Quote from: BurtW on August 05, 2019, 12:38:57 PM

For my real job I am writing all the TCG and secure boot ROM firmware for a next gen SSD controller ASIC. This SSD controller ASIC happens to have a built in hardware crypto engine for AES, SHA, HMAC, RSA, ECC, etc. I was thinking I could download a special test firmware into the SSD that would use the built in hardware crypto engine to do this calculation. It would be incredibly fast. I could justify downloading it to an entire rack of SSDs during manufacturing in order to do a "burn in test" of the crypto hardware on the drive. Should be fun.

!!!!!
than
(auto-translate)

Quote from: kzv on February 15, 2019, 09:01:38 PM

Now some super secret asic boost algorithm is being heard. Who knows what about him?

Quote from: kzv on February 16, 2019, 05:47:44 PM

Consider several ways to find a hash with a given complexity
1. Sequential nonce from zero
2. Sequential search not from zero, but from a random point
3. Random search nonce.

Now each of the methods is separate. For clarity, let nonce vary from 0 to 1,000,000 and let us look for a hash with three zeros

Sequential nonce from zero .
In this case, the miner will find the hash on average for 3845 iterations, but with a probability of 61% iterations will be less. That is, Of the 100 hashes found, 61 hashes will be found faster than in 3845 iterations.

Sequential enumeration not from zero, but from a random point
In this case, the miner will find the hash on average for 3845/2 = 1923 iterations, while with a probability of 39.4% iterations will be less (I calculated this additionally). That is, out of 100 hashes, 40 will be found faster than in 1923 iterations.

Twice as fast as in the first case! This is because the random beginning, with a normal distribution, will fall on average to the center of the segment for enumeration.
Also in this method with a probability of 259/1000000 the hash will be found immediately.

Random brute force nonce.
In this case, you can never find the desired hash at all, but with a probability of 259/1000000, the hash will be found on the first try. If the random distribution is distributed normally, then the probabilities should add up, which means after 1923 iterations the probability of finding a hash will be already 50%. That is, out of 100 hashes, 50 will be found faster than in 1923 iterations.

As a result, it turns out that the first algorithm is the slowest. The third algorithm is the fastest. If I havent messed up anything of course ...

and googling "Probabilistic Bitcoin Mining Method"
What u know about this?
(given the probabilistic nature of the kangaroo, this can help)