2021, time for a new general & diff speculation thread...

[kano]

2021-09-08 00:05 UTC workinfo_add(): DIFF CHANGE: hi=699552 delta=4.54% new=18415156832118.2 prev=17615033039278.9

Price dropped a lot 10 hours ago also (around 9% drop), so will be interesting to see if that drop leads to a diff drop this next diff, or if the price will recover.

[mikeywith]

people were mining at around 20T back in May and June and the price crashed several times under 30k

Different times - different people my friend, one of the reasons why I keep mentioning that it will be extremely difficult for all the ex-China gears to come back online is the power cost, back in May-June there were tens of thousands of mining gears running on Hydropower in Sichuan with power rates as low as 1-2 cents, do you know of any other place that has the same capacity and power rates? I don't.

This means your theory is likely invalid, the price decline post-China-ban will have a lot more impact on the hashrate, the higher the average power rate the stronger the effect of the price is, of course, that doesn't mean we will have a difficult drop because the price dipped 10% or so, but it means that with prices under 30K and diff over 20T, things won't be anything similar to May-June.

[stompix]

Different times - different people my friend, one of the reasons why I keep mentioning that it will be extremely difficult for all the ex-China gears to come back online is the power cost, back in May-June there were tens of thousands of mining gears running on Hydropower in Sichuan with power rates as low as 1-2 cents, do you know of any other place that has the same capacity and power rates? I don't.

Middle June the Chinese drama was over, the hashrate bounced back between 18 and 31 by 6%, the people who were still mining then would have no reason to stop mining now, the revenue is still higher per TH, and they weren't being evicted from anywhere anymore. I wasn't talking about people adding more hashrate on what we have now, but about the ones that were mining after the Chinese ban, for the survivors there is no way this dump would have triggered an operational loss, numbers are still better than July!

My thought was that right now, there is no way even minimal above 40k prices would trigger any serious downtrend in hash rate, although I would sure love to see the end of the epoch like this  :

Latest Block:   699636  (2 minutes ago)
Current Pace:   91.8681%  (85 / 92.52 expected, 7.52 behind)

110 x 30cents is 33 a day very over priced even with free power.

Even worse, cut 10 TH out of it as those are 100TH/s. Still available? Nobody buying?
If only I had a coupon... . Damn, I would rather buy doge for 10k than this!

[philipma1957]

Well we had a sept price slump in 2020
11.5 k aug 2020
10.3 k sept 2020
30k dec 2021

and we had a sept price slump in 2017
4600 late august 2017 to 3500 sept 2017 then took off in dec 2017 19k

both times we took off in the fall.

maybe we take off.

I just got 8x 30 amp circuits put in one spot we mine at.

[icopress]

Well, you indicated three-year intervals, so it is possible that soon we will see an even deeper correction.

And I have a feeling that the decrease in the mining difficulty against the background of the Chinese news was not the last, because if we compare the phrase "Price is increasing faster than difficulty" and the daily chart of the Bitcoin price, then I would not expect new difficulty peaks in the foreseeable future. Although as long as the price of bitcoin is above $ 40,000, mining is still very highly profitable, so the next difficulty adjustment is likely to be positive as well.

[alh]

I also think that there has been some hashrate in the USA that's been dormant because of hot summer weather in the country. As the weather cools some of that will get switched back on. I don't know how much that is in aggregate.

[mikeywith]

Middle June the Chinese drama was over, the hashrate bounced back between 18 and 31 by 6%,

Not quite like it, the drama is something and the actual drop is something else, the real drop in difficulty started exactly by the of may all the way to the end of July, the 19th of July marked the lowest difficulty, which means the Chinese were still in the game up to the point in time.

My point is that there are probably a dozen mining gears running at the edge of being profitably at current prices, old mining gears that were bought when diff was at 13-15 and the price was at around or above 40k, those folks running these gears at high power rates will soon be forced to shut down if difficulty keeps going up and prices don't.

The demand on S9s and other old gears was pretty high post the China ban, I was following the second-hand market very very closely, and it does seem like many people failed to run the math, they probably bought those gears under the assumption that prices were not going to fall and difficulty wouldn't go up as much, I am not convinced that all of those gears are running on cheap power, they will soon be making dust and will be forced out of the game.

Do I think that will have a larger impact than other effienct gears combining online? maybe it's not likely, in other words, I don't think we will see any large drop in difficulty, but those people being forced to shut down will somehow limit the difficulty spikes coming from other places.

Now my thoughts in regards to the price are that we won't be seeing any new ATHs any time soon, we won't have a happy year-end and prices will be either in the same range as they are now or much lower (20-30% lower), I have a feeling that many miners will be disappointed by the outcome of this year, especially those who bought mining gears recently, I sadly believe things will only get worse from now on, the great times are probably... over.

[kano]

Meanwhile with the 9% price drop, and still dropping more, the block rate 'seems' to have dropped also.
The numbers, of course, could well be wrong, since the number of blocks are way too small,
but at 60 blocks into the diff change, the rate was showing at +3.5%, now 140 into the diff change the rate is showing at -3.3%
Give it another day and who knows where it will be

Edit: of course, -3% is -4EH, so not some tiny amount that some random theory with nothing to back it will account for
... 40000 100TH/s miners ... or a lot more when including smaller ones ...

[stompix]

And it's back to normal levels

Latest Block:   699751  (10 minutes ago)
Current Pace:   98.4131%  (200 / 203.22 expected, 3.22 behind)

There were 3 intervals of ~30 minutes with no blocks right at the start, that's what's probably to blame for those swings at the start but probably as we pass another 3-4 days it will stabilize around 0, hopefully.

My point is that there are probably a dozen mining gears running at the edge of being profitably at current prices, old mining gears that were bought when diff was at 13-15 and the price was at around or above 40k, those folks running these gears at high power rates will soon be forced to shut down if difficulty keeps going up and prices don't.

I got the point now, the tiny difference between being bought at the x moment and being online and running at x.

Do I think that will have a larger impact than other effienct gears combining online? maybe it's not likely, in other words, I don't think we will see any large drop in difficulty, but those people being forced to shut down will somehow limit the difficulty spikes coming from other places.

Agree on this, with the mention that I think you overestimate the people who haven't done the math right.

Now, taking into consideration that since now money is involved and from totally not caring about miners revenue and just wanting the difficulty (security) of the network to grow and spread to other countries I'm on the opposite side and pray it stays the same or even drops so somebody neutral would probably find everything I say gloomy beyond redemption:

- assuming miners in Sichuan were mining at the end of the last rainy season at 0 costs, the revenue per TH was 8 cents, an S9 would have earned you $1.16 a day, now with current prices (as I type, god knows what those will be one hour from now on), an S9 with 10cents per would still make around $1.3, so if I'm doing the math right it would need bitcoin at under $32k to turn unprofitable.
Of course, if you would have paid 800$ for this it will never ROI, and it would be totally stupid to buy now, but we're talking about gear that is online, it makes no sense to turn it off under these conditions.
- hobby miners that would be stupid enough to buy 3-5 of those would also prove a lot more stubborn to shut them down, I know the feeling, even if at the end of the month you're down 30-50$ you will keep mining because...something, I see them more resilient than even large farms.
- I think we're focusing too much on old gear and too little on the new one, I went through the press releases of Riot, Mara, DMG, they combined got 80k S19 pro or j or whatever in Q2, those are enough to offset half a million of S9s alone, and these guys mine at below 4 cents, plus they mine on investors money, which sometimes can make both ROI and costs irrelevant

That said, I don't see that much second-hand gear coming online, even less being sold, but neither do I see much of the one already deployed going off until (if) we go well below 40k.

[philipma1957]

Well

https://www.bitrawr.com/difficulty-estimator

Latest Block:   699770  (3 minutes ago)

Current Pace:   99.7991%  (219 / 219.44 expected, 0.44 behind)

Previous Difficulty:   17615033039278.88                           
Current Difficulty:   18415156832118.24                           
Next Difficulty:   between 18451492817188 and 18593367467193
Next Difficulty Change:   between +0.1973% and +0.9677%
Previous Retarget:   last Tuesday at 8:05 PM  (+4.5423%)
Next Retarget (earliest):   September 21, 2021 at 8:13 PM  (in 12d 11h 33m 55s)
Next Retarget (latest):   September 21, 2021 at 8:45 PM  (in 12d 12h 6m 10s)
Projected Epoch Length:   between 14d 0h 8m 20s and 14d 0h 40m 35s
...

+ or - 1%

and prices over 45k https://coinmarketcap.com/currencies/bitcoin/

lets just do this for the rest of the year.

[alh]

..........

+ or - 1%

and prices over 45k https://coinmarketcap.com/currencies/bitcoin/

lets just do this for the rest of the year.

I am quite certain that the vast majority of folks that are looking at BTC as an "investment" have little interest in a a moderately static price of $45K to $50K. I still read articles that are pushing BTC price of $100K by years end. They may well just out for the "pump and dump" but if BTC breaks to $65K or higher, difficulty will follow in the same direction until every single S9 is restarted.

[philipma1957]

..........

+ or - 1%

and prices over 45k https://coinmarketcap.com/currencies/bitcoin/

lets just do this for the rest of the year.

I am quite certain that the vast majority of folks that are looking at BTC as an "investment" have little interest in a a moderately static price of $45K to $50K. I still read articles that are pushing BTC price of $100K by years end. They may well just out for the "pump and dump" but if BTC breaks to $65K or higher, difficulty will follow in the same direction until every single S9 is restarted.

Yeah true. But 140 eh of mining gear is very happy at this.

So far 2021 has been a very good year for miners. Not as good for johnny come lately traders.

Ie if you purchased first coins this year it is very likely you are a loser as many days are well over 50k price maybe 75 days over 50k and 100 over 46k

[mikeywith]

Agree on this, with the mention that I think you overestimate the people who haven't done the math right.

Maybe you are right, I could be overestimating these figures, I just can't see why I shouldn't. lol, I mean I do believe that most folks do NOT run the math right when it comes to mining, it's not hard to see, people paid 15k for S19s, 22k for L3s to receive some time in the future, they are still paying 2.5k for a lousy USED T17, I can't help but think that these folks are doing the math in the strongest way possible.

On the other hand, I can't tell for sure if after a few months these decisions people make now will seem like they were the best to do or not, who knows maybe 4 months from now BTC will hit 150k and difficulty drops 40% and then paying 15k for an S19 pro won't sound anywhere near WRONG as it does now, everything is possible, but as it stands right now, and since I do believe difficulty will only go up while the price won't, I can't help but think that these guys will end up losing and leaving the game.

Also, I don't think these guys make up a tiny 1% or 5%, nop, I think they represent at least half of the miners on planet earth, the once that gradually will largely offset the new mining gears added by those large cooperations, speaking of which, has anyone done the math of the total hashrate that the likes of Riot own?  

[stompix]

Also, I don't think these guys make up a tiny 1% or 5%, nop, I think they represent at least half of the miners on planet earth, the once that gradually will largely offset the new mining gears added by those large cooperations, speaking of which, has anyone done the math of the total hashrate that the likes of Riot own?  

Core Scientific doesn't say anything about hashrate but:

"Through our last purchase agreement with Bitmain in December 2020, we became the largest blockchain infrastructure and hosting provider in North America with 76,024 units in our fleet," said Kevin Turner, President and CEO of Core Scientific. "Now, just 4 months later, we have more than doubled our fleet, and intend to continue to expand our self-mining operations throughout 2021.

DMG might be around that:

As DMG secures further orders to fully occupy its 85MW flagship facility, the Company will explore the idea of multiple other Bitcoin mining sites to allow for additional hashrate growth in 2022 over and above the previously announced 2021 target of 2.0 EH/s.

Riot:

At full deployment of 20,140 next-generation miners, Riot estimates its total operational hash rate capacity will be 2.007 EH/s (2,007 PH/s) with the Company's aggregate mining efficiency of 32.65±% 5 joules per terahash (J/TH), consuming 64 MW of power.

*Surprised about this (June numbers but still)
Mara:

Existing mining fleet consists of 22,412 active miners producing approximately 2.3 EH/s

Blockfi/Blockstream, in every article they won't release any numbers.

Bitdigital:

Our 32,500 miners produce a hash rate of 1,920 PH/S

Digihost

Digihost Technology Inc. provides an update on its acquisition of 9,900 Bitcoin (“BTC”) miners (the “Miners”) that will increase the Company’s current hashrate by approximately 925PH to 1.145EH

Hut8

The anticipated production capacity increase with both purchases upon full deployment is expected to be 1.081 E/H, which alongside previously announced investments, should bring Hut 8's total installed capacity to approximately 2.5 E/H.

Hive:

With the addition of these 3,019 MicroBT M30S miners, HIVE's aggregate operating hash rate increases from approximately 566 PH/s to 830 PH/s.

Bitfarms

Bitfarms  announces  the successful installation  and operation of  its previously announced  order  of  4,500  MicroBT  M31S+  miners,  bringing  the Company’s current installed hashrate  to 1.2 EH/s

And I'm going to stop for now as I'm starting to laugh as I run through press release after press release, just how many of those are out there?
I'll make a sheet with them probably during the weekend as this thing is starting to really intrigue me.

Now, that list aside, I saw something interesting in Riot PR:

The Coinmint facility draws a significant portion of its energy from hydroelectric sources and the annual average power generation cost has been approximately 1.5 cents per kWh.

Damn...

[philipma1957]

https://www.bitrawr.com/difficulty-estimatorLatest Block:   699910  (a minute ago)

Current Pace:   96.9713%  (359 / 370.21 expected, 11.21 behind)

Previous Difficulty:   17615033039278.88                            
Current Difficulty:   18415156832118.24                            
Next Difficulty:   between 17980783508486 and 18271946723075
Next Difficulty Change:   between -2.3588% and -0.7777%
Previous Retarget:   last Tuesday at 8:05 PM  (+4.5423%)
Next Retarget (earliest):   September 21, 2021 at 11:29 PM  (in 11d 13h 42m 9s)
Next Retarget (latest):   September 22, 2021 at 6:35 AM  (in 11d 20h 47m 31s)
Projected Epoch Length:   between 14d 3h 24m 17s and 14d 10h 29m 39s

I like this a good indication of a large marginal mine either powered off or maybe upgrading.

prices are 45-46k

A drop back to 17.9 diff would be good for all established farms.

We have seen a lot of evidence that

chips are hard to come by.

That there maybe about 5eh of s9 hash in the hands of a big player.

Keep in mind the LTC/DOGE L7 is a big money earner in terms of $ per watt
of course the ETH asic gear also earns big money in terms of $ per watt

and the s19 pro earns less in terms of $ per watt.

Since bitmain is a really big gear producer of all 3 above. they may have come to the conclusion shift all 3 around don't mine against yourself.

Ie selling the gear makes less money then shutting it off and shuffling it around. Based on $ per watt.

Everything since May shows that all 3 major hash rates are locked down and controlled by very big players.

does anyone know where Jack Ma is?

https://en.wikipedia.org/wiki/Jack_Ma


My guess is someone else runs bitmain and that they have locked down the hashrate for the 3 biggest algorithms.

If you were China government you could do this.  Real gear and chip restriction. To the entire world.

Now this is all speculation but hashrate and gear production is very very very very tightly controlled at this moment.

[mikeywith]

Core Scientific doesn't say anything about hashrate but:

I found something

When added, these S19J and S19JPro mining units will bring the total 19 Series computational power hosted with Core Scientific to over 18.1 exahash.

This includes all orders CS made, which is a total of 188,824 S19j and S19 pros, keep in mind that they don't actually have these gears up and running, they are slowly getting them as they come, I found no exact date for when the last gear will be delivered but CS said:

With that purchase, Core Scientific’s total equipment on-order would amount to 188,824 units expected to be delivered throughout this year and next year.

This means this will be probably be completed by end of next year judging by the size of the order.

So all these guys combined make a little less than 30EH, of course, some of them will expand by 2022, but the 30EH is already accounting for full delivery for CS huge order, since there are probably a few other larger players out there, we can speculate the total these guys combined will have by the end of 2022 will be roughly 50EH and what they actually now is around 25EH?

Honestly, these numbers are larger than what I thought would be, but still, they are not even close to owning half of the hashrate, I still believe the vast majority of hashrate is owned by other small companies and individuals, with many of them who have no clue about what they are doing and will be eaten by these large guys pretty soon.

With that being said, no matter how much money these guys can afford to pay, the already so huge network size and the production capacity will keep the difficulty from going up in parabolic, once we hit 200EH which we got very close to back in May, CS will need another 500MW and nearly 200k miners just to bring the difficulty up a modest 9%, which is why I don't worry much about newly added gears more so than what is actually happening with nearly 80EH that went offline from May to July.

[kano]

Looks like a diff drop is getting more likely ...
Currently at -4.85% after 417 blocks ...
Quite possible evidence to my 'speculation' that there's a lot of old hardware that's borderline profitable, that's being switched off due to the price drop.

If you were China government you could do this.  Real gear and chip restriction. To the entire world.

Fortunately China does not have control of the Chip production.

[NotFuzzyWarm]

If you were China government you could do this.  Real gear and chip restriction. To the entire world.

Fortunately China does not have control of the Chip production.

Ja. There are no Foundries in China that are capable of producing 10nm chips much less 7 or 5nm ones. The major makers of mining gear and owners of their specific chip designs are Chinese-owned but,
 a) all the actual producers (Foundries) of cutting-edge much less bleeding-edge chips are located in Taiwan, Korea, and the US.
 b) if pressured by the Chinese government the miner makers will just move assembly operations to a different country.
 c) worse comes to worst, they will startup new miner manufacturing companies in a different country.

[philipma1957]

If you were China government you could do this.  Real gear and chip restriction. To the entire world.

Fortunately China does not have control of the Chip production.

Ja. There are no Foundries in China that are capable of producing 10nm chips much less 7 or 5nm ones. The major makers of mining gear and owners of their specific chip designs are Chinese-owned but,
 a) all the actual producers (Foundries) of cutting-edge much less bleeding-edge chips are located in Taiwan, Korea, and the US.
 b) if pressured by the Chinese government the miner makers will just move assembly operations to a different country.
 c) worse comes to worst, they will startup new miner manufacturing companies in a different country.

I suspect c) will happen

I can see spondollies coming back into major production.

[HagssFIN]

@Phil
Is Spondoolies really coming back to btc miner market,
or are you only speculating?
I saw their Dash miner year or few ago,
but I have not seen a new btc miner after SP20.