That's exactly the result I was looking for and demonstrates just how versatile Lilypond can be with the necessary know-how.
Although this certainly serves as a work-around, I'd still be interested in sponsoring a solution where Lilypond can work out the control points itself. The mark is quite common in the music I use Lilypond for, and I have several (hundred) short pieces in separate files which I include in layout files to make 'sets.' Often a piece can be included in more than one set. As I use 'minimal page-turning' to shrink the pieces to one page, I'd rather not have to hand-tune the coordinates each time Lilypond decides to fit the piece to, say, three lines instead of four, depending on the other pieces in the set. The time spent hand-tuning the ties could better be spent practising playing the pieces instead :) Off topic - would be interested in reading your bezier documentation - I have another un-lilypond-related reason for wanting to understand them. >----Original Message---- >From: [EMAIL PROTECTED] >Date: 01/08/2008 21:55 >To: "David Pounder"<[EMAIL PROTECTED]> >Cc: <lilypond-user@gnu.org> >Subj: Re: Loop-the-Loop Ties > > >More complicated shapes would normally be implemented as >two cubic Bézier curves spliced together. This would work >quite well for the shape you want. Here's a lash-up using >three spacer notes in parallel to carry two slurs, just to show >what could be achieved (this is hardly a practical technique >for real music, although if there are only one or two such >slurs and it is important enough, maybe it could serve.) >The 8 values are to be determined by trial and error. > >\relative c' { > << > { e4 e } > { > \override Slur #'control-points = > #'((0.75 . -3.5) (1.75 . -5) (2.75 . -3.5) (2.0 . -3.5)) > s8( > \override Slur #'control-points = > #'((-1.0 . -3.5) (-1.75 . -3.5) (-0.75 . -5) (0.25 . -3.5)) > s)( s) > } > >> >} > >The algorithm to work out 8 possible control points would not >be too difficult, and the routines to draw Bézier curves are >already available in LilyPond, but I don't know how difficult it >would be to select an optimum curve given the surrounding >environment. In any case this would need to be done by one >of the code writers (I'm not one), so that's as far as I can take >it, I'm afraid. I've only pursued it this far because I shall >shortly be documenting Bézier curves and their control-points >and I needed to understand them better. Thanks for the >incentive :) > >Trevor > >----- Original Message ----- >From: "David Pounder" <[EMAIL PROTECTED]> >To: "Trevor Daniels" <[EMAIL PROTECTED]> >Cc: <lilypond-user@gnu.org> >Sent: Friday, August 01, 2008 7:27 PM >Subject: Re: Loop-the-Loop Ties > > >Thanks for the response, but as you say, the overall shape of the tie gets >lost with the loop. > >Would this be horrendous to implement as a new feature, and if it is >feasible, how would I go about sponsoring? > >David. > >------- Original Message ------- >From: "Trevor Daniels" <[EMAIL PROTECTED]> >To: [EMAIL PROTECTED], lilypond-user@gnu.org >Sent: 1.8.08, 18:59:54 >Subject: Re: Loop-the-Loop Ties > >You -can- make a cubic Bézier "loop-the-loop" by crossing over the two >intermediate points, but then there is not enough control left to achieve >a nice downward shape on the two halves. Also you would need to tune >every such slur manually, which is quite tedious. > >Here's an example which shows what might be achieved: > >\relative { > \override Slur #'control-points = > #'((1 . -3.5) (6 . -2) (-1 . -2) (4 . -3.5)) > e( e) >} > >Trevor > > >----- Original Message ----- >From: <[EMAIL PROTECTED]> >To: <lilypond-user@gnu.org> >Sent: Friday, August 01, 2008 10:23 AM >Subject: Loop-the-Loop Ties > > >> Good morning list. >> >> Apologies for the cryptic subject header, but what I need to do is >> difficult to express in words. >> >> In short, I need to show two notes tied, where the tie forms a circle >> in the middle, performing a 'loop the loop' before ending at the second >> note. I realise that this probably makes no sense whatsoever, but such >> a tie can be seen in bar 4 and also in the final bar of the following >> link: http://www.abdn.ac.uk/scottskinner/collectiondisplay.php? >> pageNum_collection=4&totalRows_collection=13&Record_Type=EC&title=The% >> 20Elgin%20Collection&link=other.shtml%23ec >> >> This mark itself is fairly common in Scottish fiddle music and is used >> to indicate that the notes, though tied, should be emphasized >> separately by the bow, so that the two notes can be heard separately by >> the listener. >> >> I am not a great expert on bezier curves and was wondering if it is >> possible to use 'control-points' to do this, or would it have to be a >> new feature request? >> >> Would be grateful for any help, as I don't know of any other way to >> represent this bowing. >> >> - David >> >> >> >> _________________________________________ >> Get up to 33% off Norton Security only from Tiscali - >> http://www.tiscali.co.uk/securepc >> >> >> >> _______________________________________________ >> lilypond-user mailing list >> lilypond-user@gnu.org >> http://lists.gnu.org/mailman/listinfo/lilypond-user >> > > > > > >_______________________________________________ >lilypond-user mailing list >lilypond-user@gnu.org >http://lists.gnu.org/mailman/listinfo/lilypond-user > _________________________________________ Get up to 33% off Norton Security only from Tiscali - http://www.tiscali.co.uk/securepc _______________________________________________ lilypond-user mailing list lilypond-user@gnu.org http://lists.gnu.org/mailman/listinfo/lilypond-user