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

[DaveF]

I was wondering how much gear is profitable because it's paid for itself. But it has hardware issues. Once you have to move it there is a cost involved in labor for taking ti down, transport, labor in resetting it up and then hosting and power.

BUT if they owned the facility in China, and had access to cheap power then it was worth it to keep it running.
Now it's worth more @ $0.40 a pound as scrap metal.

I am kind of in the same boat. I have 5 S9 units that all have issues of bad boards. Where they are, with the cost of power in their current state they are generating about $1.00 a day in profit at $32,000 per BTC

To get to them is going to cost me over $30 in gas and tolls. So I would have to spend 1/2 a day going there and back to get them, clean them, swap boards, and then spend another 1/2 a day and $30 in gas and tolls to put them back.

So I lost hours of my time and over $60 to make $1 a day. So, as boards die and things go wrong I just leave them there.

Multiply that to a massive farm and even though yes all those old S9 units are making money, by the time you pack them and ship them and upback and set them up again you just added cost to something that had 0 cost to you at the moment. And you might be better leaving the space and power unused while waiting for new equipment.

Because yes you might be giving up $1 a day BUT if the cost to move it is going to be $50 and you have new equipment that will need the space and power coming in 90 days then you don't do it. Yes you are giving up $40 in theory but there is scrap value as I said and you don't know how much equipment will survive the trip and then you have to pay someone to take it down in a month or 2 which adds to labor cost.

-Dave

[Philipma1957cellphone]

I agree with the logic. I have a few s9 units left maybe 10. When I drive from howell to Clifton  I usually pull one or two of them bring g them home clean them and sell them a board at a time on ebay.

I am thinking of a lot of what i need to do in the future.  I have power to spare. I struggle to get new gear.

Even s17 would do. I am not sure if I dump all s9 and buy 2 s17. Mine them till they drop.

I can imagine that a lot of small Chinese farms are like what do i do.

Btw the diff is shifting upwards we are closer to -6 % vs -10 %

Have to wait and see.

[kano]

Meanwhile over the last 12 hours the history calculation has dropped to a -5.7% diff change.
Of course that means very little, since we still have 427 blocks to go, but alas interesting anyway

[PrimeNumber7]

I think a lot of the larger (or even fairly modest) farms have employees that take care of routine maintenance to keep utilization of equipment fairly high. It is also not typical to have to pay tolls to travel through most of the country. Obviously gas is an issue if you are not local to the location of your mine. Most people are not going to save money on rent if machines in need of repair/maintenance sit unused, so broken miners being in a datacenter results in the miners loosing money.

[DaveF]

I think a lot of the larger (or even fairly modest) farms have employees that take care of routine maintenance to keep utilization of equipment fairly high. It is also not typical to have to pay tolls to travel through most of the country. Obviously gas is an issue if you are not local to the location of your mine. Most people are not going to save money on rent if machines in need of repair/maintenance sit unused, so broken miners being in a datacenter results in the miners loosing money.

I was using myself as a small example. In the mega sized farm world there are different expenses. And with many relocating out of certain areas in China it's different too.
Even if transport is free and labor is free etc, how much does it cost in import taxes and fees when you cross a border i they are going to a different county? I And as Kano pointed out it does look at least at a quick glance on the surface that more hashrate is slowly coming back.

The truth is for now we just don't know, it's all speculation.

Going to be an interesting few adjustment cycles.

-Dave

[Philipma1957cellphone]

I think a lot of the larger (or even fairly modest) farms have employees that take care of routine maintenance to keep utilization of equipment fairly high. It is also not typical to have to pay tolls to travel through most of the country. Obviously gas is an issue if you are not local to the location of your mine. Most people are not going to save money on rent if machines in need of repair/maintenance sit unused, so broken miners being in a datacenter results in the miners loosing money.

I was using myself as a small example. In the mega sized farm world there are different expenses. And with many relocating out of certain areas in China it's different too.
Even if transport is free and labor is free etc, how much does it cost in import taxes and fees when you cross a border i they are going to a different county? I And as Kano pointed out it does look at least at a quick glance on the surface that more hashrate is slowly coming back.

The truth is for now we just don't know, it's all speculation.

Going to be an interesting few adjustment cycles.

-Dave

My hope was this does -5% and we are getting there.

My guess for next one is -2.to  +2

[PrimeNumber7]

I think a lot of the larger (or even fairly modest) farms have employees that take care of routine maintenance to keep utilization of equipment fairly high. It is also not typical to have to pay tolls to travel through most of the country. Obviously gas is an issue if you are not local to the location of your mine. Most people are not going to save money on rent if machines in need of repair/maintenance sit unused, so broken miners being in a datacenter results in the miners loosing money.

I was using myself as a small example. In the mega sized farm world there are different expenses. And with many relocating out of certain areas in China it's different too.
Even if transport is free and labor is free etc, how much does it cost in import taxes and fees when you cross a border i they are going to a different county?

As long as the costs make the miners a net positive to operate based on estimates of difficulty and price, miners should move the miners (or in your case, fix them). If a miner cannot afford the costs associated with moving (or repairing) the miners, or if doing so would be infeasible due to time or other constraints, the owners should sell the miners at their estimated value less the costs associated with moving/repairing them.

Miners have a limited useful life, so any downtime is money being thrown away. 

And as Kano pointed out it does look at least at a quick glance on the surface that more hashrate is slowly coming back.

Yes, it does appear we are coming off the local lows in terms of hashrate, but who knows how much of the local trough was due to variance. We had a local low in late June, that was preceded by a steep drop, but the hashrate bounced back a little bit and we have been bouncing around the same approximate hashrate since. I saw if we continue to hover around the same hashrate for another two weeks, the local low was driven in part by variance and we aren’t seeing large farms being brought back online.

In any case, we are still well below the ATH of hashrate. The price has gone up much more than difficulty did over the past year, so I would not blame the price decline on the difficulty reduction.

[kano]

... and today there's been 7 blocks from a miner who is mining to a new address and not putting anything in the coinbase to identify them.
One block 7 hours ago, then six in the last 3 hours.

So someone is definitely trying to increase the hash rate ...

See the current block 691111 coinbase.

...
We had a local low in late June, that was preceded by a steep drop, but the hashrate bounced back a little bit
...

That 23.6T, 20.6T, 25.0T in May (-12.61% +21.53%) was due to a known cause, again in China.
- it panned out to be about +3% +3% ignoring the outage that caused it - not that different to the previous 3 diff changes: +1.95% +5.82% +1.92%

The diff has simply kept dropping since the 25.0T, (25.0T 21.0T 19.9T 14.3T = -15.97% -5.30% -27.95%) but I suspect this one might be the last notable drop for a while to somewhere around 13.5T in a bit over 2.5 days.

[DaveF]

... but I suspect this one might be the last notable drop for a while to somewhere around 13.5T in a bit over 2.5 days.

Down to about 1.5 days. Still looking to be about 13.5T give or take a bit as you said.
My unscientific gut feeling is that it's going go down a bit more at the next adjustment and then bounce +/- 5% for a bit.
Bitmain (and others) are known for mining with equipment and then shipping them out to customers. And they have a bunch of promised Aug / Sept deliveries.
So there might be some stuff in transit.

Also, more and more at least here in the US, i am seeing smaller date centers put in no mining clauses. Which brings up a rant for another day.
But, they are not making you pull out existing equipment, just not letting you bring in more.


-Dave

[stompix]

Someone is doing a bit of rebranding:
https://www.bitrawr.com/difficulty-estimator

Current Pace:   94.4584%  (1821 / 1927.83 expected, 106.83 behind)
Next Difficulty Change:   between -5.4971% and -5.4485%

So it's going to be -5%, I don't see some huge farm coming suddenly online during the weekend.
And seems like more and more gear is available for next-day delivery from legit resellers, not shady deals so probably the next epoch will be the first with growth, but I doubt it will be serious.

Also, more and more at least here in the US, i am seeing smaller date centers put in no mining clauses.

Lol, you mean like not accepting miners in their data centers?
Damn, we just joined the porn branch with hosting companies not allowing adult websites!!!!

[DaveF]

...
Lol, you mean like not accepting miners in their data centers?
Damn, we just joined the porn branch with hosting companies not allowing adult websites!!!!

A bit OT but yes. There are many places that now have minimum power pulls per rack. Usually 4.8kVa (20A @ 240V)
That is the least amount of power you can purchase per equipment rack. Many times people were no where near that amount when they just had limited switch gear and such in there.
So miners were put in. Hell, if you make me pay for it I am going to use it.
The plan backfired so instead of just going back to billing for use, they put in a no miners are allowed clause. You can keep what you have but you can't bring in more.

Guess were are going to see many 4U rack servers coming in filled with GPUs. At least they look like everything else ;-)

-Dave

[NotFuzzyWarm]

There are many places that now have minimum power pulls per rack. Usually 4.8kVa (20A @ 240V)

Don't you mean maximum power per-rack? As a max it would make more sense because as you said, 4.8kva in a full rack is a lot for just powering servers and switches. But, a 42U high rack could easily pull over 20kva if filled with miner hardware.

My guess is that the data centers were forced to not only look at distributing that kind of dense power but also how to get rid of the heat generated. Most were designed and built to be data centers -- not mining farms.

edit: Got a PM and yes it is a minimum power per-rack. In short, it seems many clients op-sec requirements are demanding that only the clients and providers hardware be in a rack with no other (different) clients hardware allowed. That led to unused rack space which folks were filling with miners.

Moving back to on-topic,

Quote
Current Pace:   94.4584%  (1821 / 1927.83 expected, 106.83 behind)
Next Difficulty Change:   between -5.4971% and -5.4485%

So it's going to be -5%, I don't see some huge farm coming suddenly online during the weekend.

I'm hoping for -5.69% as that is my guess in the .01BTC "who can guess the closest" giveaway Kano is running in his Discord channel for folks mining at his pool

[Philipma1957cellphone]

There are many places that now have minimum power pulls per rack. Usually 4.8kVa (20A @ 240V)

Don't you mean maximum power per-rack? As a max it would make more sense because as you said, 4.8kva in a full rack is a lot for just powering servers and switches. But, a 42U high rack could easily pull over 20kva if filled with miner hardware.

My guess is that the data centers were forced to not only look at distributing that kind of dense power but also how to get rid of the heat generated. Most were designed and built to be data centers -- not mining farms.

Moving back to on-topic,

Quote
Current Pace:   94.4584%  (1821 / 1927.83 expected, 106.83 behind)
Next Difficulty Change:   between -5.4971% and -5.4485%

So it's going to be -5%, I don't see some huge farm coming suddenly online during the weekend.

I'm hoping for -5.69% as that is my guess in the .01BTC "who can guess the closest" giveaway Kano is running in his Discord channel for folks mining at his pool

My 1.5 ph is off for a day due to maintenance.  Maybe it fixes the diff for you

[kano]

... and today there's been 7 blocks from a miner who is mining to a new address and not putting anything in the coinbase to identify them.
One block 7 hours ago, then six in the last 3 hours.

So someone is definitely trying to increase the hash rate ...

See the current block 691111 coinbase.

...

Looks like that was fupool using that 691,111 block bc1 address (bc1q7wed...) and not identifying it as themselves again.

The last address they did that with (bc1qf274...) for about 330 blocks, stopped at block 691,063 also didn't have their name in the coinbase, but they transferred block rewards to the usual fupool mining address, so it obviously was them.
That address stopped appearing just before this new one appeared (and has now been in 35 blocks)

Meanwhile, the history calc on the diff is now -4.58% with only 67 blocks to go, thus very likely that hash rate is now increasing again.

[philipma1957]

... and today there's been 7 blocks from a miner who is mining to a new address and not putting anything in the coinbase to identify them.
One block 7 hours ago, then six in the last 3 hours.

So someone is definitely trying to increase the hash rate ...

See the current block 691111 coinbase.

...

Looks like that was fupool using that 691,111 block bc1 address (bc1q7wed...) and not identifying it as themselves again.

The last address they did that with (bc1qf274...) for about 330 blocks, stopped at block 691,063 also didn't have their name in the coinbase, but they transferred block rewards to the usual fupool mining address, so it obviously was them.
That address stopped appearing just before this new one appeared (and has now been in 35 blocks)

Meanwhile, the history calc on the diff is now -4.58% with only 67 blocks to go, thus very likely that hash rate is now increasing again.

-4.44% is better than I hoped for.

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

Latest Block:   691430  (9 minutes ago)

Current Pace:   95.5102%  (1959 / 2051.09 expected, 92.09 behind)

Previous Difficulty:   19932791027262.74                            
Current Difficulty:   14363025673659.96                            
Next Difficulty:   between 13724229954263 and 13724717918669
Next Difficulty Change:   between -4.4475% and -4.4441%
Previous Retarget:   July 3, 2021 at 2:25 AM  (-27.9427%)
Next Retarget (earliest):   Today at 6:11 PM  (in 0d 9h 56m 2s)
Next Retarget (latest):   Today at 6:12 PM  (in 0d 9h 56m 47s)
Projected Epoch Length:   between 14d 15h 46m 56s and 14d 15h 47m 41s
...

we are pulling a lot of blocks at the moment.
If next jump on July 31 or Aug 1 jumps +2 to +5 will will be back on track.

But it would mean six week of mining with diff at or near 14 on average

July 3 to July 17 at 14.36
July 17 to July 31 at 13.72
Aug 1 to  Aug 15 at 14.00?  Next guess on my part.

edit at 730 pm EST
1 block to go

and 732 pm done

dropped to  13.67

enjoy the extra coin for next 2 weeks or maybe 13 days.

I smell +2 to +4%

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

Latest Block:   691489  (3 minutes ago)
Current Pace:   522.2318%  (2 / 0.38 expected, 1.62 ahead)


Previous Difficulty:   14363025673659.96                           
Current Difficulty:   13672594272814.14


                           
Next Difficulty:   between 13694756121608 and 54690377091256
Next Difficulty Change:   between +0.1621% and +300.0000%
Previous Retarget:   Today at 7:32 PM  (-4.8070%)
Next Retarget (earliest):   Tuesday at 11:52 AM  (in 2d 16h 16m 31s)
Next Retarget (latest):   July 31, 2021 at 6:59 PM  (in 13d 23h 23m 50s)
Projected Epoch Length:   between 2d 16h 20m 21s and 13d 23h 27m 40s

The drop was about -4.807%


So if we go up 2% we go to 13.946

and if we go up 4% we go to 14.219

While I guess that as a slot. I could be wrong.

[kano]

DIFF CHANGE: hi=691488 delta=-4.81% new=13672594272814.1 prev=14363025673660.0

Heh didn't match that site's "Between" values

Edit: time was 2021-07-17 23:32:17

[philipma1957]

DIFF CHANGE: hi=691488 delta=-4.81% new=13672594272814.1 prev=14363025673660.0

Heh didn't match that site's "Between" values

their between misses almost always  


I just look at the current pace number which is real time.

I expect at least 40 above the pace when we are at the 1900 block number which should be 12,13 days from now.

[mikeywith]

I smell +2 to +4%

Gear prices in China went up about 15-20% in the last couple of days, and since the price has been stable or more to the downside than up, and the difficulty hasn't changed in the past few days and thus profitability wasn't a factor this means

1- The Chinese are relocating to different regions in China.

Or / And

2- The demand from outside of China spiked.

The reason is somehow irrelevant, the outcome however is rather clear, at 32k Bitcoin we will unlikely witness any difficulty drop below the current levels, so if I had to bet, I would bet on the difficulty to go up (assuming the price doesn't collapse), and then remains the question of how much up? if it's the Chinese relocating within China then the rise will be fast, if it's overseas, it will be as small as you think.

My long-term projection remains the same, we will go up slowly, we won't see the ATH difficulty level any time soon, probably not until 2022.

[philipma1957]

I smell +2 to +4%

Gear prices in China went up about 15-20% in the last couple of days, and since the price has been stable or more to the downside than up, and the difficulty hasn't changed in the past few days and thus profitability wasn't a factor this means

1- The Chinese are relocating to different regions in China.

Or / And

2- The demand from outside of China spiked.

The reason is somehow irrelevant, the outcome however is rather clear, at 32k Bitcoin we will unlikely witness any difficulty drop below the current levels, so if I had to bet, I would bet on the difficulty to go up (assuming the price doesn't collapse), and then remains the question of how much up? if it's the Chinese relocating within China then the rise will be fast, if it's overseas, it will be as small as you think.

My long-term projection remains the same, we will go up slowly, we won't see the ATH difficulty level any time soon, probably not until 2022.

I would love to see a slow climb to 22.22th by jan 27 2022

[kano]

Meanwhile while there's no chance of it actually being accurate - since it's only 89 blocks so far - the diff calc on those 89 blocks is +8.71%

Will be interesting to see what it is at around 1008 blocks/half way into this diff change ...