Topic: Well the 2018 btc diff jumps are over so what will 2019 bring us? (Read 9460 times)

Even thoguh I have 5 hours to midnight here in NJ.. I will lock it before I start drinking for new years.  I don't want to say something stupid.

 

Just about 24 hours are left for next difficultity adjustment, we are at a base of 105% ( 90 blocks ahead) and it has been this way for quite sometime, so it is highly unlikely that we would see any change in the expected  +5%  diff, which is not enough to bring us above all time high of 13.7t.

I honestly expected things to be worse for year end, where did all Dec batch of S17/T17 go?? What happened to the 20EH which whatsminer was suppose to ship according to one of the members here?  Have all the S9s been replaced and this is why the jumps are small?


Phill, 2019 is over , it has been a pleasure speculating and analyzing 2019 difficulty with you and the rest, do you plan on locking this topic and starting a new one for 2020 or we would just use this one for the rest of our mining journey?

We have about 209 blocks to go and are at about 5.2% If we stay at 5.2% or less  we will not have climbed back to the all time high diff of 13.7t
we will be at 13.6 t

Yeah it has backed off from 13-15%. 4 or 5% is much better number.

Mini Christmas bonus with everything still being extremely early we are now looking at a 3.5-5.3% jump up. That's not bad at all.
I'm amazed that we've floated these low numbers for so long. I get that price isn't rallying upward but I just really assumed there would be a lot more pressure for future profits by mining as much as possible for the next few months.
I've pretty much given up on looking for what the best move is, I've gone back to a plan that ignored the halving. Sell/trade some coin hold a portion, and try to get all of my available hashrate online while I'm free/offset on power - been a shit winter for that. This way the Diff doesn't have much bearing on my little farm until summer, and then I can just wait until Fall if need be.

Well a  big jump finally is happening


this is a plus 13%    my guess is a major deployment of m20's or s17s if   get them up and running in 2019   this income is small for 2019 year, but will allow expensing of the gear in 2019 for taxes.

If they use 3 year depreciation method in usa law  they could deduct 33% of build cost  against the income of only 2 weeks.  That equals a loss for the 2019 tax year.

To be a true 13%  it is about 10-15 eh of gear.  if all s17's doing 50th  20 are 1ph 2000 are 1 eh  20000 are 10eh

This is all speculation of course  20,000 x 2.4 kwatts  = 48 megawatts

2020  is going to be very interesting year for us.  This move is right out of mikeywith's post a bit earlier.

we went from 12.87 to 12.96  a jump of + 0.699%

price is about    7132 usd

last jump price  7400 usd

basically  a small decline in earnings maybe 4% most of  earnings decline was in price not  diff.

I am hopeful next jump is under 2%  so far it is higher.

We have 23 blocks to the next adjustment.
We are at +0.5%
https://diff.cryptothis.com/
Coin price is about
7100

100th earns about $13.50 usd a day before power cost.

if you have 100 watt  per th or 240kwatts a day  at 5 cents a kwatt   you clear $1.50 a day this would be 7 s9's
if you have 90 watt              or 216   "      "    "    "      "     "   "   "    you clear $2.70 a day
if you have 80 watt    "     "  or  192   "     "     "  "  "  "  "  "     "   "   you clear  $3.90  a day
if you have 70 watt   "     "   or  168   "    "     "  "  "   "  "  "    "   "    you  clear $5.10 a day
if you have 60 watt  "      "   or  144   "    "    "   "  "  "   "  "    "   "    you  clear $6.30 a day
if you have 50 watt   "    "    or  120   "   "    "    "  "  "   "  "    "   "    you clear  $7.50 a day
if you have 40 watt   "   "     or   96   "    "    "    "  "  "   "  "    "   "    you clear  $8.70 a day  this would be 2 s17's

Yeah they have a lot of ways to plan how and when to switch over from s9's to s17s or t17s or a1166 or m30s.

But doing a very slow switch out  right now does make some sense.

A fast switch over is really stupid  it is just about the only way they can lose. (other then fires earthquakes floods and or a government shuts them down)

So they would be sitting and waiting or very slow switch over to the better gear.

We just slipped into a negative diff of around 0.05%

Yeah, if they can't get more power and the price for new equipment is low enough, it makes sense.

But if you look at current prices, a T17e will replace 2 S9s and use about the same power. It will only give you an extra 22TH for about $1350 shipped to USA, maybe $1100 if coming from Malaysia.

If you did a one S17 pro running on low power mode (maybe 40TH at 1440W) to replace each S9, then you get an extra 26TH for $1900 shipped to USA, or $1600 if coming from Malaysia.

Both options are expensive and will take a very long time to ROI.

A 3 cent farm has the luxury of being able to wait. If price and profitability goes up, they are still making good money. The higher the profitability goes, the less advantage the higher efficiency miner has over their S9s. If price and profitability goes down, they can buy up the next level up of efficient gear for a deep discount.

Price needs to go below $3400 for the 3 cent farm to go unprofitable with S9s at current diff, so they have plenty of buffer at the moment.

There is a case for a 40megawatt  farm of all s-9's  at 3 cents to just add s17's slowly.

Even if they are at maxxed power use.

As the mining industry expands with these huge farms that have super cheap power, the average power cost for the network drops. The price/diff ratio will drop from this as well as the increase in miner efficiency. It may eventually reach an equilibrium where all the super cheap power capacity is used up.

The saving grace for the small guy now is that it doesn't make sense for these huge farms with super cheap power to pay the premium for the highest efficiency gear.

Break-even for S9s now solidly below 6 cents.

we are close to 0%


price is ugly  6498.

Except for the fact that mining is noway close to the average business model you described here, mining is heavily based on speculations, many people mine in the hope of price jumps, they try to secure cheap gears in bad market conditions, some go all the way to mining at break-even or even a slight loss.

I am not saying that those people are making or will make a good profit or even see a ROI , that is not the case, a dozen of miners including large mining farms ended up broken, there is no guarantee that anyone is going to win at the end, but the fact that what some people would give for free, worth a lot to others, proof is, Antminer T9 or even V9 still hold some value to some people.

if everyone was to follow a business plan as they do in other business models, i am sure half the hashrate wouldn't have come to existence to begin with, but miners are not just professional business men, there are newbies, stupid people, greedy people and speculative people who would settle down for tiny profit in an attempt to be a head of the curve when the "moon" comes, all of these things make it kind of hard for anyone to say "if price does this-miners will do that".

I hope you can follow the logic of my theory, which of course could end up being a complete B.S  

No, I don't think it would have been possible. If we look at the single time the hash rate went down for a longer period of time, not just one adjustment, it seems that 50exa was the max for 6k and 3k could sustain 30exa.  If we would be on the same efficiency 7k wouldn't be able to hold anything more than 70exa.


I find this way of estimating revenue too simplistic, you don't just plug miners on a field and leave them there.

You pay extra for cooling, you pay for maintenance, you pay rent (even if you don't pay you to have to subtract that as you could simply rent the space to somebody and get no headache money) you need staff, some sort of security, you need to ROI and you need to make a profit that is worth all the troubles otherwise you would simply go and flip hamburgers at McDonald's.  (trivia, right now the entry wage at McDonald's here is worth the profit of about 35-50 S9, which is above any kind of house mining operation)

It just like a store, you buy stuff with 2 cents and sell it for 5c but the profit won't be 3c per piece.

Bottom line, with a stable price around 7K after the halving I don't see something beyond 100-130 exa possible (assuming no new gear). But I know I suck at predicting things  

Price can not be predicted with any reliability at all simply because it can change minute to minute.

On the other hand difficulty can be predicted with a fair degree of accuracy. Due to it changing 1 time every 2 weeks

yeah sometimes 10 days and some times 16 days.  Mostly 13,14,15 days.

So if you mine and understand difficulty you can plan  2,4,6 weeks in advance which means $$$ in your pocket.

Philipma1957 (that is me)  has done a lot of work at difficulty prediction simply because it is easier to do.

I think we may be in the rare moments of time when price follows difficulty not the other way around. I see this more often over the next decade in the 2020 jump the 2024 jump and the 2028 jump.

Last December price dropped and dropped and dropped  along with difficulty.  I agree for the most part diff chases price.
I am certain of it up the way up.
But on the way down it sometimes seems that diff leads price.
If we are there price will tumble to under 6000 maybe under 5000.

We bottomed out last december in the 3100-3200 range.

Except the fact that there is no way for us to know if the doubling happened only because gears efficiency has doubled, there is another huge factor that many seem to be missing which is "settling down for less profit".

to make the quoted part above a little bit clearer, let's just pretend that the most efficient gear the world has ever known is still the Antminer S9 , do you think that difficulty wouldn't have increased or even doubled?

as we speak, the guys with 6 cents per kWh still make profit on S9 , let alone those with 2-3 cents and even free power, which proves the fact that regardless of efficiency miners still have a lot of room to grow, profitability will decrease, but that won't stop them, a guy with 3 cents power can now buy antminer S9 with psu for 90$ , makes 90 cents a day or  27$ a month , that's about 3 months to ROI , someone with free power makes 50$ from S9, that's less than 2 months ROI ,what stops those guys from adding more S9s to their farms?

The halving will only cause so much troubles if everyone is mining at close to break-even point, the further they are from that point the less drop in hashrate/diff , and all of this is based on the assumption that price will actually be a lot higher than it's now for this to work this way, so to sum up again, it really depends on the price ( I don't even know why are we trying to speculate something that is nearly 100% price based  )


6600$ as I type, 700$ drop, nothing major, just an ordinary move as far as bitcoin is concerned 

Well peak has in OCT  2018 was 57eh but closing number was 51eh to give us a peak diff was 6.9

we are around 93eh
diff is about  12.9

so 6.9 to 12.9 is an 86% jump

we did drop well below the 6.9 when prices bottomed at 3100  we dropped to 5.1 diff

I see good short term numbers here.

a -+1% in two days

followed by  a low number  two weeks later.

so dec 19th = 1%
and Jan 2    = 1%

1% jumps are 26%  a year

with no ½ ing   my guess for dec 2020 is a diff under 17
with the ½ ing my guess for dec 2020 is a diff of under 15

I don't do long term numbers for diff  due to price jumps which can blow away any prediction.

those diffs are based on price staying under 10k for the 2020 year.

While I agree with the math, I disagree a bit with the model. Yeah, the price was reaching 14k but it did so at the end of June while at the beginning of the month we were at 8k. Miners wouldn't have had time to adapt to the price rise and plug more machines and even if they did had spare equipment this would have been most likely 1/2 or 1/3 efficient compared to what's right not being sold.

Probably we should look as a starting point at what happened in last October, base the model on over 80% of the hashrate being s9 or equivalent models, and a price collapse from 6k to 3k. Right now we're at nearly the same price, double the efficiency > double the hash rate. So, if a halving would occur tomorrow probably we would see a drop to 70exa. Last year Oct-Nov average was 50exa and it dropped to 35, now we're at 100 so that's where the 70 is coming.

One thing  , your post was made while the price was above 7300, mine...oh crap!

I have three s9 that do 40 th they cost me 145 a month to host all three.

They earn about 40 x 0.000019 = 0.00076 btc a day

30 x 0.000076 = 0.0228 btc a month that is 157 usd just in the black.

I may stop these on Jan 1 May not depends.