[r-t] Cambridge Major Tenors Together - The Full Monty

[Andrew Johnson]

I have created a collection on CompLib 
https://complib.org/collection/11125 based on 
https://changeringing.co.uk/cambridgefm.htm
I also went through all the compositions on CompLib with the same 
characteristics:
Only Cambridge Surprise Major
>= 5000 changes
Tenors together
Tenors strictly together
Standard calls
Normal start

https://complib.org/composition/search?backstrokestart=false&midleadstart=false&variablehunt=false&variablecover=false&standardcalls=true&tenorstogether=true&strictlytenorstogether=true&minimumlength=5000&nummethods=1&method-1-value=true&method-1-title=Cambridge%20Surprise%20Major&sort=length

and found 74 compositions. Some of those compositions didn't seem to be 
small variations of the 48 representative compositions, so I added all the 
compositions to the collection, and categorised the ones over the 48 as 
extra or similar. [I think the 48 were just grouped by parts, length and 
number of befores].

For similarity I used the following:
Generate the rows for each composition.
When comparing a composition to a reference composition, chose the 
rotation / reversal which maximises the number of row + next row pairs in 
the composition which match a row + next row pair in the reference.
See 
https://lists.ringingworld.co.uk/pipermail/ringing-theory_bellringers.org/2019-February/026966.html

For each composition, try all the other compositions as a reference 
composition and find the one with the best match.

That worked reasonably well, and found compositions such as 
https://complib.org/composition/35968 which wasn't close to anything in 
the Full Monty list.

./c80532.rows=5346 rows c35968.rows=5024 rows rotate 4706:>15643278 
start=16542378 same rows=2630 52.35% same next row=2471; 93.95% of same 
rows; 49.18% of all rows

As a comparison
https://complib.org/composition/67542
https://complib.org/composition/33105
c67542.rows=5120 rows c33105.rows=5120 rows rotate 3072:>14563278 
start=16523478 same rows=5120 100.00% same next row=5112; 99.84% of same 
rows; 99.84% of all rows

look similar according to my calculation.

These look very similar:
https://complib.org/composition/10029
https://complib.org/composition/67414
./c10029.rows=5600 rows c67414.rows=5184 rows rotate 0:>12345678 
start=12345678 same rows=5184 100.00% same next row=5181; 99.94% of same 
rows; 99.94% of all rows

If the conductor was calling #67414, but you thought the conductor was 
calling #10029 only 3 times would you be surprised by the next row if you 
were just considering the current row.

Perhaps that calculation should just look at the lead end / lead head 
pairs, or lead end / lead head pairs at a tenors together calling position 
but it works okay for this situation. Consider just the lead end / lead 
head might be better for methods where a different course could be singled 
in at one of two places - should every row pair in the singled in course 
be counted as a difference or just the calls?

Andrew Johnson
Twyford







Unless stated otherwise above:
IBM United Kingdom Limited - Registered in England and Wales with number 
741598. 
Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://bellringers.org/pipermail/ringing-theory/attachments/20210604/ff492723/attachment.html>