I would expect about 9% variance due simply to random chance if this metric were calculated with a sliding window of one day. I would expect even larger variance if that sliding window were smaller (as appears to be the case here).
Note that there is also a strong periodic daily cycle causing swings up and down, and almost certainly due to those miners turning their machines off at night.
Topic: Hash rate question - WTF variance? (Read 1039 times)
Quote
This has indeed happened before, and it will happen again. Like you, I wonder why.
We used to explain it by saying "mystery miner", kind of like "the manipulator" of today.
We used to explain it by saying "mystery miner", kind of like "the manipulator" of today.
Oh I wasn't saying it's a new phenomenon, it happens all the time. The pic I posted is just a good example of it.
As I understand it the "hash rate" is a guess based on the frequency of blocks, and because the blocks are mostly random intervals that's why the hash rate swings all over.
This is the familiar screenshot taken from Bitcoinwatch.com.
Why would the range oscillate within about 1.5 terahashes for a few days, then soar almost 3 terahashes in one day? That is an increase of like 35%
I would like some explanation from any smart people out there... you would think the rate would follow a much more normalized 24-hr pattern. Is it a week vs. weekend thing? wtf?
Jump to: