2023 Diff thread now opened.

[philipma1957]

Is there really a Malaysia factory? I have read articles that one is used to get around trump tax.

Yes they do have a factory in Malaysia, I have personally seen some Malaysian made Antminers mostly the 17 series, which to my surprise were a lot better quality than the ones made in China.

And ya those gears are subject to only 2-3% tax in the U.S if IIRC.

However, ever since the 17 series, I think the vast majority of their gears are produced in mainland china.

@joker_jouse, depending on what country it is, DDP shipping could be available from China/HK which would be cheaper than your regular shipping.

iirc he gets high tarriffs in his country.

[joker_josue]

@joker_jouse, depending on what country it is, DDP shipping could be available from China/HK which would be cheaper than your regular shipping.

Few do this for Portugal. At least you have the option of including these fees at the time of purchase, but it is rarely worth doing this as you pay even more.

Just to give you an idea, I pay: 23% of the amount paid for the item + 10~20€ processing fee + if the item has a value greater than 150€, I have to pay another fee that can vary between 5 ~15%. I know that around the world, there are much worse countries. But, in Europe, Portugal is one of the least "favorable" in this field.

Furthermore, customs services are a bit complex, so unless it is a company that is already used to shipping to Portugal, it will do DDP. Otherwise, they rarely DDP go to Portugal.

[BitcoinSoloMiner]

https://newhedge.io/terminal/bitcoin/difficulty-estimator


Latest Block:   817944  (7 minutes ago)

Current Pace:   103.4589%  (1465 / 1416.02 expected, 48.98 ahead)

Previous Difficulty:   62463471666732.73                            
Current Difficulty:   64678587803496.61                            
Next Difficulty:   between 66750216791256 and 66923265466373
Next Difficulty Change:   between +3.2030% and +3.4705%
Previous Retarget:   November 12, 2023 at 11:22 AM  (+3.5463%)
Next Retarget (earliest):   Sunday at 12:08 AM  (in 3d 16h 45m 47s)
Next Retarget (latest):   Sunday at 12:59 AM  (in 3d 17h 36m 7s)
Projected Epoch Length:   between 13d 12h 45m 59s and 13d 13h 36m 20s

4 days to go. solid gain +3.45%

Having worked with every bitmain miner except the s11.

If I was buying 50,000 new units I would insist on at least 5,000 being burn tested for two-four weeks.

I would want my people to pick them at random as I would want an idea of just how shitty they are.

So I have to think these are burn tested for quality control for big buyers.

Is there really a Malaysia factory? I have read articles that one is used to get around trump tax.

Maybe not a factory but definetly a distribution warehouse. In late 2021 we received a s19j pro from "seberang jaya industrial estate" which is in malaysia

[stompix]

Current Pace:   104.4559%  (1957 / 1873.52 expected, 83.48 ahead)

So this night we're going to have the six positive streak, looking at blocktime in the meddle of the period the second week was way faster so as I said before, we're going to start with a bit of extra momentum, probably another positive adjustment. And Bitmain is sold out, a wave of S21 bought by almost everyone from Mara to Cleanspark is going to come online in January so I'm flipping right now to wonder if the grow will ever stop till the halving.
~250+ exa this year, on path of exceeding a 100% growth, insane if we look at power requirements and cost of this, so insane that if we want those e good old days of late 2021 we need a price of 200k per coin!

Quite funny, with these growth you look at a statement from Mara saying they plan on powering up already installed 4.1 exahash of gear in Garden City, TX and you're saying (at least I do) meh, seen worse, 7 Exas in UAE, keep them coming, lol! I'm out in the snow playing! 

[philipma1957]

Current Pace:   104.4559%  (1957 / 1873.52 expected, 83.48 ahead)

So this night we're going to have the six positive streak, looking at blocktime in the meddle of the period the second week was way faster so as I said before, we're going to start with a bit of extra momentum, probably another positive adjustment. And Bitmain is sold out, a wave of S21 bought by almost everyone from Mara to Cleanspark is going to come online in January so I'm flipping right now to wonder if the grow will ever stop till the halving.
~250+ exa this year, on path of exceeding a 100% growth, insane if we look at power requirements and cost of this, so insane that if we want those e good old days of late 2021 we need a price of 200k per coin!

Quite funny, with these growth you look at a statement from Mara saying they plan on powering up already installed 4.1 exahash of gear in Garden City, TX and you're saying (at least I do) meh, seen worse, 7 Exas in UAE, keep them coming, lol! I'm out in the snow playing! 

or hodl hodl hodl we may go to that 200k.

[philipma1957]

newhedge.io

Latest Block:   818494  (8 minutes ago)
Current Pace:   105.0086%  (2015 / 1918.89 expected, 96.11 ahead)
Previous Difficulty:   62463471666732.73                            
Current Difficulty:   64678587803496.61                            
Next Difficulty:   between 67923661861338 and 67923662699840
Next Difficulty Change:   between +5.0172% and +5.0172%
Previous Retarget:   November 12, 2023 at 11:22 AM  (+3.5463%)
Next Retarget (earliest):   Today at 7:21 PM  (in 0d 0h 9m 31s)
Next Retarget (latest):   Today at 7:21 PM  (in 0d 0h 9m 31s)
Projected Epoch Length:   between 13d 7h 58m 26s and 13d 7h 58m 26s

+5%

5.07

67.9 t

five percent is over 23eh even if all s21s it is 115,000 units  math check needed as I am a bit tired.

[mikeywith]

five percent is over 23eh even if all s21s it is 115,000 units  math check needed as I am a bit tired.

Yup, more like 24eh actually, but ya, let's say 23eh, that is 24000000th / 200th (S21 hashrate), then your math checks out, if we are talking power consumption, those are over 400MW, not counting cooling and whatnot, I find it hard to believe that 400MW came online in 2 weeks out of nowhere,  the numbers don't seem to add up.


current diff status

Current Pace:   99.3947%  (412 / 414.51 expected, 2.51 behind)
Previous Difficulty:   64678587803496.61                           
Current Difficulty:   67957790298897.88                           
Next Difficulty:   between 67616333558533 and 67832087777919
Next Difficulty Change:   between -0.5025% and -0.1850%
Previous Retarget:   last Sunday at 2:11 AM  (+5.0700%)

transaction fees have been going down steadily and slowly, we are now at 38 sat/vbyte, I think we are headed back to <10 sat in just a couple of days.

[stompix]

Yup, more like 24eh actually, but ya, let's say 23eh, that is 24000000th / 200th (S21 hashrate), then your math checks out, if we are talking power consumption, those are over 400MW, not counting cooling and whatnot, I find it hard to believe that 400MW came online in 2 weeks out of nowhere,  the numbers don't seem to add up.

Hihi, just as we talk about saying millions of th instead of exas, we should start using nuclear reactor instead of 1000MW, we've reached some levels where numbers alone are not really telling all the picture behind this growth. It's more power than what Uganda or Jamaica are using in a year, and this is just growth and we assume the most efficient gear and no cooling for them

transaction fees have been going down steadily and slowly, we are now at 38 sat/vbyte, I think we are headed back to <10 sat in just a couple of days.

We're sitting on over 140 blocks of tx with over 10sat/b in the pool, doubt it can be cleared until Sunday night early Monday morning, but the decrease will also trigger some exchnges to consolidate their dust, last time Binance dumped 40 blocks worth of tx alone, so probably it will take more than just a few days.
Of course Btc going over 40k might also trigger some additional traffic, so I doubt really cheap fees are that close.

Pace is just under the normal but I blame the really slow start, we had a really unlucky day on the second one, with just 131 blocks in 24 hours, so I'm still thinking 2% is easy doable.

or hodl hodl hodl we may go to that 200k.

Gonna be a bumpy ride, even if I try shooting the bear in me 100 times, I still see this a little hard to achieve, too many are waiting to cash out some 2x 3x profit margins so probably 70k and 100k will be really hard psychological price levels.

[mikeywith]

Ya like for real, is it even possible for a minimum of 500MW worth of gears to go online in a few days? I just can't buy into this theory, there must be something else that we do not know, we will probably never get to understand what "was actually happening".

I think, however, it is almost certain that we can't sustain this for too long, whatever game there is, will have to come to an end.

[joker_josue]

Ya like for real, is it even possible for a minimum of 500MW worth of gears to go online in a few days? I just can't buy into this theory, there must be something else that we do not know, we will probably never get to understand what "was actually happening".

I think, however, it is almost certain that we can't sustain this for too long, whatever game there is, will have to come to an end.

But did these machines, equivalent to 500MW, come and go, suddenly on the grid?
Was this just a moment, or did it happen several times?

Another question, for those of you who are more attentive, did this increase appear shortly before, the brutal increase in fees? Or were the fees already high?

[philipma1957]

a little bit after so they fired up to chase the high fees.

they backed off a bit as fees dropped a lot.

[joker_josue]

a little bit after so they fired up to chase the high fees.

they backed off a bit as fees dropped a lot.

So I would say that this is some miner, perhaps with older machines, who only turns them on when there is an increase in fees. Taking the opportunity to make these machines profitable. When the fees drop, it switches off again. But, this is also a basic analysis. 

I asked if this increase appeared before the increase, thinking it was someone trying to make money by causing this increase in fees. And if so, it could be a way of trying to predict when fee increases would appear. Well... maybe I'm already rambling too much with the same thoughts. 



_{Look at our PMs.}

[philipma1957]

https://newhedge.io/terminal/bitcoin/difficulty-estimator



Latest Block:   819283  (8 minutes ago)
Current Pace:   98.6078%  (788 / 799.13 expected, 11.13 behind)
Previous Difficulty:   64678587803496.61                           
Current Difficulty:   67957790298897.88                           
Next Difficulty:   between 67047625308580 and 67382470049484
Next Difficulty Change:   between -1.3393% and -0.8466%
Previous Retarget:   last Saturday at 7:11 PM  (+5.0700%)
Next Retarget (earliest):   December 9, 2023 at 10:10 PM  (in 8d 13h 47m 46s)
Next Retarget (latest):   December 9, 2023 at 11:56 PM  (in 8d 15h 33m 22s)
Projected Epoch Length:   between 14d 2h 59m 1s and 14d 4h 44m 37s

[kano]

Ya like for real, is it even possible for a minimum of 500MW worth of gears to go online in a few days? I just can't buy into this theory, there must be something else that we do not know, we will probably never get to understand what "was actually happening".

I think, however, it is almost certain that we can't sustain this for too long, whatever game there is, will have to come to an end.

But did these machines, equivalent to 500MW, come and go, suddenly on the grid?
Was this just a moment, or did it happen several times?
...

There isn't 500MW of miners waiting for high fees.

Bitcoin blocks are random, and thus ALL calculations based on block finding have variance.
Using 2016 blocks to determine each diff change is not highly accurate - so daily and weekly swings can be very high, unrelated to hash rate.

Here think of it this way ... if you went out and asked 144 people who the next president would be, would you take that as knowing the answer?
Yeah it's not really an accurate valid comparison, but it explains the idea that the daily 144 blocks are really no accurate representation of the world hash rate.

[joker_josue]

Here think of it this way ... if you went out and asked 144 people who the next president would be, would you take that as knowing the answer?
Yeah it's not really an accurate valid comparison, but it explains the idea that the daily 144 blocks are really no accurate representation of the world hash rate.

Yes, it's true, they are not really a worldwide representation.
But that's how statistics are made. Take a sample and make a more comprehensive estimate.

Of course, the data is not 100% viable, it only allows for a generalized analysis. Although numbers can be presented in many ways, and lead us to conclude what we want.

[philipma1957]

yeah all blind random luck.  well we were at -16 blocks and we rallied from 38.5 to 39.5 and we are now -19 blocks.

If we wander around like this it means random theory is pretty good.

Or companies are fully on line as of last week. So plus or minus 1% is nothing.

If we now take off like mad and go to +5% it fits a pattern.

last jump we were near 0% and surged last week

so do we do this near 0% and surge due to higher prices.

[kano]

Here think of it this way ... if you went out and asked 144 people who the next president would be, would you take that as knowing the answer?
Yeah it's not really an accurate valid comparison, but it explains the idea that the daily 144 blocks are really no accurate representation of the world hash rate.

Yes, it's true, they are not really a worldwide representation.
But that's how statistics are made. Take a sample and make a more comprehensive estimate.

Of course, the data is not 100% viable, it only allows for a generalized analysis. Although numbers can be presented in many ways, and lead us to conclude what we want.

Alas, unfortunately, statistics is actually quite clear and exact.
The problem is 2 fold:
1) most people don't understand it and don't have a chance in hell calculating valid information
2) you get statistics thrown at you by the media all the time that is blatantly wrong
3) yeah one extra - the ranges shown on diff estimators are all bullshit

[philipma1957]

https://newhedge.io/terminal/bitcoin/difficulty-estimator

Latest Block:   819589  (2 minutes ago)
Current Pace:   99.2818%  (1094 / 1101.91 expected, 7.91 behind)
Previous Difficulty:   64678587803496.61                            
Current Difficulty:   67957790298897.88                            
Next Difficulty:   between 67495959887426 and 67592036924124
Next Difficulty Change:   between -0.6796% and -0.5382%
Previous Retarget:   November 25, 2023 at 7:11 PM  (+5.0700%)
Next Retarget (earliest):   Saturday at 9:07 PM  (in 6d 10h 16m 11s)
Next Retarget (latest):   Saturday at 9:37 PM  (in 6d 10h 46m 41s)
Projected Epoch Length:   between 14d 1h 55m 19s and 14d 2h 25m 50s

so last jump at 1000 blocks we were near 1% and finished at 5%

this jump we were near -1.5 % at 1000 blocks

we went right near 40k and we pick things up 🆙.

If we finish out at 42k and + 5% I would not be surprised.

It is not that gear comes on line that fast.

a lot is clocking ability.

an xp can do 110th to 160th with aftermarket controller the epic model

And whatsminer m50s can do 80th to 110th

economy of scale allows for a wide clocking swing

160/110 is  a 44% jump

110/80 is a 37% jump


no new gear just change clocks 🕰️

we know new gear does come online.

but clocks can really make a change in %

edit we got over 40 k and newhedge.io   says we are plus

"Latest Block:   819650  (13 minutes ago)
Current Pace:   100.1759%  (1155 / 1152.97 expected, 2.03 ahead)
Previous Difficulty:   64678587803496.61                            
Current Difficulty:   67957790298897.88                            
Next Difficulty:   between 68076189693234 and 68102659987159
Next Difficulty Change:   between +0.1742% and +0.2132%
Previous Retarget:   November 25, 2023 at 7:11 PM  (+5.0700%)
Next Retarget (earliest):   Saturday at 6:36 PM  (in 5d 23h 14m 52s)
Next Retarget (latest):   Saturday at 6:42 PM  (in 5d 23h 21m 20s)
Projected Epoch Length:   between 13d 23h 24m 35s and 13d 23h 31m 3s"

this is and argument for clocking and it was bumped a bit.  But who knows as it could be random enough


we are now 22 blocks ahead. which means a clear correlation to the price and likely due to the same amount of gear with lots of clocking play..

there is a good part to this. if we shoot to 80k or 100 k we likely would run of of clocking space. and growth would slow for a bit.

mean while enjoy the higher prices.

[mikeywith]

Using 2016 blocks to determine each diff change is not highly accurate

if you calculate the probability of a 5% diff increase without an actual 5% hashrate increase in 2016 blocks, or in other words, what are the chances 2016 blocks take 5% less time to mine than the expected 10 mins without any real hashrate change, I think using the exponential CDF to get a clue, the probability is close to zero, ya it gets better if you were to assume that it's possible that 50% (2.5% of that increase was caused by an actual increase in hash power) while the other 2.5% was caused by pure randomness, but still the probability of that 2.5% is also close to zero.

I agree, there is no definite way to measure the network hashrate, nothing you can take to court and present as clear cut evidence, but generally speaking, a 5% increase in difficulty is most likely the result of an almost 5% increase in hashrate, ya it could be 5.1% or 4.9%, nobody knows, but 5% in 2016 blocks is fairly accurate.

[philipma1957]

you can see trends that make sense.  

we were -7 blocks we are + 11 blocks

you can correlate it to the price going from 388xx to 422xx

but it is small not huge the small it is the more likely it is noise not a true trend.


144 blocks set to be made in a day using 1 million machines doing 500eh =  2023

144 blocks set to be made in a day using 100k machines doing 21th =  2012

we used to have a lot of variance back in 2012 to 2013


120 block days followed by 200 block days filled by 130 block days

we do not have 200 block days and we rarely go under 120.

but if I say it is always a prefect six sided die being rolled a lot to get 144 blocks

(think of the difficulty software doing the perfect rolls)

as the diff goes up more rolls are needed for a block so is the mechanism generating flatter variance.


Ie 100,000 spins of a die = average to hit a block due to a diff of 100k
vs 1 trillion spins of a die = average to hit a block due to a diff of 1t

or is the random software generator looking for the block  not acting that way.

My small stupid mind has never fully understood this aspect of random software generation.

I know that we set limits on the software saying 1 trillion or 65 trillion hits a block

but as diff expands do we even out variance since we approach infinity. ( I know we are never getting to infinity)



I keep thinking the larger diff gets and to more machine running it should flatten variance a little.