[possibly not an issue] unexpected long-time generation on m1 from 9+ digits

[lebed2045]

Readme stays that on M1
(8-core GPU) performance is 45.0 MH/s,
while it shows only 18 MH/s (tested on 4 different m1 air 16Gb), for 8 digits in the matching pattern it takes 1-2 minutes on average, as expected, example of command:

./profanity2.x64 \
--contract \
--matching 7777XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX7777 \
-z 94243fddbb2f775084b9fa0600a10632b3122266dac2b88d438388b82ce1b8888055551cb3c0e39391d61bb48c2643dad6af7d0f97b989f16ea2d6c2df5b41c2

however, just one digits more takes not 16x times as expected, but whole 100x more, example of command:

./profanity2.x64 \
--contract \
--matching 77777XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX7777 \
-z 94243fddbb2f775084b9fa0600a10632b3122266dac2b88d438388b82ce1b8888055551cb3c0e39391d61bb48c2643dad6af7d0f97b989f16ea2d6c2df5b41c2

might be the case of thermal-throttling, but result that for m1 the effective hashrate is merely 2 MH/s

[k06a]

Hi, --contract flag decreases speed due one more hash computation, moreover check out Settings => Battery => Energy Saving option, if enabled it limits speed significantly.

[k06a]

Take a look at the thing we built on Ethdam.com:
https://github.com/1inch/address-token
https://github.com/1inch/ERADICATE3
https://opensea.io/collection/1inch-address-nft-1

Gonna redeploy it one more time with improved attributes.