thanks, its almost as complicated as actual time-travel

I think I get it tho.
but I don't understand this part:
Now here's the weird part, we retarget after blocks 3, 7, 11, 15, and for block 3 we use 0 as first and 3 as last, for 7 we use 4 as first and 7 as last, ...
so what happens if an attackers chain has blk timestamps like this:
blk# 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
time 0 1 2 30 4 5 6 70 8 9 10 110 12 13 14 150
?
first period (#3 - #0) is 30s as before
2nd period is (#7 - #4) ... 66s
3rd period is (#11 - #8) ... 104s
what does minusing #4 from #7 have to do with it? why does it use only the timestamps from particular blocks to calc the diff adjustment?
how can u do less work but do more work? its because its not counting all the work its skipping some and only counting others towards the adjustment/ ?
plz excuse my ultra nub