in its lifetime upfront without the need to keep them running. So they may not be able to maintain a large portion of the network with their own solomining but with hardware sales they could easily earn much more than 10% would earn mining alone. |
|
|
|
Does AM frequently disappear from blockchain.info's mining charts? I don't see it on the piechart..
Yes because it has just enough hashrate to show up (1%). Even less than 1% recently. Will likely jump back to 10% with the launch of gen3. |
|
|
|
I'm still baffled on how AM predictions for power consumption are so low using 40nm process node. PFM.  Proof will be in the pudding. Till then its just wishful thinking. Can performance improvements be had in a revised hashing algorithm? Must be possible if a 55nm chip can be as efficient as a 28nm chip. I doubt it is as simple as one algorithm fits all. I know diddly squat about designing asics so I could be wrong. But yeah AM gen3 specs are unbelievable and if FC can pull it off this might be an interesting year for them. |
|
|
|
i read somewhere that because Russia bans bitcoin, metabank cannot produce bitfury? Is it true?
And that has to do with Black Arrow... how? It means our miners will survive a bit longer if true. Lets be honest, BA cannot compete with the efficiency of a nextgen bitfury considering it barely competes with a current gen bitfury (0.7w vs 0.8w/gh) |
|
|
|
Bitfury chips at 55nm can reach almost 0.7W/(GH/s) under-volt'ed. I believe ASICMiner could improve on that at 40nm with under-volt'ed chips. Although 0.2W/(GH/s) sounds similar to the bravado of all the other ASIC manufacturer's claims at simulation. Even using large scale immersion cooling.
0.2W/GH does sound unreal for 40nm but asicminer is known to deliver. Also I do not believe immersion cooling will have any effect on the chip efficiency. It is used to for efficient cooling with high density. Also something you forgot to take in to consideration in your calculations is that large scale mining operations (like AM) will be able to get electricity for a fraction of the price you pay. Possibly because there are no 20nm foundries in China, 28nm fabrication only just became available these past few weeks.
I think you have 28nm confused with 20nm which just became available in the past few weeks. Also is there any reason AM needs to use a fab in china? |
|
|
|
Okay, now I am confused as hell^^ If I would buy miners (in Stock miners) so that I have about 3-4 TH/s, that would be about 6-9 BTC per month (to some calculators out there) If I take the lower average of this number, 7 BTC per month. 7 BTC per month would be about 3850$ per month. A buddy of mine would buy the exactly same amount. So we would have again 7 BTC for him. So i Could buy every 3 month more mining power. But how much is the decrease? I know that the longer it goes on, the difficulty increases. I want to have at the end of the Year about 10 TH/s in power. With my buddy combined 20 TH/s. Would that not be enough of making a living out of it? Would we need more? Or is it just mining to make a profit in high balances suck?  Kind regards P You could probably end up with 20TH by the end of the year. Problem is you will be earning around what you pay in electricity by that time. |
|
|
|
in 2-3 months (...)
Learn the math first. or maybe you should research stuff before mouthing off - try some of advanced mining calculators which take into consideration difficulty rise - you're looking at jun-july tops, to be in green, after that you will not generate enough to even cover electricity cost. if you got access to free electricity good for you, but most people dont. You are mistaken if you think 28nm chips will not even pay for their own electricity use for an average kWh cost by June. Otherwise the network hashrate would shrink dramatically as even the most efficient chips couldn't even cover their own power costs  I remembered hearing this last September when average difficulty growth was much higher than now and said the same then. 28nm chips will stay profitable until a much smaller fabrication size chips is the majority of the network hashrate. Why the obsession with process node size? Asicminer 40nm will be twice as efficient as bitmines 28nm chips. Basically means there is much room for improvement/optimization left with 28nm chips before the efficiency has reached its limit. Proof or go home. Unless they use immersion technology on their hashing chips, they have no chance at all of even just being as efficient as Bitmines chips at 40nm. You should read about Moore's Law and why the race to use smaller fabrication sizes has so many billions spent on it. Edit: The only reason ASICMiner is using 40nm and not 28nm is to save money on fabrication costs, fact. Optimization is more important than rushing to a smaller process size. Bitmine 28nm chips on low power: 0.35w/gh Asicminer 40nm chips on low power: 0.2w/gh Although the AM specs are only simulated I think they will end up pretty close. Saving money as well as time is why they went with 40nm. Why spend 10 million in nre to make 20nm chips when you can make equally efficient 40nm chips? I have no doubt that eventually smaller process sizes will come out on top, but it is not with these chips(maybe v2). We will probably have 14nm chips by the time a 20/28nm chip is fully optimized. With all that said it does look like bitmine does have the most efficient chips as of now. Much more impressive than cointerra/hashfast. |
|
|
|
Edit: looks like the inefficient is due to overclocking. An antminer is 1w/gh at 80gh.
but that would involve more chips to achieve 180GH, costs are kept down by clocking the chips to the max performance. Yea but the fact is they can achieve the same gh/w ratio on 55nm chips. Bitmain can always keep lowering the price but they wont be able to compete with knc 20nm or asicminer 40nm. Unfortunately neither will BA. But we have a long time before efficiency plays a critical role so it might be better to go with a cheap miner like Blackarrow/Bitmine rather than an efficient one. |
|
|
|
ASICMiner is the company that made the Block Erupter USB, Block Erupter Blades, and the Block Erupter Cubes. They are working on a 40nm chip now for their next generation miners (which I hope will finally support Stratum directly).
40nm is already on the market on AntMiners.. I thought next gen chips are 28nm ?! BleackArrow also promise since long time ago to come with 28nm chips but all these may come too late when the difficulty will increase too much !! Bitmain/Bitfury 55nm are both more efficient than knc 28nm. KNC 20nm will be less efficient than Asicminer 40nm. What a load of bull. I have a 2 jupiters giving me 790GH for 820watts to power the boards, more like 900w with all the extra fans and control board. That is much more power efficient than Bitmains 180GH for 380w. Bitfury, yes they are power efficient. You are right about Bitmains devices. They claim 0.68w/gh at chip level compared to bitfury 0.8w/gh. Not sure why antminers are so power hungry but they are dirt cheap. Edit: looks like the inefficient is due to overclocking. An antminer is 1w/gh at 80gh. |
|
|
|
|
Show Posts
