Something unexpected happened during testing.
A wallet that had never interacted with the protocol before submitted a failed claim and then a successful claim. The successful solution was Tier 10 (roughly a 1 in 16^10 event), with a nonce value of 18.
The interesting part was the timing:
- The failed claim used nonce 17 but was rejected because the round had already expired.
- The failed claim was submitted 300 seconds after the round started, even though the submission window is only 12 seconds.
- The next claim used nonce 18 and succeeded.
- The successful claim was submitted 480 seconds after that round started (42 Ethereum blocks later), about 180 seconds after the failed submission.
At first glance this looked strange because a very low nonce satisfying a Tier 10 target is extremely unlikely if only a small sequential search range is tested.
After investigating, the more likely explanations are:
- the miner used a custom implementation rather than the reference miner,
- the miner was able to precompute because the challenge became available before the round opened (extra 12 seconds window before submission period starts),
- or it was simply a rare probabilistic event.
Shardhash contract: https://etherscan.io/address/0x295121422b9d0fd3cbddc9e203ae9b4a1eff0082
0 comments