[r-t] Spliced surprise major: John M Goldthorpe

[John Goldthorpe]

Philip Earis wrote:

>I rang in a peal of 45 spliced surprise major yesterday
>(http://www.campanophile.co.uk/show.html?Code=31364), composed by John
>Goldthorpe.  The peal consisted of all the methods from Smith's 23 and
>Chandler's 23 (41 distinct methods), with 4 additional methods inserted
>to take the composition up to 10k. The composition is copied below.
>
>I think it's a very impressive compositional achievement to be given 45
>different treble-dodging major methods and fit them all into a
>composition. The atw peals of more spliced major methods that I've seen
>contain many (random) methods that fit a composition, rather than being
>a composition that contains only a specified group of methods.
>  
>
Hi Philip,

I'm pleased you liked it.  I think perhaps you have provided the answer 
to why many compositions of atw spliced have random methods.  It's 
because the methods have been fitted in to a composition, rather than 
the composition being fitted to the methods.  If you look at Norman 
Smith's or Chandler's 23, it is obvious that the methods were fitted in 
to a predefined, and consequently neat, composition.  I could be wrong, 
but I assume that Chandler simply took Norman Smith's composition and 
substituted harder methods.  It also appears that he gave up trying to 
find a substitute for Pudsey.

>I wonder if John wants to explain how he put his composition together?
>He must certainly have some impressive resources in his toolkit - the
>use of singles adds to this impression! And what is the maximum number
>of methods that I could give him for him to stand a reasonable chance of
>putting an atw composition together, I wonder?
>  
>
I doubt my "toolkit" is as impressive as many other subscribers to this 
list.  I basically just have a fairly general search program that does 
an exhastive search of a 'map' (if that's the right term).  The map is 
generated using the list of methods, the list of part heads, and a list 
of list of allowable course or lead heads.  Also, the search program can 
be given a composition to start with which it will then attempt to 
extend by finding somewhere to insert a new block.  In fact, it normally 
just inserts a block in to a 'null' composition.  There is a certain 
amount of skill in using it, but 7 parts of Surprise Major are among the 
easier problems.

Having said that, the 45 method composition in question was pushing it 
to the limit.  I starting by trying to create a neat composition along 
the lines of Norman Smith.  When I'd got something that looked quite 
nice I used that as the base composition.  I extended the list of lead 
heads that were allowed and searched again.  Eventually this produced a 
block containing all the remaining methods that could be inserted in to 
the first block.  Only one answer came up, though if I'd left it going 
longer there probably would have been more.  I doubt the whole process 
took more than an hour or two.  (I'm not very patient when it comes to 
composing).  Anyway, that explains why the composition has a neat 
section, and a more random section with singles.

John