Disklavier

I would like to offer my comments about the e-competition and the 
MIDI files from that competition.

First, it is a shame that Wen-Yin Chan's preliminary files were 
lost.  I hope that Yamaha will take steps to make this occurrence 
less likely in the future, for the sake of the contestants as well 
as us file users.

Second, can anyone shed light on why finalist Tao Lin was not 
awarded anything.  From what I hear on the MIDI files, his playing 
was rather good if not excellent. Was there a big boo-boo on the 
quintet or concerto portion of the competition?  Was he disqualified 
for some reason?  

Third, it is a testimony to the robustness of the Disklavier Pro 
recording and editing process that the files were able to be 
segmented, edited, and posted so quickly -  to say nothing of how 
magnificent they sound.  

Finally, some suggestions for downloads.

1.  I believe all of the files are OK, but a number of my initial 
download attempts only captured part of the file.  The telltale sign 
(at least for me)is a file size of 64k in Windows Explorer.  In 
every case, I got "incomplete" errors when trying to play files of 
that size, even though the incomplete file would begin to play.  A 
second attempt to download usually captured the complete file. So I 
advise everyone to check your e-competition downloads and try to 
play all of the files of size 64k and smaller. If you get an error, 
it is likely you have an incomplete file, and you should keep trying 
to download it until you get an error-free file.

2.  Errors occurred if I tried to change the file name during 
download.  So I kept the original file names and truncated them 
later as needed in order to make all file names unique in the first 
8 character positions (for bulk ESEQ conversion using MID2ESEQ).  To 
my pleasant surprise, the resulting ESEQ files were well-labeled for 
display on the Disklavier control unit.

3.  The Bach-Busoni Chaconne, one of Tao Lin's preliminary 
performances, contains 1264 measures of dead air at the beginning of 
the file.  If you play this file in MIDI format, the empty measures 
are ignored.  However if you convert it to ESEQ format uusing 
DKVUTILS MID2ESEQ utility, all of those dead measures are there - 
over 40 minutes of silence before the Chaconne begins to play. On  
the EExplore utility, this 15-minute piece displays as 58 minutes 
long.  I fixed it by using a sequencer to remove the first 1264 
measures ffrom the original MIDI file, then converted the modified 
file to ESEQ.  If anyone else faces the same situation, I will 
gladly forward the modified MIDI file to you. (Or possibly the MIDI 
file has been corrected at the e-competition site by now?)

4.  I downloaded the HTML page for each semifinalist and finalist, 
as a single file (web archive).  I found that I could open the page 
in an HTML editor such as Word, and edit the HREF references so that 
when I click on a link, the downloaded MIDI file plays.  It is a 
matter of removing everyhing in the link definition (if that is the 
right term) except the MIDI file name, and then saving the page 
again.  If that MIDI file resides in the same folder as the saved 
HTML page for that artist, then it will play your downloaded file 
rather than going to the internet to retrieve the file from the e-
competiton web site, or returning an error if you are not connected 
to the internet. Thus you have a playlist for each artist and keep 
it visible when you are playing the files.

That's all I have.  Enjoy!

Mike McGregor
Double Oak, TX