...
4. Increasing freq won't increase output of found nonces, and won't increase quality nonces (which has higher diff), this is the key to success in mining. We want the nonce which can reach the target or higher diff. Now, too many nonces with below 10k difficulty.
5. Higher freq will produce more failed nonces which can not pass 1st test under asic boost mode. ASIC boost mode in BM1397 chip may limit the quality of nonces. I turn off asic boost, and got more quality nonce or more higher diff nonces. I set freq at 350MHZ and got 320 GH/s. No nned higher freq.
...
Ah sorry, this isn't correct.
There's no such thing as making changes to get better 'quality' nonces.
If it mines twice as fast it will get twice as many nonces.
Due to Proof of Work, you expect that twice as many nonces will have, on average, twice as many of those nonces above any given difficulty.
This result can be achieved by either 1) mining twice as fast or 2) mining with twice as many miners at the same speed.
The ticket mask decides which nonces the chip returns and I set that to 16.
If you set it to 32, you'd get half as many nonces in the same time, with the same average expected difficulty.
Setting it higher just reduces the amount of data that travels back from the chip to cgminer by intercepting it.
It has no effect on the 'quality' of the data, it just affects the average difficulty of the data by removing the lower difficulty nonces from the returned data, the chip still generates this data, it's just removed from the returned data, i.e. less data returned.
Read here, coz it would appear you may not understand the meaning of difficulty:
https://kano.is/index.php?k=workdiff