I know I'm jumping in here kind of late but just wanted to say thanks for your SeedSigner/ Sparrow Wallet series. I have a question about using the dice method for generating the seed. I took 5 identical dice and rolled each die individually then recorded the results and repeated this 10 times for a total of 50 dice rolls. I would have expected the distribution of digits to be somewhat uniform but there appears to be an uncanny direct inverse relationship between the individual digits and their frequency. That is: 1 had the highest frequency and 6 the lowest and if you plotted it on a graph it would be almost linear. Are my dice loaded? Is my seed compromised? I have not repeated the experiment to see if I get different (more uniform) results.
Thank you so much, really helpful video. Can you drop a link to where you got your seedsigner kit, or did you buy the unit whole? Its look great, also seems to have a solid build.
I think there is a problem with the way seedsigner uses dice input. Using the numbers 1-6 and omitting 0,7-F gives biased entropy output. This is something addressed by Ian’s tool, at the expense of additional rolls.
Maybe raise this with the developers in telegram, i'm sure they can give an answer. I have had similar thoughts, it might relate to this: x.com/coletu/status/1648444126059442177?s=46
