1 hour delay in the blockchain - is this unusual?

[solex]

Today (Jan 19th) there was a 1 hour delay between blocks 217134 and 217135.
Should the frequency of deviation from the ideal 10 minutes decease over time - or could this happen at any point in the future...?  It may matter a lot if/when BTC is widely used in global commerce.

Block         Elapsed                trans         Value
217135   3 minutes                777    48480.28491324 BTC        BTC Guild   243
217134    1 hour 3 minutes      405    10,197.86 BTC               76.112.4.167    159.00
217133    1 hour 16 minutes    109    425.40 BTC                    BitMinter    37.38
217132    1 hour 17 minutes    594    5,399.05 BTC                 82.72.179.204    255.34
217131    1 hour 14 minutes    11     5,843.06 BTC                  Deepbit    3.46
217130    1 hour 33 minutes    236    12,299.30 BTC                EclipseMC    115.76
217129    1 hour 53 minutes    570    7,370.18 BTC                 BTC Guild    220.05

[The Fool]

Likely it's an indication of decreased difficulty and hashing power in the bitcoin network. It could be a fluke of chance that will be made up with a 1 second block. I don't know the specifics.

Generally, if hashing power increases drastically you will have blocks resolved under 10 minutes and the network adjusts the difficulty upwards. If hashing power decreases, you'll see blocks like this.

Let's see if someone with knowledge of the source code has a better answer.

[Stephen Gornick]

Let's see if someone with knowledge of the source code has a better answer.

Someone with the knowledge (and desire) to compute the probability could say how often that occurs.  It isn't that rare, ...  maybe occurring once a week (wild-assed guess).

[The Fool]

Let's see if someone with knowledge of the source code has a better answer.

Someone with the knowledge (and desire) to compute the probability could say how often that occurs.  It isn't that rare, ...  maybe occurring once a week (wild-assed guess).

Probability is a factor, definitely.

https://en.bitcoin.it/wiki/Target

For reasons of stability and low latency in transactions, the network tries to produce one block every 10 minutes. Every 2016 blocks (which should take two weeks if this goal is kept perfectly), every Bitcoin client compares the actual time it took to generate these blocks with the two week goal and modifies the target by the percentage difference. This makes the proof-of-work problem more or less difficult. A single retarget never changes the target by more than a factor of 4 either way to prevent large changes in difficulty.

[deepceleron]

Someone with the knowledge (and desire) to compute the probability could say how often that occurs.  It isn't that rare, ...  maybe occurring once a week (wild-assed guess).

Maybe someone could have already said... (or someone could search before starting a thread.):

So the chance a block won't be found after 10, 20 minutes, etc. (and the 1 in x chance)

Code:

>>> for x in range(10,131,10) :
	print x, '%2.5f%%' % (100*math.exp(-x/10)), '%1.2f' % (math.exp(x/10))

10 36.78794% 2.72 (average block length)
20 13.53353% 7.39
30 4.97871% 20.09
40 1.83156% 54.60
50 0.67379% 148.41
60 0.24788% 403.43
70 0.09119% 1096.63
80 0.03355% 2980.96
90 0.01234% 8103.08
100 0.00454% 22026.47
110 0.00167% 59874.14
120 0.00061% 162754.79
130 0.00023% 442413.39

And every day we can reasonably expect a block longer than 49 minutes, every week one longer than 69 minutes, every month 83 minutes:

Code:

>>> for x in [ 10*math.log(144), 10*math.log(144*7), 10*math.log(144*365.25/12)]:
	print x, '%2.5f%%' % (100*math.exp(-x/10)), '%1.2f' % (math.exp(x/10))

   
49.6981329958 0.69444% 144.00
69.1572344863 0.09921% 1008.00
83.8548870042 0.02282% 4383.00

So if hashrate maintains at a level as predicted by difficulty, you would expect a >60 minute block every three days (red), however if hashrate drops to 83% of the predicted rate, a >60 min block averaging one a day would not be unusual (orange) (don't forget, many long blocks one day and none for many days in a row also would not be unusual, variance tells us the unlikely is just as likely).

You can get a rough idea if the hashrate may be decreasing here: http://dot-bit.org/tools/nextDifficulty.php
It has both difficulty predictions for Bitcoin and Namecoin based on several periods. As Namecoin is also mainly merge-mined by Bitcoin miners, it can be a secondary indicator to evaluate the true direction of hashrate (as block finding, even over many blocks, is random, and can be non-predictive of actual hashrate).


In addition, I just asked myself "Does it matter if it's the weekend or not?". Well, here's the number of blocks found by day of the week for Jan 16 2012 (Monday) to Jan 13 2013 (Sunday); it's hard to assign any significance to any day:

[Stephen Gornick]

Someone with the knowledge (and desire) to compute the probability could say how often that occurs.  It isn't that rare, ...  maybe occurring once a week (wild-assed guess).

Maybe someone could have already said... (or someone could search before starting a thread.):

LOL, sorry   Nice update to the title.   

Fantastic info, thanks!

[21after2]

I sent a payment for a sale to a user here on the forum and it's taking quite a while to show up in his Electrum wallet. Are there more delays going on right now or something?

[DeathAndTaxes]

I sent a payment for a sale to a user here on the forum and it's taking quite a while to show up in his Electrum wallet. Are there more delays going on right now or something?

There is no such thing as "more delays going on right now".  The time between blocks is random.  The time between any two blocks is random.  The time until the next block is random.  The only thing which is certain is that in the long run on average the block interval will be ~10 minutes.  BTW the last block was found ~1 minute ago however it may not contain your transaction.

[21after2]

I sent a payment for a sale to a user here on the forum and it's taking quite a while to show up in his Electrum wallet. Are there more delays going on right now or something?

There is no such thing as "more delays going on right now".  The time between blocks is random.  The time between any two blocks is random.  The time until the next block is random.  The only thing which is certain is that in the long run on average the block interval will be ~10 minutes.  BTW the last block was found ~1 minute ago however it may not contain your transaction.

I gotcha. I've been checking the transaction on the blockchain, which has four confirmations: http://blockchain.info/address/1LHmGT39uLJYSfeKVbgUP5vAgrhf1GaqMK

Blockexplorer shows not redeemed, however. My own wallet simply says seen by one peer, but not in the chain. I don't have a lot of technical knowledge on the subject, so I'm trying to confirm that the seller gets the money I sent him.

[DeathAndTaxes]

Weird your wallet would indicate it is unconfirmed.  It was included in block #217371 which was almost 30 minutes ago.  If your bitcoin wallet up to date (has all blocks, most recent block at time of post is #217374).

[21after2]

Yeah, I'm not sure why it's not updating. My wallet still says not seen in chain (I'm using MultiBit). The seller is using Electrum, and he says the payment hasn't showed up at all in his history. Blockchain has four confirmations now, so I don't get why it's not showing up in either of our clients. I could see MultiBit taking a bit, since it loads a part of the blockchain, but my understanding is that Electrum doesn't load the chain at all. Am I right in thinking that?

EDIT: Here's the Blockexplorer link, if it helps: http://blockexplorer.com/tx/470593e7e59dbf4d3efe900e590fdef6f907c735702e34b2bc2a10fbe123388a
EDIT 2: My wallet is now showing full confirmations, so I messaged the seller to see if it's shown up in his now.

[DannyHamilton]

. . . The seller is using Electrum, and he says the payment hasn't showed up at all in his history. Blockchain has four confirmations now, so I don't get why it's not showing up . . .

There was an Electrum server issue:

there was a server-side problem. I restarted my server, and the transactioni should show up now.

[molecular]

relevant informative thread with lots of math and charts: extremely long blocks - what are the chances?

from that thread:



so yes, it happens

[solex]

Impressive analysis.

I imagine then that one significant function for banking institutions in a bitcoin future would be to facilitate zero confirm transaction processing (for those merchants with a real-time need to allow them, up to certain value limits). By doing so the banks accept the risk that a small percent will be double-spends, reflecting these losses in their fees, which are spread widely and average out.

[mygodpp]

I think it is normal