vvjmg
short pre-cutover
An app start too short to pin down a seed. Some have candidate seeds, listed on their pages; their counts are plotted by draw with the draws numbered from the first post (an assumption).
- Period (UTC)
- 2012-07-01 11:34:49 → 2012-07-01 23:35:09
- Posts
- 4 (short: 4)
- First / last post
vvjmg/vwd7a- Draws
- Not known: no seed was found, so the posts are numbered from the first post, at draw 0, as consecutive draws (an assumption; the 2012 build starts at draw 0, the 2013 build at draw 1 after its startup draw)
- Posting interval
- 4 h
- Title clock
- UTC+9 at the first post
- Started
- Sun 2012-07-01 00:34–04:34 Pacific (estimate: within one posting interval before the first post)
Notes
- 4 posts 4-hourly (zone +9) before run 10; undecidable: both June boots excluded under the 7-min launch rule, new-boot hits chance-level
Possible seeds
Undecidable. Under the launch rule that 2012 posts follow, both June boots are excluded. On a new boot after the burst of 1 July, six seeds fit, but they are correlated and amount to about three chance events, so the result is chance-level and no seed is listed as a candidate.
The run
Four posts on 1 July 2012, 11:34:49 to 23:35:09 UTC, every four hours, with the title clock on UTC+9. The 2012 build draws Next(8, 64) per post, and the first post is draw 0.
| Draw | Post | Posted (UTC) | N |
|---|---|---|---|
| 0 | vvjmg | 11:34:49 | 37 |
| 1 | vvqf7 | 15:34:53 | 14 |
| 2 | vw1d2 | 19:34:58 | 9 |
| 3 | vwd7a | 23:35:09 | 19 |
Why it is hard
Each post has 56 possible values, so four posts carry about 23 bits: roughly 200 chance matches per starting draw over all seeds. June 2012 seeds are not on the tick grid, so there is no grid factor to help.
The neighbours
Two machines were active in late June. One is on the folded boot of 15 May, pinned to 19:12:40 (runs 4 to 9, including vv2q4 at 02:49 on 1 July). The other is on a boot of 14 June, no later than 10:26:51 (runs 1 to 3). Run 10 is on a new boot of 2 July, between 09:51 and 10:51. The continuation test of every known stream is at chance, and the stream of the burst does not contain this run.
How it was narrowed
| Hypothesis | Candidates | Expected by chance | Seeds found |
|---|---|---|---|
| Folded 15 May boot, launch within 7 min | about 420,000 | 0.04 | 0, so excluded |
| 14 June boot, launch within 7 min | about 830,000 | 0.09 | 0, so excluded |
| Folded 15 May boot, launch within 264 min | about 16 million | 1.6 | 2 (empirical probability 0.42) |
| 14 June boot, launch within 264 min | about 16 million | 1.65 | 0 |
| A new boot after the burst (1 July, 02:49:48), launch within 7 min | about 32 million | 3.2 | 6 (empirical probability 0.15) |
The six hits on the new boot are not six independent chances. Several of them are exactly 970,845 apart, and seeds a fixed distance apart share their first draws in the .NET generator. So they amount to about three independent chance events.
What could decide it
Only a tighter boot: for example, evidence of when the folded machine rebooted between 02:49 and 11:27 on 1 July.
Counts
Posts
| Post | Title | Posted (UTC) | Draw | N | Mask seed | Status | Note |
|---|---|---|---|---|---|---|---|
vvjmg | 201207012036 | 2012-07-01 11:34:49 | 0 | 37 | 587 | explained | 4 posts 4-hourly (zone +9) before run 10; undecidable: both June boots excluded under the 7-min launch rule, new-boot hits chance-level |
vvqf7 | 201207020036 | 2012-07-01 15:34:53 | 1 | 14 | 877 | explained | 4 posts 4-hourly (zone +9) before run 10; undecidable: both June boots excluded under the 7-min launch rule, new-boot hits chance-level |
vw1d2 | 201207020436 | 2012-07-01 19:34:58 | 2 | 9 | 458 | explained | 4 posts 4-hourly (zone +9) before run 10; undecidable: both June boots excluded under the 7-min launch rule, new-boot hits chance-level |
vwd7a | 201207020836 | 2012-07-01 23:35:09 | 3 | 19 | 187 | explained | 4 posts 4-hourly (zone +9) before run 10; undecidable: both June boots excluded under the 7-min launch rule, new-boot hits chance-level |