On 2019-Nov-14, at 9:59 PM, Brent Hilpert via cctalk wrote:
> On 2019-Nov-14, at 9:30 PM, Jim Brain via cctalk wrote: >> On 11/14/2019 11:05 PM, Brent Hilpert via cctalk wrote: >>> >>> (Without having gone through the code presented in full detail, but >>> thinking from root premises.) >>> >>> ASCII A = 0x41 --> 2 bits on >>> ASCII T = 0x54 --> 3 bits on >> >> I agree, though I believe aT,At,AT,and at are all allowed, which complicates >> things a bit aT is the bad one, where the calc for 7E1 and 7S1/8N1 yield the >> same values... > > ... right, was forgetting about case. > But if you're trying to account for mixed case, I don't think the CR resolves > it anyways: > > ASCII A = 0x41 --> 2 bits on > ASCII T = 0x54 --> 3 bits on > ASCII a = 0x61 --> 3 bits on > ASCII t = 0x74 --> 4 bits on > ASCII CR = 0x0D --> 3 bits on > > In sequence "Ta<CR>", all characters have an odd number of bits on and would > produce the same parity value, and collectively would be indistinguishable > from a corresponding constant M or S parity. Whoops, that should have been: in sequence "aT<CR>".
