ANN: BITMAIN has Tested Its 28nm Bitcoin Mining Chip BM1382

[hurricandave]

This page will give you some idea of what you should use as a difficulty increment :- https://bitcoinwisdom.com/bitcoin/difficulty

I don't understand why people keep referencing this website.  I proved it was wrong last change.  It called a 30%+ increase 3 days before the change, when the numbers showed about 24% - 25%.

Right now the numbers are showing a decrease in difficulty.. but this website seems to be arbitrarily picking 11% increase.

M

It's not arbitrary at all. Bitcoinwisdom bases it's estimation on the time it took to find the last 2016 blocks just like bitcoin does.

Do the math yourself.  I did for the last change, and proved it wrong.

M

There is no such thing as "doing the math".

What you are doing is called astrology.

Quite the contrary.  There is such a thing as doing the math.

Get the timestamp of the last block solved before the difficulty change.
Get the timestamp of the last block solved.
Subtract the 1st one from the 2nd one.
Divide that by the numbers of blocks solved since the last change.
That's how often the network is solving blocks since last change.

You'll see that number works out to 10.59 minutes per block so far.  

Last change, 3 days before, bitcoinwisdom was saying 30%.  My math was saying 23.6% at the time, and I figured it would likely increase a little more, but certainly not 6% more.  Actual increase?  24.93%.

https://bitcointalk.org/index.php?topic=656461.msg7517267#msg7517267

M

A single sample will not prove you true or its not scientific.
It can be coincidence. You put you current calculations and look at next change.
At the near of difficulty change they too come to the same %.

I Wonder, if they get their guess of next diff change based on the last 504 blocks, instead of 2016, they don't say that, but the hashrate average is based on 504 blocks, hover over the "?" and that's what it says.

[mdude77]

Quite the contrary.  There is such a thing as doing the math.

Get the timestamp of the last block solved before the difficulty change.
Get the timestamp of the last block solved.
Subtract the 1st one from the 2nd one.
Divide that by the numbers of blocks solved since the last change.
That's how often the network is solving blocks since last change.

You'll see that number works out to 10.59 minutes per block so far.  

Last change, 3 days before, bitcoinwisdom was saying 30%.  My math was saying 23.6% at the time, and I figured it would likely increase a little more, but certainly not 6% more.  Actual increase?  24.93%.

https://bitcointalk.org/index.php?topic=656461.msg7517267#msg7517267

M

This is just a much less accurate way of estimating the hashrate.

Everyone knows that difficulty adjusts to 10 minutes per block based on the time it took to find the last 2016 blocks.

With your estimations you will always see a ~0% increase immediately after difficulty increases because that is the whole point of the difficulty adjustments.

Right now your "math" is suggesting that there will be a 6% decrease. I am willing to bet that that will not happen.

I'm not seeing how it's much less accurate.  How would you accurately predict what the next change will be?  Without math, it's guess work.

My math gets more accurate as we get closer to 2016. 

It's only been 4 days since last change.  Odds are we won't have a decrease.  But suggesting that the 11% predicted on bitcoinwisdom is accurate is crazy.

M

[mdude77]

A single sample will not prove you true or its not scientific.
It can be coincidence. You put you current calculations and look at next change.
At the near of difficulty change they too come to the same %.

I put this math is my app (M's Pool Monitor) quite a while ago and have been watching it as the difficulty changes.  It's spot on every time, because it's math.  That's why I said with certainty in the post above that I don't see how it would be 30% increase 3 days before when the math said otherwise.

M

EDIT: I put the math in my app because none of the sites I looked at gave accurate estimates.  Every single one was wrong, every single time.

[Caesium]

Using block data from the previous difficulty to try and estimate the next one is crazy talk.

It assumes that hashrate added or removed in the last difficulty is part of a trend and that trend will continue for the next one. Fortunately for bitcoinwisdom this is usually true but it is clearly not this time. Right after the last difficulty change they were predicting nearly 20% and I called BS on it then. Its still correcting downwards even now I see.

bitcoinwisdom even blatantly contradict themselves, they're saying the 6 block timespan right now (over the last 504, as mentioned above) is 1h which is bang on target, yet that's going to result in an 11% difficulty increase. :/

So they're using a different number of blocks to work that out, probably 1008, or maybe even 2016, can't be bothered to work it out.

[allcoinminer]

A single sample will not prove you true or its not scientific.
It can be coincidence. You put you current calculations and look at next change.
At the near of difficulty change they too come to the same %.

I put this math is my app (M's Pool Monitor) quite a while ago and have been watching it as the difficulty changes.  It's spot on every time, because it's math.  That's why I said with certainty in the post above that I don't see how it would be 30% increase 3 days before when the math said otherwise.

M

EDIT: I put the math in my app because none of the sites I looked at gave accurate estimates.  Every single one was wrong, every single time.

If it so, that's great, I will watch your app it for some time.
But sampling is not a fool proof way of prooving something, it can be wrong too.
I will check the math behind it too.

[jimmothy]

I put this math is my app (M's Pool Monitor) quite a while ago and have been watching it as the difficulty changes.  It's spot on every time, because it's math.  That's why I said with certainty in the post above that I don't see how it would be 30% increase 3 days before when the math said otherwise.

M

EDIT: I put the math in my app because none of the sites I looked at gave accurate estimates.  Every single one was wrong, every single time.

If your estimations were spot on then why is it saying there will be a 6% decrease?

If you look at the second graph on bitcoin wisdom you'll notice that they are pretty accurate.

[mdude77]

I put this math is my app (M's Pool Monitor) quite a while ago and have been watching it as the difficulty changes.  It's spot on every time, because it's math.  That's why I said with certainty in the post above that I don't see how it would be 30% increase 3 days before when the math said otherwise.

M

EDIT: I put the math in my app because none of the sites I looked at gave accurate estimates.  Every single one was wrong, every single time.

If your estimations were spot on then why is it saying there will be a 6% decrease?

If you look at the second graph on bitcoin wisdom you'll notice that they are pretty accurate.

The estimate gets more accurate as we get closer to the next difficulty change.  If difficulty were to change now, there would be a 6% decrease.

M

[Caesium]

The estimate gets more accurate as we get closer to the next difficulty change.  If difficulty were to change now, there would be a 6% decrease.

This is exactly the same as ignoring the previous difficulty entirely and just using the time delta between blocks since it changed - the only difference is it produces wild swinging numbers for the first few days, instead of an arbitrary steady number that relies on irrelevant data.

[philipma1957]

Made my first bitmain order. Anyone know how long it takes to get a payment verification email?

I don't think that I ever got a 'payment verification email' from Bitmain.  Just log onto your account on the Bitmain web site, if on the order list it says that the status is 'Paid Unshipped Valid' then you are fine.  If you click on 'view' for that order, you'll see the order details and that should say 'Pay confirm'.  I usually print a pdf copy of the order 'just in case'.
When it ships, sometimes you may get email notification, however the order status on the Bitmain page will provide the shipping status and tracking information, which you should then follow with whoever ships your miner (my preference is UPS, as they have been much more reliable than DHL, usually UPS delivered it ahead of the projected ETA).

Cheers

After I place order (which I pay in 1-3 minutes)  I always get the billing email from them 10 to 20 minutes later.  I reply to the billing email  with  "I paid this order"  and give them the tx id

[mdude77]

The estimate gets more accurate as we get closer to the next difficulty change.  If difficulty were to change now, there would be a 6% decrease.

This is exactly the same as ignoring the previous difficulty entirely and just using the time delta between blocks since it changed - the only difference is it produces wild swinging numbers for the first few days, instead of an arbitrary steady number that relies on irrelevant data.

lol.  Actually it's only wild swinging numbers for about 24 hours, then it starts to steady out.  And it gets more accurate as time goes by.  Last difficulty within a few days of change it was saying 15%.

And you have to ignore the prior difficulty entirely...

M

[philipma1957]

The estimate gets more accurate as we get closer to the next difficulty change.  If difficulty were to change now, there would be a 6% decrease.

This is exactly the same as ignoring the previous difficulty entirely and just using the time delta between blocks since it changed - the only difference is it produces wild swinging numbers for the first few days, instead of an arbitrary steady number that relies on irrelevant data.

lol.  Actually it's only wild swinging numbers for about 24 hours, then it starts to steady out.  And it gets more accurate as time goes by.  Last difficulty within a few days of change it was saying 15%.

And you have to ignore the prior difficulty entirely...

M

I go here: 

http://bitcoincharts.com/                            they say 16,818,461,371 jumps to 16,819,948,837  that is under 1 percent

https://bitcoinwisdom.com/bitcoin/difficulty  they say 16,818,461,371 jumps to 18,726,927,772 that is 11.35 percent



I do this 1 time each day for the 11-12 days between jumps.       It just about always ends up in between the two.

So that will be 6%  based on today's numbers.   

[rograz]

We are also entering the period of when mining gear has started to be turned off completely, even under optimal conditions with dirt cheap power I can't see anyone keeping 110/130nm gear running for much longer. Next to go will be all the BFL 65nm gear (to be fair after they did start to ship they actually did ship quite a few PH), after that it will be Avalon gen 2 etc. It could be as much as 10-15% of the network from all "old tech" sources in the next couple of difficulty jumps.

[s1gs3gv]

I must say, in the nicest way possible, that this thread now sounds like a bunch of fellas peering (no pun intended) at the tea leaves hoping against hope to find some glimmer of a reason to believe that the S3, and bitcoin mining in general with current generation (or next generation) equipment will retain the smattering of microscopic profitability that machines like the S1 had for just long enough that  they can rationalize buying more gear to themselves before they have to accept that it is utterly and absolutely impossible to make even a few satoshis in profit over the lifetime of new gear.

You can always just tell the missus: Sorry honey, the guys on bitcointalk talked me into it.

[mdude77]

I must say, in the nicest way possible, that this thread now sounds like a bunch of fellas peering (no pun intended) at the tea leaves hoping against hope to find some glimmer of a reason to believe that the S3, and bitcoin mining in general with current generation (or next generation) equipment will retain the smattering of microscopic profitability that machines like the S1 had for just long enough that  they can rationalize buying more gear to themselves before they have to accept that it is utterly and absolutely impossible to make even a few satoshis in profit over the lifetime of new gear.

You can always just tell the missus: Sorry honey, the guys on bitcointalk talked me into it.

The price is too high.  I'm not in.  Bitmain needs to lower it a few notches if they want my coin.

M

[s1gs3gv]

We are also entering the period of when mining gear has started to be turned off completely, even under optimal conditions with dirt cheap power I can't see anyone keeping 110/130nm gear running for much longer. Next to go will be all the BFL 65nm gear (to be fair after they did start to ship they actually did ship quite a few PH), after that it will be Avalon gen 2 etc. It could be as much as 10-15% of the network from all "old tech" sources in the next couple of difficulty jumps.

Unlikely, as long as the urban myth regarding an explosion in BTC/fiat exchange rate lives. People will run them at a loss hoping someday to turn their new bitcoin dust into crisp $100 bills.

[s1gs3gv]

I must say, in the nicest way possible, that this thread now sounds like a bunch of fellas peering (no pun intended) at the tea leaves hoping against hope to find some glimmer of a reason to believe that the S3, and bitcoin mining in general with current generation (or next generation) equipment will retain the smattering of microscopic profitability that machines like the S1 had for just long enough that  they can rationalize buying more gear to themselves before they have to accept that it is utterly and absolutely impossible to make even a few satoshis in profit over the lifetime of new gear.

You can always just tell the missus: Sorry honey, the guys on bitcointalk talked me into it.

The price is too high.  I'm not in.  Bitmain needs to lower it a few notches if they want my coin.

M

Ditto. Its a very cynical and hard nosed game these miner manufacturers are playing these days. Every last one of them.
Greed.

[vortexz]

S3 is still too expensive, especially when delivery is not immediately.
I hope they will release coupons soon !

[oskuro]

then dont buy anything. dont buy S3 or whatever hardware out there.

Life is a shit.

Ill be right back, im going to suicide  

S3 is so beautiful, how its not going to ROI? xD

[denniscdunbar]

S3 is still too expensive, especially when delivery is not immediately.
I hope they will release coupons soon !

At the moment, what other device under 1BTC is a better price than AntMiner s3?  It may not be the perfect price for ROI but I can't seem to find anything better.  Please post other options in the 500GH range.

Really like the simplicity of AntMiners with only a network cable and 2 power plugs.  My 2 s1's run none stop without issues.  The new s3 looks sexy with the new cover.


BTCBTC
DoubleDD

[oskuro]

S3 is still too expensive, especially when delivery is not immediately.
I hope they will release coupons soon !

At the moment, what other device under 1BTC is a better price than AntMiner s3?  It may not be the perfect price for ROI but I can't seem to find anything better.  Please post other options in the 500GH range.

Really like the simplicity of AntMiners with only a network cable and 2 power plugs.  My 2 s1's run none stop without issues.  The new s3 looks sexy with the new cover.


BTCBTC
DoubleDD

yeah and paying in BTC is so cheap, considering you have 1 S3 for 599 EUR in the "famous" online shops