It is 2020 time for a new diff thread.

[philipma1957]

we are at an all time high in diff

Latest Block:   639097  (4 minutes ago)
Current Pace:   93.7036%  (26 / 27.75 expected, 1.75 behind)
Current Difficulty:   17345948872516.06
Next Difficulty:   between 17397042235313 and 19591094591950
Next Difficulty Change:   between +0.2946% and +12.9433%
Previous Retarget:   Today at 7:03 AM
Next Retarget (earliest):   July 27, 2020 at 7:38 AM  (in 13d 19h 57m 14s)
Next Retarget (latest):   July 28, 2020 at 5:38 AM  (in 14d 17h 57m 10s)
Projected Epoch Length:   between 14d 0h 34m 42s and 14d 22h 34m 38s

[alh]

It seems to me that as long as BTC price trades within a small range (say roughly $9100), it will take gear that is lower priced to replace less efficient gear. Most of the hash rate will move to the most efficient producers which include gear efficiency and electricity cost. Hig priced electricity pretty much precludes any significant mining in those areas, even with massively efficient. An upswing in BTC price will also push hash rate and hence difficulty over time.

The whole combination of BTC price, difficulty, electricity cost all come together to make the most efficient mining operations produce a SMALL profit. If they were to make a large profit, then others would join which pushes up hash rate and hence difficulty. Gear vendors can only charge a premium on more efficient gear for a while until thing settle.

There is no long term plan to make big profits with BTC (IMHO) mining. It's only done on the margins and for a short (i.e. a few months maybe) period of time. After the system will settle, with possibly a slightly different set of marginal profit makers. Folks with high electricity cost are effectively unable to mine.

[philipma1957]

lets use 1 s19 pro and 2 s19

they add to 110 + 95 + 95 = 300 th
watts.     3250 + 3250 + 3250 = 9750 watts +-5%. make it 10 kwatts for sake of easy math

so 300th x 0.0671 =  $20.13 usd a day as per current  viabtc earnings per th

10 x 24 = 240 kwatts a day

10 cent power = 24 usd  loser
 9 cent power =.  21.60usd       loser
 8 cent power =.  19.20.          0. 93 daily profit
 7 cent power =.  16.80           3.33  daily profit
 6 cent power  =. 14.40.          5.73  daily profit
 5 cent power. =.  12.00          8.13  daily profit
 4 cent power  =.   9.60         10.53  daily profit
 3 cent power =.    7.20.        12.93. daily profit
 2 cent power =.    4.80.        15.33 daily  profit
 1 cent power =.    2.40         17.73 daily  profit

these assume flat diff and flat price.

2 s19 and 1 s19pro cost 5977 plus shipping and import taxes if they were in stock and they ship in oct.  if you get a reseller in stock now they are about 8200 plus shipping and import taxes.

lets say you get all 3 for 8400 and they show this Friday .  lets say you get around the import and pay only 600 vs 2200. you are at 9000 for the 3 pieces

at

8 cent power. 9000/.93   =  9677 days to break even
7 cent power  9000/3.33  = 2702 days to break even
6 cent power  9000/5.73  = 1570 days to break even
5 cent power  9000/8.13  = 1107 days to break even
4 cent power  9000/10.53 =.  854 days to break even
3 cent power  9000/12.93 =  696 days to break even
2 cent power  9000/15.33 =  587 days to break even
1 cent power  9000/17.73 = 507 days to break even

to be fair I used reseller prices for this week delivery
if you wait for bitmain gear from Bitmain and not a reseller the price would be 1/3 less (maybe)

still numbers are nuts.

and yet the diff rises 

[alh]

Just out of curiosity, has there ever been s stretch of 507 days where difficulty has NOT risen, much less 700 or more days?

[philipma1957]

Just out of curiosity, has there ever been s stretch of 507 days where difficulty has NOT risen, much less 700 or more days?

we had flat spells of 200 days

maybe 330 days

oct 2018    7.183t

may 2019  6.705.

 this was a drop. and was about 210 days

[stompix]

With that being said, after a second thought on the situation, I strongly believe that difficulty has most likely hit the top, for now, starting from the next epoch my guess is that we start consolidating to the downside for a few months (unlike I thought a while earlier for many reasons which I will talk about in the near future), the correction won't be major so if you have an S9 and pay 5 cents per Khw, it's impossible for you to become profitable again, however, the adjustment will be a small push in profit, meanwhile, I wouldn't buy any mining gear at these current prices.

I waited a bit before replying to this, to see what the trend is at the end of the targeting period and not be completely off from the start. Even if things are not really going the way I expected I still stand on my previous assumption that we might still see an increase even with the insane ROI period and not a small one.
And this would happen not because your math is wrong or your reasoning but it might because by people who have not the same reasons, not the same plan, and are not making their plans on their own financial power.

One of the things that made me think mining will go the same way as shale, not really caring about today's prices or even the future is the news about Core Scentific with their purchase of ~17 000 miners, those are companies that received funding from investors, they present a business plan, they fail, it's not their money they lose. In the shale industry there are hundred like these, not risking their own money and the 3rd mortgage on their house, these guys can keep buying and putting hashpower online even if it earns them pennies, not even talking about ever ROI.

With no serious price swings which would upset the balance too much, I would venture and say we're going to see at least 50% increase in hashrate based on the same price by the end of year.

[philipma1957]

So if you are correct.
the insane number below will become 1.5 times worse

[...]

the earning number above will be not 20.13 a day with free power but 20.13/1.5 =

13.42 with zero other cost which does not exist
11.02 with  1 cent= free power+infrastucture taking 6000/11.02 to 9000/11.02  = 544 to  816 days. to break even
  8.62 with. 2 cent power/plus infra                          6000/8.62 to 9000/8.62     = 696 to 1044 days
  6.22 with. 3 cent power/ plus other cost                 6000/6.22 to 9000/6.22.    = 964 to  1446 days
  3.82 with  4 cent power/plus other cost                  6000/3.82 to 9000/3.82.    = 1570 to 2356 days
  1.42 with 5 cent  power/plus other cost                  6000/1.42 to 9000/1.42.    = 4225 to 6338 days

I know there are stupid people that invest, but if you are correct that price stays at 9200 and diff moves to 1.5 x 17.5 = 26.25 diff

then the industry will collapse. The numbers don't exist for right now no less with a diff of 26.25

No-one has a true free power setup so large independent farms are between 2-3 cents at best.
this is :

power cost
gear breakage
salary
building

[stompix]

I know there are stupid people that invest...

Yes, I'm putting all my bets in this scenario on pure stupidity, I know it's there, and I know there are lots of money there also, how many bubbles have burst in the past, how many industries have seen overinvestment with no demand, with billions poured in? All of Eastern Europe was flooded with ostrich farms, and people were raising thousands of them till they found out nobody wants to buy their eggs and meat. One such so-called business in the UK got £ 20m, one!

If we would talk rationally and on numbers, and if we think from the perspective of buying the gear with our own money, yeah, it makes ZERO sense. Even the last few increases make little sense but I'm seeing more and more news about companies betting on mining and as previously mentioned actually buying thousands of miners.

Would I purchase gear with 5-7 cents per kWh (not that I could get such a  deal anyhow), no, would you, no, would any on this board? Nope. But investors that throw millions on a company that only has a whitepaper would. Bitmain is sold out till October!! They are out of (finally) S9s! That gear is going to come online, it's either going to make a cent a day or get sold ...to somebody who is still going to plug it in.

Again, I'm not having my scenario based on reason, but on stupidity, and the biggest manufacturer having gear sold for 4 months in advance shows there is little reason left in the industry.

[alh]

I'll add my two cents to the discussion on the current ASIC pricing, BTC price, and difficulty. While I agree with Phil that the industry can't support the combination, that doesn't mean that won't continue. Round #1 of crazy investors buy gear from Bitmain at whatever Bitmain is asking. After some months bills come due, and the venture figures out that it isn't going to fly. They sell their gear at a major discount to the original price to Round #2 or somewhat less crazy investors. The round # 2 folks might make it, BTC price might go up, or after several more months, the Round #2 guys throw in the towel. They then sell what gear is still working at a discount to their cost to the Round #3 guys. Eventually the price of that once new gear has now gotten down to a more reasonable price and the Round #N guys are able to continue for a while. The various investors in first few rounds take a loss, which is essentially "paying down" the original price of the hardware.

I think eventually the push back up the chain to Bitmain happens and they drop the price. Hard to tell at what point, if ever, their costs for developing a new generation of gear becomes too high.

All of this happens within the context of difficulty changes, and BTC price changes. But nobody doing actual mining can actually get rich for very long. Something will adjust to essentially negate their advantage over time.

This overall process seems to be repeatable for an indefinite period of time. If for some reason BTC won't rise enough, then the hash rate and difficulty will have to adjust downwards to support whatever the rest of the industry is willing to provide in terms of blockchain processing.

[philipma1957]

the Oct 2018 drop  to May 2019 in diff can reoccur and  if price stays flat it will reoccur.

For now I don't make any buys and mine on with the 1.7 ph I control.

New gear is simply not very efficient.

S17pro on low = 35 watts a th with stock firmware.

Newest gear = 29 watts a th

A simple translation is not much improvement.

I have seen used s17 pros offered for under 1000 usd.

[mikeywith]

One of the things that made me think mining will go the same way as shale, not really caring about today's prices or even the future is the news about Core Scentific with their purchase of ~17 000 miners, those are companies that received funding from investors, they present a business plan, they fail, it's not their money they lose. In the shale industry there are hundred like these, not risking their own money and the 3rd mortgage on their house, these guys can keep buying and putting hashpower online even if it earns them pennies, not even talking about ever ROI.

You have a valid point, in fact I have talked about this point several times in this very topic, I simply categorize all of these groups under the "Newbies" who enter a losing business, the only slight difference between newbies who take bank loans or spend their savings to buy a few miners and those investors is that in general, investors will make better decisions in most cases, the reason is the fact that they have so much money which most likely is the result of their proper planning or in the worst case scnerio - they would hire a proffesinal who would make business plans for them.

17,000+ S19s might sound like the end of the world but it's merely 1.615EH or 1.3% of the current hashrate, yes Core Scentific are not the only host who bought a dozen of these gears, and yes there are a dozen of newbies who have no clue about anything and they keep buying expensive gears that don't make profit, but there can't be so much stupidity to push difficulty any further, or at least I still have faith in humen beings and that the majority of them will do the right thing, as both phill and I have brefiely explained, with such numbers it makes 0 sense to mine, if someone is gambling on the price, then buying the coin would be a superior choice, if you are counting on other miners to leave the network before you then you might get the wrong bet.

Of course as I mentioned my speculation is based on the fact that most people use common sense, but even if the majority were plain stupid, those who have more brain cells will be able to kick those with less brain cells out of the game because by the end of the day you can only lose so much before you give up.

[philipma1957]

[...]

I normally do not quote myself but I have been running these renewed threads looking for slots like the Oct 2018 to May 2019 slot to happen and predict them.

I did very well with the Oct 2018 to May 2019 move.

Of course. I also look for a monster bull run to come.

Harder to predict.

I can say this I feel one or the other will happen this year or early 2021

After 2020 USA election  if Trump wins

After 2021 Jan if Biden gets in and their is no coup by Trump.

Of course I am hoping Covid-19 does not kill us all.
Oh my wife had a lung x-ray again she has damage from Jan-Feb (Covid-19)

Doctor will setup a different regime to her up then what we have been doing.

So to all be safe this shit is real.

[stompix]

17,000+ S19s might sound like the end of the world but it's merely 1.615EH or 1.3% of the current hashrate, yes Core Scentific are not the only host who bought a dozen of these gears, and yes there are a dozen of newbies who have no clue about anything and they keep buying expensive gears that don't make profit,

Yeah, I know it's not much,

Assuming they are S19Pro, as I doubt they bought normal s19s with no discount it's around ~ 42 million but I assume they spend way less since they probably got a hefty discount. Nevertheless, as high as it might seem currently the whole farm will make for around 1.5% of the hashrate.

for my worst-case scenario of insanity, we would need another 30 like them..
But, they are indeed one of many, and a lot of those companies sit on a pile of cash with almost nothing real to spend on, you are a blockchain company you have 170 million from investors, what do you do? By Macs for the crew, a fancy office, you have to invest in something that you can actually show to the investors.

the reason is the fact that they have so much money which most likely is the result of their proper planning or in the worst case scnerio - they would hire a proffesinal who would make business plans for them.

Wow, interestingly that you mentioned this, actually the reason I went, how can you call it, bullish on the hashrate?
I recently got word about a business plan for a mining farm, I can't give details as my friend already breached the terms disclosing it to me but at one point in the study, they showed labeled as something like  (I have to reword this a bit as I said)  "maximum numbers of players on the market" where they took the daily reward, multiply it by the price of a kWh (they used 3.5Ecents), then showed how many miners would be able to run while not losing money.
As you can assume, the conclusion was...insane!!! They estimated there is a place on the market for almost 3 times more players !!!!
This is why I mentioned this possibility. There are companies that have money to burn, there are companies that lie to them with bs studies and there are greedy manufactures that would sell anything with fake advertisements.

Of course as I mentioned my speculation is based on the fact that most people use common sense, but even if the majority were plain stupid, those who have more brain cells will be able to kick those with less brain cells out of the game because by the end of the day you can only lose so much before you give up.

Well, I hope I am wrong, a total collapse with companies having 40-50% of the hasrate and shutting down going bankrupt and with extra gear lying around at scrap metal place is nothing I want to experience.

[mikeywith]

for my worst-case scenario of insanity, we would need another 30 like them..

Yup 30 of them will differently turn the mining game upside down, but the real question is does Bitmain and other major mining gears manufacturer have the capalbity of supplying that huge number of mining gears? does TSMC have enough chips to sell? even if the answer was YES! on both questions, then time factor will take care of the problem since they can only make so many gears, also increase demand on products will take prices way to high which will make late commers reconsider their options PLUS the internal issue with bitmain (Zhan Vs Jihan) doesn't seem to be ending soon, all of those factors combined make me feel "bearish" ( I like the terms you used btw  ) on hashrate.

of a kWh (they used 3.5Ecents), then showed how many miners would be able to run while not losing money.

Many people including so called "professionals" work with the wrong numbers, it's hard for someone who live in EU or US to believe that there are hundernds if not thousands of megawatts worth of gears that run on FREE power.

So in short, you think that many investors are so stupid and hashrate will keep rising, I think that most people aren't stupid and hashrate will either stay flat or more likely drop slightly for months to come (given that BTC price doesn't make any considerable highs), since I am a miner myself - I am of course biased towards my thoughts and i hope you will be proven WRONG. 

[philipma1957]

A lot of people don't understand that the guy that purchased 1 megawatt for 3 years and prepaid all the power is common all over the USA.

Our warehouse has a 3 year power contract for 1 megawatt .  It is paid off.  He lost a major freezer user that used 3 walk-in freezers year round.

This gave him 200 kwatts 24/7/365 for 29 months paid for not in use.  This happens a lot all over the USA.
People paid the power and now don't have a user for it.

Sure 200 kwatts 24/7/365 is a drop in the bucket.

it can run 2ph of s9's. since it is 'free' power this guy can run s9's and earn 2000 x 0.00000725  = 0.0145 a day or  133usd a day. it seems small but a lot of people can do this.

If you look at a pay out sheet for viabtc if they make 900 payments in 1 transaction 800 of them are guys in the 0.01 to 0.1

or 1.4 ph to 14ph size. I estimate there are 2,000 farms like this

say 1ph to 20ph on 'free' power

that is 2eh to  40eh  of people that can turn a profit on an s9

And they slowly swap out to better gear

we are now at 1.7ph.  almost all gear is 37 to 50 watts a th

The big investors can not win out over these miners very likely they will bust out and feed into the core of miners I am describing.

I have now seen the s17 pro used as low as 990 usd plus 60 to ship it or 1050.

So I still go with the large diff drop like what happened oct 2018 to may 2019 or a big price jump.

I am guessing stompix is wrong with his 9200 price and 25-27 t diff

But I have made mistakes maybe he is correct.

[Last of the V8s]

https://medium.com/@BitOoda/bitcoin-mining-hashrate-and-power-analysis-bitooda-research-ebc25f5650bf

interesting new report on mining, with a link to the full pdf

In our assessment, 50% of Bitcoin mining capacity pays 3c or less per kWh, a steady decline over the past several years. Anecdotal evidence suggests this number was closer to 6c / kWh in 2018.

[DaCryptoRaccoon]

That is a fantastic report with lots of solid info been looking for a report like this for some time great find LOTV8's.

Just shows really the amount of cheap power in china really has a monopoly over the network now and I worry that if prices in the west continue to rise the way they have been then it's going to put a lot of pressure on miners and hand control over to the east.

Thanks for the share.

[mikeywith]

A lot of people don't understand that the guy that purchased 1 megawatt for 3 years and prepaid all the power is common all over the USA.

Indeed, that's great point phill,  you would include those under the "newbies" who make profitability calculation off by a good portion, really it's like slowing down when the traffic light is green because you can't possibly trust that all cars from the other roads will stop for the red light, it's common sense, assuming that all other miners run their gears at the same exact conditions as you are is plain stupid.

[...]

That's a nice read buddy, despite the fact that it could be off by 50% i tend to agree to many points.

... we believe that the Bitcoin Network Hashrate could reach 260EH/s in 12 months and 360 EH/s in 24 months. However, this is somewhat reliant upon the price of Bitcoin appreciating or being expected to appreciate at 25–35% annualized.

So by July 2021 IF bitcoin price is at about $11,250 - 12,150$ hash rate should be 260EH or nearly double the current hashrate, sounds reasonable to me, but the price factor is CRUTIAL, taking their estimation of the average cost being 3cents per Khw, let's run some numbers.

S19         95TH  > $2150 for September batch > Makes $145 a month > ROI in 15   Months
S19pro    110TH  > 2900 for September batch > Makes $175 a month > ROI in  16  Months

The above numbers don't include pool fees, other operation and maitninace fees, it only takes powest cost of 3cents into account, these numbers are pretty terrible, assuming addition cost add 10-20% it will take about 18 months for these gears to ROI, 18 months is probably above their techincal lifespan anyway, and this assumes that difficulty states flat, it will be even worse with those miners who have FREE power keep adding more gears with 100% felixblity, with free power you ROI an S9 in 2-3 months, the incentive to mine at 0 power cost will always be there, it will only make thing works for anyone who pays a thing for power, even those 3 cents guys, let alone 4-5-6-7 cats.

The research is good, I wish it was done in the assumption that bitcoin price isn't going to increase, so if we have +100% difficulty and +35% price, if price is +0% then difficulty is +65% or about 200EH in a year from now.

[philipma1957]

I personally I think the median price is more then 3 cents for power.

Now If you look at it they add 10% for cooling and infrastructure which means 3.3 cents.

I also do not see broken gear anywhere in the calculation.

i do about 10% for broken gear.   so this means  3.6 cents is a better number to use.

not 3 cents.

But what do I know.

If you use 3.6 cents and price stays flat they are wrong.

We will be more like 180-210 eh.

[Searing]

I thought China was 1-2c kWh during these months of the rainy season (if the miners or the power or the Inet is not wiped out by floods in whatever/whichever the downstream province you may be storing your china ASICs at. Or is the above 3.3c kWh above best deal in USA/Canada the best in those countries you can pull off. I still think MASSIVE amounts of miners are mining for free on hydropower with electric utility handshake deals...the utils have electric in the rainy season they can't fully use and Bitmain/Innsilicoin/etc have miners..thus 'handshake deal' via rainy season. Utility supplies electric and Bitmain or whoever supplies ASIC equipment of whatever flavor and split the coins at the end of the month 50/50 till the rainy season is over in I think 2 months. win/win it seems..if indeed the 1-2c kWh in China is more or less correct with and/or alongside the 50/50 split arrangement above. Anyway ....like I know anything..but have heard the above as rumors off/on for last few years.

Brad