Topic: 2022 Diff thread. - page 16. (Read 9656 times)

One uneventful period


Less than 1% change in difficulty, as I see the price was also in the 30k-32k area, so minimal deviation thee too, I hope we've reached a point where at least when it comes to difficulty the double digits changes are almost gone just as higher single digits ones. I would have expected and obviously bet on a 1-2% up instead of down but I'm definitely happy with the outcome.

One thing I'm trying to think it's true is that no matter how much those big guys have ordered in advance from Bitmain and Canaan the May-June deliveries are going to make much difference, probably higher rates are putting others out of mining as well as gears breaking down are negating the effects of everything new delivered. I just hope I'm not proven wrong in a week and we see a spike of some 10% up cause it will really make me believe even the hashrate game is rigged.

There was some talk about NY baning again non-renewable mining but probably is just another non-event without any effect either.



 

 I will start with this post.



if we go negative it will be 2 straight diff drops.

if you study the diff in 2018 from Jan 2018 to Oct 2018 this never happened.

I am hoping it means the 2018 to 2022 coupling is not that solid a thing.

I will continue on this theme.

I do not want to mark this account with a ton of merits from  philipma1957 so I will give it a merit from my main account and hope to earn merits as I go.

Yeah this is me and I am going fully formal and will use the business account  a1 hashrate LLC2022

I am  a philipma1957 alt. I will be using this newly formed account.

I finally formalized with my partner buysolar.

I have this new ebay account

https://www.ebay.com/usr/a1hashratellc


I will post with this account most of the time from now on.


I will verify with philipma1957 in 1 minute.

well we are about 440 blocks left and about -1%

if we do a negative -1% if would be a very good sign of sideways vs bear for miners.

We actually could be in  a zone of crab for miners and bear for traders. We have been flirting with this and its new Go back to dec 2020 we were 28k and miners were making money

its now april 2022 we are at 29k and miners are making money.

and inbetween this entire time miners are making money.

traders not so much.

Well past the halfway point n the current epoch, and it's about as close to zero diff changer as you can get:


BTC price is still below $30K at: $29,869.

Start building now and deploying gears after 2024 Q2 would be a good idea, at least, a lot better than those who are building now and expecting gears to arrive mid to late 2023, right before the halving, I think many people do not understand how brutal the halving can be, it's probably because of the myth that many people spread which says " difficulty will always follow profitability".

While the truth is, BTC went down 60%, difficulty went up like 36%, so we are now mining 36% less bitcoin that is worth 60% less in fiat value, the same thing will happen after the halving, there is no such thing as half the miners shutting down, people need to understand this or else, they are in for troubles.

We know the next halving is going to be somewhere in April 2024 or maybe earlier, which means, by the end of this year, it will be extremely risky to buy a mining gear, it's either you do it now, or a few months after the halving when gear prices would adjust to the new profit scale.

meanwhile, we are back to positive on the difficulty estimation

in 2018 diff went up Jan to Oct while price unwound.  We finally became saturated in October and diff lowered and flattened for 6 or 7 months.

we appear to be copying that pattern.

the wild card is mr v and his eth machine I guess we will see what he does this summer.

One good thing about this year we has a few good drops over 4% I think in early 2018 we never had any good drops.

I went back and looked at 2018 and every diff jump from Jan 1 to july 21 was a plus then 1 drop then all plusses til oct 10 or so.

basically like 19 or 20 to the plus side. all with dropping prices.

while 2022 is 10 up with 3 down during a market dropping prices.  So maybe we are not a real mirror of 2018.

It would be nice to get a good rally from now til fall simply surprise all people.

all my stats came from here.


https://www.coinwarz.com/mining/bitcoin/difficulty-chart

...which will be more than compensated by the weed-like growth of U.S. corporate miners like Whinstone, which is building a gigawatt facility in northern Texas. I don't know how many more facilities can be built there, since ERCOT demand is ~50GW on average, and 1-2GW already goes to mining.

These corp-o-mines care very much about the long term, which is why they HODL most of their coins. To their investors, mining the coin comes at a 20-30% discount compared to buying the coin. Once it's built, it will never be shut down even if the megacorp goes bankrupt.

The bigger the farm, the longer it takes to build, so I expect network diff to keep growing at +4%/month unless BTC goes to $15k. Even then, it will take months for diff growth to slow down.

Not for long, I guess, although 4% is a serious drop and I wonder what caused it, I'm not that convinced profitability rates would be the culprit, at least not in the middle of the month and not so long after we've been hovering around these prices.

While not making it for 8 exa, at least 1 Exa was lost by bitdigital:
from a previous press release:

Probably compass and their mess has also a say in this drop although it did happen earlier but maybe the miners got disconnected later than the official statements.

On the other side:
2025?? Somebody is indeed going for the long run!

Looks like a definite decrease in difficulty is in the cards for tomorrow:


This suggests that the current difficulty may be the new All Time High difficulty for a bit. I wonder how long that will be true?

BTC price: $30,082.

went from -60 blocks to - 54

that pace means maybe we do -2%

Meanwhile, finally a nice drop:

I included cooling in the power usage of the ASICs as well as the buildout cost. I assumed that the climate is Northern Texas, so I used 150 CFM/kW as a conservative estimate. Air conditioning or chillers are more expensive than just shutting down when the weather is hotter than 100° F. The downtime would only be 1-2% and electricity is very expensive during those hot days anyway.

well if he is lucky and can use evaporate coolers it does cut cooling cost a lot.

a bigassfan cooler can do a lot in the right space.


https://www.northerntool.com/shop/tools/product_200874633_200874633?


https://bigassfans.com/evaporative-coolers/cool-space-500/

You left out cooling. 1MW = 3410000 BTU/h need to move and cool a lot of air. Which adds to electric cost. And big air efficient handlers are not cheap.
Once again sticking it in northern states will help that. Putting it in the south may save you money for rent & other things but you will be running the chillers longer and harder.

-Dave

Excellent dig. It was interesting reading their SEC K-1 filings, particularly the higher-than-expected payroll expenses. Rolf Verslusius might have been right after all: the lone wolf miner who has no employees might have an advantage.

Let's pit them against a typical small miner in the U.S. I assumed all capital costs are amortized over 3 years.

• Warehouse rent + fixed costs: $3500 + $500 = $4k/month
• Equipment: 300x Bitmain Antminer S19 95T = 1MW load | $8000 x 300 = $2.4m --> $66.7k/month
• Electricity price: 6¢/kWh all-in x 720,000 kWh = $43.2k/month
• Equipment repairs: 15%/year net failure rate x $1.8m = $22.5k/month
• Buildout cost: $200/kW x 1MW = $200k --> $5.6k/month
• Farm hashrate: 300x 95 Th +5% custom FW = 30,000 Th

• All expenses: $142k/month
• Average difficulty in Q1 2022: 26.7T
• Average BTC earned in Q1 2022: 4.19BTC/month
• Average cost per BTC (pre-tax): $33,890
• Average cost per BTC (after-tax, 24%): $25,756

If the power cost is 4¢ instead of 6¢, the cost per BTC is 8-10% cheaper, so $31k pre-tax or $24k after tax. This would make the per-Bitcoin cost of a small miner comparable to MARA or RIOT.

I do think Rolf was onto something when he hypothesized that small miners have many advantages. This doesn't count the flexibility that a smaller operation has to sell/buy equipment, not to mention the ability to do GPU mining.

I do hope that you were saying that as an exaggerated point... Miner chips simply cannot match the revenue generated that Foundries make by producing cutting/bleeding edge chips for just about everything NOT related to mining.

As I've often said before, the only reason that TSMC, Global, UMC and Samsung Foundries produced advanced-node miner chips so cheaply was to use them to fine tune their low nm node chip production processes. Miner chips are dirt-simple and very forgiving of production faults making them a perfect process refinement testbed. Once those processes were nailed down and had yields good enough for their main customers to run full speed with their very much more complex and non-fault tolerant chip designs- which produce immense profits for the Foundries - the advanced equipment availability needed for mining chip production was slashed and chip prices for the remaining allocation raised. Foundries know their cash-cow markets very well and making mining chips ain't one of them.

I stumbled on this sheet while browsing another forum:
https://docs.google.com/spreadsheets/d/11hd-Q6KdkNETiRBs_JYwxOEAEfiEM4YoYad__grb5lE/edit?usp=sharing

Basically, this is an attempt to calculate the cost of mining 1 btc for Marathon and Riot (the infamous big players).

The numbers were surprising to me, and I've noticed some errors, like the hashrate was not 225 EH on average in Q1, and a little more bitcoins were produced than 900*90 since, I guess, when hashrate is constantly growing we have more blocks generated than expected 2016 per 2 weeks. This lowers final numbers by 10-15% but still way too high IMO. The SEC filings are useful, but I'm not familiar enough with the terms to be able to verify them.

Anyway, thought it would be useful to share it here.

The only problem is we won't see as much growth per cycle as we did in the past.

2011: $1 --> $30 ATH, $5-10 bear market
2013-2014: $100 --> $1000 ATH, $300-500 bear market
2017-2018: $20000 ATH, $3000-6000 bear market
2021: $65000 ATH, $30000 bear market so far

The second bull market (2013-2014) saw the price increase 100x compared to the 2011 bear market. The 2018 gold rush was a 50x increase over the previous bear price, while the 2021 gold rush was a 20x increase over the last bear market at best. I think that the new ATH will only be 5-10x higher than whatever the bear pricing this cycle is, which I predict to be $20-30k.

That suggests the next peak will be $100k-300k, which lines up with the HODL price model. Assuming it takes a year for the next bull market, we can expect $200k as the most likely scenario, with $400k as the upper boundary.

The only event which would cause price to break $400k in the next 1-2 years is institutional investors, like large pension funds or emerging tech indexes, to allocate 2%-5% to BTC. This would require some regulation from the SEC (yes, regulation is bad for the little guy) and economic instability to get worse.

Can manufacturers produce enough ASICs to keep up with the price as it soars 5-10x? No way; it should take 2 years minimum for hashrate to go up fivefold. By that point, half the chipmaking capacity will be devoted to mining instead of consumer electronics, putting a cap on difficulty growth. This means even people who bought Bitmain S19's for $10k a piece and have 7-cent power can make a killing, so long as they can keep the doors open during this bear market. Block halving? Not a problem until the next bear cycle.