Topic: It is 2020 time for a new diff thread. - page 8. (Read 14618 times)

@windjc, that is a lot of "IFs" which might not happen, I wouldn't count on stimulus to be a life-changing factor for bitcoin, a good portion of that money will surely end up being invested in bitcoin but it won't be something huge, we have a ton of other cryptocurrencies as well as a ton of other markets where most of that money will go to, the good thing about bitcoin, however, is its relatively small market cap.

U.S stock market cap = 27,000 Billion
Bitcoin market cap     = 200 Billion

so if the stimulus is going to be 2000 billion and 1% (20 billion) of that goes into stock, it doesn't add much with only 0.-07%, however, 20 billion for bitcoin is 10% which isn't that bad.


Watch the S&P 500, if it breaks its ATH (3% from current prices) then bitcoin will likely hit 16k by end of the year if the stock market fails to do so and start collapsing, we will see 9k again.

Current status:


The rainy season in China is about to finish, thousands of S9s will go offline by the next diff adjustment, let's hope that drop won't be wiped by all the S19/S19 pro miners.

hmm “we shall see what we shall see.”

Frankly we want betterment some how some way.

at the moment we have about 2.95% jump and are near 11500.

I really would like to see a strong move to ATH.

Biden is going to win which means that there is going to be a huge stimulus on the horizon. Markets will react positively to this. By 2022 we might see inflationary result from this. Inflation or fear of inflation would be the most potential bullish market condition for bitcoin. The US hasn't had inflation since the 1970s. During that time gold had its biggest rally in history by far.

What is even worse is that markets aren't showing any real signs of recovery, Covid-19 cases are hitting a new ATH almost daily



https://www.worldometers.info/coronavirus/coronavirus-cases/

It will probably take another year for things to slow down with the pandemic, and this could potentially lead to lower bitcoin prices or in the best case, flat price for an extended period of time, both scenarios aren't good for miners who paid a premium for the new gears, remember that these miners have a relatively short lifespan so even if you have free power there is no guarantee you will ever be able to ROI, this is why my main focus has been fast ROI.

I was talking to a large miner last night (in an attempt to get more information about how bitmain assemble their gears in Malaysia) and in the middle of our conversation we did talk about the exact topic we discuss here, and he brought a very interesting point which is the hydro season in China, many large players are counting on a nice difficulty drop by mid or end of this month when the rainy season in China is over and there is no more dirt-cheap power, we know that there are tens of thousands of gears which are now hashing at almost free power somewhere near the rivers in China, so it does make sense that these will have to go offline pretty soon, we don't know exactly how much the total hashrate there is, but I can only assume that it's at least 10-15%.

I believe a drop of 10-15% is almost certain by the end of the month, the question which remains is that are these S9s going to be replaced by other more efficient gears, in other words, will we have 10-15% of the hashrate leave the network and that's it, or will that be replaced by more efficient gears and thus keep the difficulty flat or take it even higher, I tend to believe that the drop will be more than the new hashrate, so we will very likely have a drop of some kind.

Having said that, the drop which I now expect isn't going to be a life-changer for those who paid 2.5-3.5k for S19/S19pro, but that should at least reduce their overall ROI by a bit.

And amazing as it is we are up 3%


No real improvement  for a miner. The stats for the s19 pro are

110 x 0.0744 = 8.18 usd a day earned as per viabtc.com

the s19pro burns 3.3 x 24 = 79.2 kwatts lets say 80 kwatts

so

8.18 - 8.80 = 62 cent loss at   11 cent power.
8.18 - 8.00 = 18 cent profit at 10 cent power. at a 2500 price for the s19pro 13888 days to break even
8.18 - 7.20 = 98 cent profit at   9 cent power. at a 2500 price "     "     "         2551 days to break even
8.18 - 6.40 = 1.78 usd profit at 8 cent power. at 2500 "  "   "     "   "    "         1404 days to break even
8.18 - 5.60 = 2.58 usd profit at 7 cent power. at 2500 ..........................         968 days to break even
8.18 - 4.80 = 3.38 usd profit at 6 cent power. at 2500 ..........................         739 days to break even
8.18 - 4.00 = 4.18 usd profit at 5 cent power. at 2500 ..........................         598 days to break even
8.18 - 3.20 = 4.98 usd profit at 4 cent power. at 2500 ..........................         502 days to break even
8.18 - 2.40 = 5.78 usd profit at 3 cent power. at 2500 ..........................         432 days to break even
8.18 - 1.60 = 6.58 usd profit at 2 cent power a 2500 unit .....................         379 days to break even

those numbers assume:
no breakage.
no maintenance cost.
no infrastructure cost.

they are grim numbers.

Actually I would say they are more grim for asic builders then anyone.

It's done, and it's almost flat with just -0.0858% which doesn't mean anything because that tiny drop could be the affect of luck and not a single miner has left the network, very bad news for those who are counting on either a drop in difficulty or a spike in price, flat price and difficulty works for me, but to many people, this only prolongs their ROI given that most of them think if the price doesn't go up then the difficulty will have to drop which hasn't been the case for quite some time now.

Current diff status:

just about flat with 60 blocks left.


To project what the most efficient gear does

1 s19pro 110 th about 110 x 0.0719 usd at via htc = $7.91 before power

it uses 3.3 x 24 = 80 watts rounded a day

10 cent power = 8.00 a loss
  9 cent power = 7.20 a 71 cent profit   not counting labor, maintenance , breakage and infrastructure
  8 cent power = 6.40 a 1.51 usd profit  not counting labor,   "                 "        "
  7 cent power = 5.60 a 2.31 usd profit  not counting   "
  6 cent power = 4.80 a 3.11 usd profit  not    "
  5 cent power = 4.00 a 3.91 usd profit  "
  4 cent power = 3.20 a 4.71 usd profit
  3 cent power = 2.40 a 5.51 usd profit

so if you got them for 2000usd and have 3 cent power 363 days to breakout only if your labor maintenance breakage and infrastructure is zero cost.

so basically a terrible way to do a business.

and it indicates no growth for diff (caveat price stays around 10-11k)

time will tell.

You are 100% correct on this one, we all have different ways to run our businesses and to each their own goals, mine has always been on the low-risk side, in other words, I always focus on fast ROI, and the only time I did something different I was screwed.

Telling people that I have 300KW to use and they all are S9 seems stupid, some would suggest that since my power is limited but free I should go for more efficient gears.

In order for me not to push things too much, I usually keep the load around 200ish KW, with just about 180 S9s in that farm, what I can alternatively do is use that 200KW for S19 pros for an example, and then I will have about 70 of them that hash at  7700TH instead of just 2500TH, I am sure many people would have gone with S19 pros or T17 at least, but my maths won't work that way, and here is why:

S9i 14th costs about 60$ shipped
S19 pro costs about $3500 shipped

This is $10,800 vs 245,000, in fact, even if I were to go with T17s, the maths will still favour the S9s in every single way.

Also, it isn't just the ROI factor, it is the robustness of these gears, in my other mini-farm since power is pretty limited to about 180amps, I can only use about 28 S9s, which is a bit less than 390TH and isn't worth the trouble, so I went ahead and filled with more efficient gears (a mixture of the 17 series) and now I have a ton of dead boards and PSUs, after having re-done the math, I would have been better of with keeping just S9s in there, but well, a lesson learned.

I plan to close down this mini-farm by next year and expand the other one, I would like to get a 600Kva transformer (won't be easy but doable) and I will only place S9s in there, my initial plan was to get rid of those Antminer 17 series and replace them with MicroBT M21s, but two things got in my way

1- M21s price increased after the 14k bull run and the shutdown of bitmain shop.
2- People here are heavily focused on ROI as well, so nobody is willing to pay $1000 for an S17 pro that does 52th when they can buy about 15*S9s for the same price and mine at 200TH, so I am stuck with these 17 series gears and looks like I am going to milk them until they die on me.

Well diff looks to drop a bit.  WE ARE about 300 blocks to go.

An interesting twist here is Potus and Fotus getting covid-19. Markets dropped a bit as more 2020 uncertainty has occurred.  A long serious bout with this covid for trump could shake market confidence a lot.

Picking diff is very very hard.


Not as hard as picking price.

I have run many many many diff threads 🧵.

I have some in site due to my power cost which is a 50-50 coin split.

Not to insult a free power guy like mikeywith but he will always prefer cheaply priced mining gear over new.

Not to insult a five cent power guy but he is looking for efficient newer gear and possibly with aftermarket firmware to be more efficient.

A big farm with 10,000 s19pros is not even looking at the power to hash to cost angle.

He looks at how many will buy into my company and at what price.

and a three thousand dollar unit with write offs etc is under 2000 cost for him.

bitmain is looking for suckers to buy their cloud mining contracts.

They all have a different business plan then a guy like me.
I would love to be all s9 but it screws the warehouse guy.
He would love me to be a s19pro and gpu mix mining at nicehash as this give me the most $ per watt.

So for me I look how to balance
$ per watt earnings vs
$ per hash vs
gear not dying too quickly.

So I look at all cases above and try to guess the diff.

always knowing that btc mooning = alters all calculations.

and always realizing that I dont over extend if price crashes.


The big farm that used a lot of OPM simply does not give a shit.

Here is why it was Other Peoples Money.

 If coin moons he gets rich.
If coins crash he loses other people’s money 💵.

I spend an hour or two each week trying to track new farms being built and trying to guess if they used OPM

opm guys wreck the diff. As they dont lose if it crashes.

Predicting future difficulty isn't easy, I too probably make more wrong predictions than right, simply put, the mining industry isn't rational, one would assume that paying $3500 for S19 pro which makes about $150 at 5 cents per Khw is something that nobody would do because this takes 2 whole YEARS to ROI at best case scenario, and there is zero guarantee that the miner will not toast by then, no matter how you look at it, paying this much is plain STUPID unless you are betting on the price to get extremely bullish, which in that case, buying the coin directly would yield better results with much less risk.

However, guess what? people are actually paying $3500 for S19 pro and $2500 for S19, they might not be stupid, they could have been laundering some money, tricking investors to invest in losing business, it could be any other reason but they all fall into the same category of "irrationality" which makes difficulty prediction a tough job.

Looking back at the moment I opened my big mouth and made a really, god how really stupid that prediction was, even with the huge jump we had and a bit of price retrace for the 12.5k high, we're still below the level when I said it will grow 50%. So, on the 15 July when I said that. the price was 9241, with a difficulty of 17,345T, now the price is 10903 but the difficulty is 19,314T with a downward trend, instead of the 20,465T to match the price growth, so rather than 50% up is 5% down, man I messed up so bad. Even if bitmain does deliver some October and November batch which seem sold out unless the price crashes there is no way for it to come true even by half of that.

As for the current stats, two days to go and we're 6 blocks behind, with a bit of extra miner luck we might see another 0.0x% change.

Yeah buying asic gear ⚙️ is next to impossible to justify at the prices asked.

Well I will continue doing my mining ⛏ in a mixed gear manner. Hoping for the best.

That's one hell of a jump given the price crash we had recently.

At this point mining with 6 cents per kwh hardly makes any profit on anything that is less effienct than the S19, heck even the S19 pro makes only 3.9$ a day at 6 cents, and 5.5$ daily at 4 cents, not sure in which universe does the maths for paying $3500 for an S19 pro actually work.  

Yeah. big move up.  we are now over 140eh. about 143 to be more accurate.  which was a jump from around 129 eh

so 14eh was added or

14000ph or

14000000th. which is 127,272 s19pros. that is a lot of gear

the oct s19 pros are 2407 each. so 2407 x 127,272 = 306,343,704 in sales

[2020-09-20 12:34:04.194+10] workinfo_add(): DIFF CHANGE: hi=649152 delta=11.35% new=19314656404097.0 prev=17345997805929.1

Ugh.. 11.35% jump

Yes, S19 and S19-Pro started being delivered last week.

Well he would need a lot of power and infrastructure to stealth mine.

          110th =         3 k-watts         1 machine
   110,000th =   3,000 k-watts  1,000 machines
1,100,000th = 30,000 k-watts 10,000 machines this is also 1,100 ph or 1.1eh. and 1.1eh is only .9% diff jump

so it would need to be 100,000 s19pro's

if I recall correctly those big orders are for 18,000 s19 pro's

let me see if I missed a zero.

1 = 110th
10 = 1100th or 1.1 ph
100 = 11000th or 11ph
1000 = 110000th or 110ph
10000 = 1100000th or 1.1eh
100000 = 11000000th or 11eh

and 11 eh would be a 9% jump . I am not sure this is just bitmain adding gear. They could have done some of it but I think whatsminer must have sold a lot of gear.

bit deer is bitmain maybe they are selling the contracts below.

https://www.bitdeer.com/pricing

It looks like the price drop didn't have any effect on difficulty.


Did bitmain deliver some of the pending S19/S19+ orders? hitting a new ATH with BTC price stuck in the 10k levels suggests that some people have got their hands on a large order of the latest gears.

What are the chances that Micree Zhan is actually mining with all the gears which they were supposed to ship? we know Micree Zhan control the manufacturer and warehouses, while Jihan Wu controls the headquarters and all the finance-related stuff, so if Micree has the gears and can't sell them on the official website, what stops him from using them to generate BTC or simply sell them directly to some local Chinese farms?

I tested quite  a few pieces of different gear since they did the 'locked' in style of psu's.

I settled on bitmain t17 s17 s17pro as best for my setup. and I have 20 of them total and 1 spare psu.

these all work fine after 16 to 9 months of usage

But the scattered  gear below  is annoying.

whatsminer
m10's
m20's
m21's  should be 246th is 190th and I have purchased 4 psus

avalon
a1047
a1066 should be 160 2 of each but  is 110 a dead a1066 due to psu

innosicion
t2 30th
t3 39th dead due to psu
t3 50th should be  119 is 80

almost every piece of gear above is pushed too hard. which kills the psu.

so the gear with issues should be 525th and is 380th  so  a loss of 145th

I am not the only miner with these issues.  this partially keeps the diff in check.

The issues above made me decide to get some gpus I spent 4000 in gpu to mine alts to convert to etc.

I use less power and I mine more coin. I have a 3 year gpu warranty and six months to pay with PayPal.

I mention this because in sept of 2017 I was typing very much the same words on older threads.

Then the rally hit.

so math shows 15 cards at 267 cost 4000 they will mine at 600mh for eth and use about 2000 watts.

add 3 mobos with cpus and ram 465 on ebay

4465+ 3 cheap coolers = 60 so for 4525 3 year warranty  six months to pay use 2000 watts

so the 2000 watts earns about 24.60

now buy 2 s19's for 4525 they use 6000 watts 17.01

This is a huge hedge factor.

As a miner do I buy 4525 in gpus and mine eth spend 2000 watts. earn 24 a day.

or do I buy 4600 in s19 pros and miner btc spend 6000 watts. earn 17 a day.

we have not seen this pro favor  gpu ratio since 2017.

These are very much pushing me at gpus.

Since I am a coalition miner and will mine any coin with btc/ltc/doge my key coins it is easy for me to shift new money at gpus.

Gpu dies =warranty
mobo dies = all kinds of cheap replacements
psu dies= my mobos work with get this apw3+ psus

Now I know history does not always repeat
and I know past performance does not guarantee future performance.
but it is unloading just like it did in 2017

just saying.