So I did a little analysis on the number of blocks generated in the past 3 days (since the timewarp was 'set in motion'). I've broken them down into 12-hour (0.5 day) increments from 11:00 to 23:00 and vice-versa. I then compared the number of blocks we'd expect in these increments given one block per minute, vs the number of blocks we have actually had.
I further considered both cumulative variance (the total blocks generated between now and then vs what we would have expected) and the period variance (the number of blocks generated in any 12 hour increment vs what we would have expected). For any 12 hour period we should be getting, on average, 720 blocks.
| Days | | No. Mins | | Block height | | Total Blocks | | Cumulative Variance | | Period Blocks | | Period Variance |
| 0 | | 0 | | 236012 | | 0 | | 0.00% | | 0 | | 0.00% |
| -0.5 | | 720 | | 235287 | | 725 | | 0.69% | | 725 | | 0.69% |
| -1 | | 1440 | | 234495 | | 1517 | | 5.35% | | 792 | | 10.00% |
| -1.5 | | 2160 | | 233792 | | 2220 | | 2.78% | | 703 | | -2.36% |
| -2 | | 2880 | | 233083 | | 2929 | | 1.70% | | 709 | | -1.53% |
| -2.5 | | 3600 | | 232353 | | 3659 | | 1.64% | | 730 | | 1.39% |
| -3 | | 4320 | | 231678 | | 4334 | | 0.32% | | 675 | | -6.25% |
As we can see, the major number that jumps out was the 10% period variance in the 12-hour interval between block 235287 and block 234495. This means there was 10% more blocks generated than we would have expected.
Personally I don't believe this to be outside of the bounds of normal generation-rate fluctuations, especially given the prolonged increase in difficulty over that period of time (see difficulty on
http://chainradar.com/xmr/chart between 11am and 11pm on 26th September). This suggests there was an increase in network hashrate over this period, which could account for the 10% extra blocks generated while difficulty caught up.
Obviously this is just my opinion and it's possible that it may be the symptom of some sort of timewarp attack, but I do not believe this to be the case. Especially given that there has only been 0.32% more blocks generated than there should have been in total over the past 3 days.
I'll continue to analyse the block generation rate in 12-hour intervals going forward, and post the results on these forums, pointing out anything I spot that seems anomalous.
I welcome the opinion of others on these figures, or any suggestions on how to improve this analysis going forward.
tl;dr there's nothing wrong with the block generation rate since the 'attack' started. Will keep an eye on it going forward.
~pst~