It's a neat pattern but the alignment looks pretty contrived. Sure you're not just saying "there are repeated nibbles"?
-
-
-
Definitely overfitting is a possibility. But notice the ordering of values in the yellow columns.
-
Regarding contrived alignment, see https://twitter.com/brian_____/status/836338084903976960 … and https://gist.github.com/briansmith/48029577ce4ecad994dc6b22e20158a7 …, which make it seem less silly.
-
Ok. I agree maybe a little less.
End of conversation
New conversation -
-
-
is it statistically significant? We humans are very good at finding nonexistent patterns.
-
Given 5 or 10 20-byte sequences, can we do meaningful statistics? I am not sure.
-
yes, there are formulas to calculate it. Empirically you can generate a random set of similar-sized numbers and search patterns
-
sounds like you might have the background to do that analysis? If so, tell us what you find!
-
I didn't have time to dig the formulas but did an empirical comparison: https://gist.github.com/arisada/cb1dd9e9edbe6ab433a75a8a88cebc28 …pic.twitter.com/8PCFbSXlHq
-
That's good. I also ran the prime curve seeds & randomly-generated values through `ent` w/ similar results.
-
I also found some way to link, for example, parts of the seeds for p384 & p521 to the values 384 & 521, resp.
End of conversation
New conversation -
-
-
399 chars / 16 hex-digits means we should have about 25 repeating nibbles. Your example has 26. I don't see any unusual patterns
-
Yes, I also did this calc too. Definitely overfitting is possible. But notice 22/44/88... sequence repeated in each set of 5.
End of conversation
New conversation -
-
-
prime fields can be even? That's odd.
-
Sorry, I'm not sure what you mean. Please elaborate.
End of conversation
New conversation -
-
-
@matthew_d_green this is numerology. On average, will get repeating digits every 16 bytes. Everything is alignment. -
Yep. But how likely would the 22/44/88... AA sequence repeat?
-
sequences not byte-aligned, so pairs will appear every 8. Much brainpower wasted in 19th century from this.
End of conversation
New conversation -
-
-
what's with the weird alignment? Tbh, that looks like a bit of a stretch
-
Omg DaVinci code that shit!
-
that's what I was thinking. When we want to see patterns, we see patterns :(
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.