Disklavier

Hi all,
 
I recently had a recording session that used a Disklavier.  We needed 
to record the actual piano sounds as well as get the corresponding 
MIDI.  For the most part, it worked well.  Our resulting audio files 
are great, and the corresponding MIDI almost always matches 
perfectly.  There were a few exceptions though.  In a few of the 
files, the MIDI would cut out for almost exactly 30 seconds, then 
after that silence, it would return, still in sync with the audio.
 
There are a few possible explanations for this.  It's certainly 
possible that the MIDI interface going into the computer was faulty 
in some way (audio and MIDI both recorded into Reaper).  Me feeling 
though is that it was probably something getting stuck on the 
Disklavier's end.  The exact devices were Disklavier II, and the 
controller that spit out the MIDI was a DKC100R.
 
Has anyone else had similar issues?  Any suggestions for how to fix 
this?  We're planning on going back to the same studio, with the same 
setup, in a few days, so we were hoping to have something new to try.
 
I did try searching previous posts for "lost MIDI", but wasn't able 
to find anything pertinent.

Any help would be greatly appreciated.  Thanks very much!

I would think there are 2 possibilities.

1. there could be an intermittent MIDI connection. This could be the cables, the sockets these plug into or the interface.

2. you may have a problem with the key/hammer sensors. If you try to record again, put the piano into record mode as well. If the piano plays the recording back OK, it's your connections/interface, if the playback on the piano is intermittent, you have a problem with your sensors.

athomik

On Aug 26 2008, slothmachine wrote:

>Hi all,
>
>I recently had a recording session that used a Disklavier. We needed
>to record the actual piano sounds as well as get the corresponding
>MIDI. For the most part, it worked well. Our resulting audio files
>are great, and the corresponding MIDI almost always matches
>perfectly. There were a few exceptions though. In a few of the
>files, the MIDI would cut out for almost exactly 30 seconds, then
>after that silence, it would return, still in sync with the audio.
>
>There are a few possible explanations for this. It's certainly
>possible that the MIDI interface going into the computer was faulty
>in some way (audio and MIDI both recorded into Reaper). Me feeling
>though is that it was probably something getting stuck on the
>Disklavier's end. The exact devices were Disklavier II, and the
>controller that spit out the MIDI was a DKC100R.
>
>Has anyone else had similar issues? Any suggestions for how to fix
>this? We're planning on going back to the same studio, with the same
>setup, in a few days, so we were hoping to have something new to try.
>
>I did try searching previous posts for "lost MIDI", but wasn't able
>to find anything pertinent.
>
>Any help would be greatly appreciated. Thanks very much!
>
>

Good morning, everyone.

I recommend looking at the MIDI data carefully on the computer and  
determine with certainty whether that MIDI data is present.

Assuming that the MIDI data is present, I think that it is worth  
determining:

--whether the dropouts always occur at the same places on playback

--determine what happens if you start playback at a location where  
there is a dropout

--look at the MIDI status LEDs on the MIDI interface (if there are  
any) during playback and see if the interface shows MIDI data going to  
the piano during the dropout

How are you synchronizing the audio and the data? Are they both being  
played by the same sequencer?

Regards,
PianoBench

On Aug 25, 2008, at 7:00 PM, slothmachine wrote:

> Hi all,
>
> I recently had a recording session that used a Disklavier. We needed
> to record the actual piano sounds as well as get the corresponding
> MIDI. For the most part, it worked well. Our resulting audio files
> are great, and the corresponding MIDI almost always matches
> perfectly. There were a few exceptions though. In a few of the
> files, the MIDI would cut out for almost exactly 30 seconds, then
> after that silence, it would return, still in sync with the audio.
>
> There are a few possible explanations for this. It's certainly
> possible that the MIDI interface going into the computer was faulty
> in some way (audio and MIDI both recorded into Reaper). Me feeling
> though is that it was probably something getting stuck on the
> Disklavier's end. The exact devices were Disklavier II, and the
> controller that spit out the MIDI was a DKC100R.
>
> Has anyone else had similar issues? Any suggestions for how to fix
> this? We're planning on going back to the same studio, with the same
> setup, in a few days, so we were hoping to have something new to try.
>
> I did try searching previous posts for "lost MIDI", but wasn't able
> to find anything pertinent.
>
> Any help would be greatly appreciated. Thanks very much!

Athomik & George,

Thank you very much for replying with such good suggestions.  
Athomik, I think your suggestion is best to actually record on the 
piano itself, in addition to sending it through the MIDI outputs.  I 
think the reason we didn't do that the first time is that those 
floppy disks don't hold that much information, so we would have ended 
up continually replacing them.  We could be wrong though about how 
much info they hold.  About how much time does one of those disks 
hold?  (I know that with MIDI it's not time that takes the space, but 
the number of notes played, but just assuming that he's playing some 
normal piano music, how much time...)

George, to answer your questions, the dropouts don't always occur in 
the same place, but they seem to almost always last for about 33 
seconds.  This is kind of what made me rule out faulty cables, 
because it would just be too much of a coincidence that cables would 
fail, several times, each lasting 30 seconds.  The behaviour would 
seem to me that it's some device along the way, whether it's the 
piano sending to the DKC100R, the DKC100R sending to the MIDI 
interface into the computer, or the MIDI interface into the computer, 
that somehow gets overloaded with info and takes 30 seconds 
to "restart" itself at which point it continues to spit out MIDI info.

The MIDI and audio both get recorded simultaneously into the 
sequencer REAPER, and yes, we're certain that for those dropouts 
there is no MIDI at all.  You just have to look at an event list to 
see that there's nothing there.

Also George, as per your suggestion, I will instruct the engineer to 
pay close attention to the flashing lights on the MIDI interface 
while he records to make sure that that is being recorded correctly. 

Thanks again for your help, I now have some very useful information 
for when we go into the studio next.

Sincerely,
Tobin



--- In disklavier@yahoogroups.com, "George F. Litterst" 
<PianoBench@...> wrote:
>
> Good morning, everyone.
> 
> I recommend looking at the MIDI data carefully on the computer and  
> determine with certainty whether that MIDI data is present.
> 
> Assuming that the MIDI data is present, I think that it is worth  
> determining:
> 
> --whether the dropouts always occur at the same places on playback
> 
> --determine what happens if you start playback at a location where  
> there is a dropout
> 
> --look at the MIDI status LEDs on the MIDI interface (if there are  
> any) during playback and see if the interface shows MIDI data going 
to  
> the piano during the dropout
> 
> How are you synchronizing the audio and the data? Are they both 
being  
> played by the same sequencer?
> 
> Regards,
> PianoBench
> 
> On Aug 25, 2008, at 7:00 PM, slothmachine wrote:
> 
> > Hi all,
> >
> > I recently had a recording session that used a Disklavier. We 
needed
> > to record the actual piano sounds as well as get the corresponding
> > MIDI. For the most part, it worked well. Our resulting audio files
> > are great, and the corresponding MIDI almost always matches
> > perfectly. There were a few exceptions though. In a few of the
> > files, the MIDI would cut out for almost exactly 30 seconds, then
> > after that silence, it would return, still in sync with the audio.
> >
> > There are a few possible explanations for this. It's certainly
> > possible that the MIDI interface going into the computer was 
faulty
> > in some way (audio and MIDI both recorded into Reaper). Me feeling
> > though is that it was probably something getting stuck on the
> > Disklavier's end. The exact devices were Disklavier II, and the
> > controller that spit out the MIDI was a DKC100R.
> >
> > Has anyone else had similar issues? Any suggestions for how to fix
> > this? We're planning on going back to the same studio, with the 
same
> > setup, in a few days, so we were hoping to have something new to 
try.
> >
> > I did try searching previous posts for "lost MIDI", but wasn't 
able

> > to find anything pertinent.
> >
> > Any help would be greatly appreciated. Thanks very much!
>

Greetings slothmachine,

a couple of other possibilities to consider.

i have had systematic (seem random at first glance but have some sort of pattern) failure of MIDI interfaces occasionally. the MIDI spec says that a receiver of MIDI data should pass on anything that it can not process and get on with the rest. however it seems (no real proof of this but it has happened enough to suggest so) that some devices can choke on something they do not process properly, a sysex message possibly. they may recover or may need to be rebooted. for this reason, i have a few different interfaces (different brands and models) and when one just doesn't seems to work for some bizarre reason that i can not figure out, i just try another one. of course it could be something in the DKV firmware that does not come into play often enough to have been identified by other users. cyclical failures can be problems with data buffer management. if there is some strange bug in the firmware, changing a mode or setting might allow the firmware to operate differently and not hit the bug. try an update if there is one,
-- 
Best regards,
Spencer_Lists Chase        mailto:lists@...
67550 Bell Springs Rd.
Garberville,  CA 95542    Postal service only.
Laytonville, CA 95454    UPS only.
Lists@spencerserolls(dot)com
http://www(dot)spencerserolls.com 
replace (dot) with a .
(707) 984-8356

Good afternoon, everyone.

I am a little confused, Tobin. On the one hand you indicated that the  
dropouts on playback do not always occur in the same places each time  
you play back a file. That tells me that all of the MIDI data is  
present in the file. However, you also said that you checked the file  
and saw no MIDI data in the event list where dropouts occur.

The first and most important thing is to figure out if data is being  
continuously recorded during the performance. The event list should  
confirm this. If the data for the entire piece is present, you should  
be able to listen to the entire piece without dropouts by directing  
the MIDI output of the sequencer to an internal tone generator on your  
computer (QuickTime GM Synth on the Mac or Microsoft GS Wavetable SW  
Synth on Windows). Playback by those 2 soft synths may be muddy  
(because they don't understand incremental pedaling), but you will be  
able to confirm aurally that all of the notes are there.

Assuming that all of the notes are there, there should be no droupouts  
on playback by the piano. If there are, the most likely culprits are:

--sequencer
--the MIDI cables or interface
--the Disklavier

In the case of the Disklavier, overheating can cause playback  
dropouts. However, consistent overheating usually results in an error  
message on the control unit.

If you choose to record to high density floppy disks (formatted as  
HD), you should be able to get 2 hours of normal piano playing on a  
floppy (assuming that the piano is not a version called the Pro).

Regards,
PianoBench

On Aug 26, 2008, at 11:26 AM, slothmachine wrote:

> Athomik & George,
>
> Thank you very much for replying with such good suggestions.
> Athomik, I think your suggestion is best to actually record on the
> piano itself, in addition to sending it through the MIDI outputs. I
> think the reason we didn't do that the first time is that those
> floppy disks don't hold that much information, so we would have ended
> up continually replacing them. We could be wrong though about how
> much info they hold. About how much time does one of those disks
> hold? (I know that with MIDI it's not time that takes the space, but
> the number of notes played, but just assuming that he's playing some
> normal piano music, how much time...)
>
> George, to answer your questions, the dropouts don't always occur in
> the same place, but they seem to almost always last for about 33
> seconds. This is kind of what made me rule out faulty cables,
> because it would just be too much of a coincidence that cables would
> fail, several times, each lasting 30 seconds. The behaviour would
> seem to me that it's some device along the way, whether it's the
> piano sending to the DKC100R, the DKC100R sending to the MIDI
> interface into the computer, or the MIDI interface into the computer,
> that somehow gets overloaded with info and takes 30 seconds
> to "restart" itself at which point it continues to spit out MIDI info.
>
> The MIDI and audio both get recorded simultaneously into the
> sequencer REAPER, and yes, we're certain that for those dropouts
> there is no MIDI at all. You just have to look at an event list to
> see that there's nothing there.
>
> Also George, as per your suggestion, I will instruct the engineer to
> pay close attention to the flashing lights on the MIDI interface
> while he records to make sure that that is being recorded correctly.
>
> Thanks again for your help, I now have some very useful information
> for when we go into the studio next.
>
> Sincerely,
> Tobin
>
> --- In disklavier@yahoogroups.com, "George F. Litterst"
> <PianoBench@...> wrote:
> >
> > Good morning, everyone.
> >
> > I recommend looking at the MIDI data carefully on the computer and
> > determine with certainty whether that MIDI data is present.
> >
> > Assuming that the MIDI data is present, I think that it is worth
> > determining:
> >
> > --whether the dropouts always occur at the same places on playback
> >
> > --determine what happens if you start playback at a location where
> > there is a dropout
> >
> > --look at the MIDI status LEDs on the MIDI interface (if there are
> > any) during playback and see if the interface shows MIDI data going
> to
> > the piano during the dropout
> >
> > How are you synchronizing the audio and the data? Are they both
> being
> > played by the same sequencer?
> >
> > Regards,
> > PianoBench
> >
> > On Aug 25, 2008, at 7:00 PM, slothmachine wrote:
> >
> > > Hi all,
> > >
> > > I recently had a recording session that used a Disklavier. We
> needed
> > > to record the actual piano sounds as well as get the corresponding
> > > MIDI. For the most part, it worked well. Our resulting audio files
> > > are great, and the corresponding MIDI almost always matches
> > > perfectly. There were a few exceptions though. In a few of the
> > > files, the MIDI would cut out for almost exactly 30 seconds, then
> > > after that silence, it would return, still in sync with the audio.
> > >
> > > There are a few possible explanations for this. It's certainly
> > > possible that the MIDI interface going into the computer was
> faulty
> > > in some way (audio and MIDI both recorded into Reaper). Me feeling
> > > though is that it was probably something getting stuck on the
> > > Disklavier's end. The exact devices were Disklavier II, and the
> > > controller that spit out the MIDI was a DKC100R.
> > >
> > > Has anyone else had similar issues? Any suggestions for how to fix
> > > this? We're planning on going back to the same studio, with the
> same
> > > setup, in a few days, so we were hoping to have something new to
> try.
> > >
> > > I did try searching previous posts for "lost MIDI", but wasn't
> able
> > > to find anything pertinent.
> > >
> > > Any help would be greatly appreciated. Thanks very much!

Hi George,

A little confusion there:  We recorded about 30 separate files.  The 
MIDI dropout occurs in about 15% of those files at different places 
in each file (in other words, in one file it may occur 15 seconds 
into the file, in another file it may occur 5 minutes into the file) 
but in every case, the dropout lasts for 30 seconds.  In any one 
given file where the dropout occurs, however, yes, obviously the 
dropout occurs in the same place all the time.  MIDI data is 
definitely not present during those dropouts.

That's excellent news that a floppy should be able to hold about 2 
hours.  I think we're going to use that as a backup.  If we find that 
we're experiencing dropouts again, we'll immediately check the floppy 
to see if it also occurred on that.

Again, thanks for your help.

Tobin




--- In disklavier@yahoogroups.com, "George F. Litterst" 
<PianoBench@...> wrote:
>
> Good afternoon, everyone.
> 
> I am a little confused, Tobin. On the one hand you indicated that 
the  
> dropouts on playback do not always occur in the same places each 
time  
> you play back a file. That tells me that all of the MIDI data is  
> present in the file. However, you also said that you checked the 
file  
> and saw no MIDI data in the event list where dropouts occur.
> 
> The first and most important thing is to figure out if data is 
being  
> continuously recorded during the performance. The event list 
should  
> confirm this. If the data for the entire piece is present, you 
should  
> be able to listen to the entire piece without dropouts by 
directing  
> the MIDI output of the sequencer to an internal tone generator on 
your  
> computer (QuickTime GM Synth on the Mac or Microsoft GS Wavetable 
SW  
> Synth on Windows). Playback by those 2 soft synths may be muddy  
> (because they don't understand incremental pedaling), but you will 
be  
> able to confirm aurally that all of the notes are there.
> 
> Assuming that all of the notes are there, there should be no 
droupouts  
> on playback by the piano. If there are, the most likely culprits 
are:
> 
> --sequencer
> --the MIDI cables or interface
> --the Disklavier
> 
> In the case of the Disklavier, overheating can cause playback  
> dropouts. However, consistent overheating usually results in an 
error  
> message on the control unit.
> 
> If you choose to record to high density floppy disks (formatted as  
> HD), you should be able to get 2 hours of normal piano playing on 
a  
> floppy (assuming that the piano is not a version called the Pro).
> 
> Regards,
> PianoBench
> 
> 
> On Aug 26, 2008, at 11:26 AM, slothmachine wrote:
> 
> > Athomik & George,
> >
> > Thank you very much for replying with such good suggestions.
> > Athomik, I think your suggestion is best to actually record on the
> > piano itself, in addition to sending it through the MIDI outputs. 
I
> > think the reason we didn't do that the first time is that those
> > floppy disks don't hold that much information, so we would have 
ended
> > up continually replacing them. We could be wrong though about how
> > much info they hold. About how much time does one of those disks
> > hold? (I know that with MIDI it's not time that takes the space, 
but
> > the number of notes played, but just assuming that he's playing 
some
> > normal piano music, how much time...)
> >
> > George, to answer your questions, the dropouts don't always occur 
in
> > the same place, but they seem to almost always last for about 33
> > seconds. This is kind of what made me rule out faulty cables,
> > because it would just be too much of a coincidence that cables 
would
> > fail, several times, each lasting 30 seconds. The behaviour would
> > seem to me that it's some device along the way, whether it's the
> > piano sending to the DKC100R, the DKC100R sending to the MIDI
> > interface into the computer, or the MIDI interface into the 
computer,
> > that somehow gets overloaded with info and takes 30 seconds
> > to "restart" itself at which point it continues to spit out MIDI 
info.
> >
> > The MIDI and audio both get recorded simultaneously into the
> > sequencer REAPER, and yes, we're certain that for those dropouts
> > there is no MIDI at all. You just have to look at an event list to
> > see that there's nothing there.
> >
> > Also George, as per your suggestion, I will instruct the engineer 
to
> > pay close attention to the flashing lights on the MIDI interface
> > while he records to make sure that that is being recorded 
correctly.
> >
> > Thanks again for your help, I now have some very useful 
information
> > for when we go into the studio next.
> >
> > Sincerely,
> > Tobin
> >
> > --- In disklavier@yahoogroups.com, "George F. Litterst"
> > <PianoBench@> wrote:
> > >
> > > Good morning, everyone.
> > >
> > > I recommend looking at the MIDI data carefully on the computer 
and
> > > determine with certainty whether that MIDI data is present.
> > >
> > > Assuming that the MIDI data is present, I think that it is worth
> > > determining:
> > >
> > > --whether the dropouts always occur at the same places on 
playback
> > >
> > > --determine what happens if you start playback at a location 
where
> > > there is a dropout
> > >
> > > --look at the MIDI status LEDs on the MIDI interface (if there 
are
> > > any) during playback and see if the interface shows MIDI data 
going
> > to
> > > the piano during the dropout
> > >
> > > How are you synchronizing the audio and the data? Are they both
> > being
> > > played by the same sequencer?
> > >
> > > Regards,
> > > PianoBench
> > >
> > > On Aug 25, 2008, at 7:00 PM, slothmachine wrote:
> > >
> > > > Hi all,
> > > >
> > > > I recently had a recording session that used a Disklavier. We
> > needed
> > > > to record the actual piano sounds as well as get the 
corresponding
> > > > MIDI. For the most part, it worked well. Our resulting audio 
files
> > > > are great, and the corresponding MIDI almost always matches
> > > > perfectly. There were a few exceptions though. In a few of the
> > > > files, the MIDI would cut out for almost exactly 30 seconds, 
then
> > > > after that silence, it would return, still in sync with the 
audio.
> > > >
> > > > There are a few possible explanations for this. It's certainly
> > > > possible that the MIDI interface going into the computer was
> > faulty
> > > > in some way (audio and MIDI both recorded into Reaper). Me 
feeling
> > > > though is that it was probably something getting stuck on the
> > > > Disklavier's end. The exact devices were Disklavier II, and 
the
> > > > controller that spit out the MIDI was a DKC100R.
> > > >
> > > > Has anyone else had similar issues? Any suggestions for how 
to fix
> > > > this? We're planning on going back to the same studio, with 
the
> > same
> > > > setup, in a few days, so we were hoping to have something new 
to

> > try.
> > > >
> > > > I did try searching previous posts for "lost MIDI", but wasn't
> > able
> > > > to find anything pertinent.
> > > >
> > > > Any help would be greatly appreciated. Thanks very much!
>

Good afternoon, everyone.

Tobin, just checking: has everything been working correctly?

Regards,
PianoBench

On Aug 26, 2008, at 5:02 PM, slothmachine wrote:

> Hi George,
>
> A little confusion there: We recorded about 30 separate files. The
> MIDI dropout occurs in about 15% of those files at different places
> in each file (in other words, in one file it may occur 15 seconds
> into the file, in another file it may occur 5 minutes into the file)
> but in every case, the dropout lasts for 30 seconds. In any one
> given file where the dropout occurs, however, yes, obviously the
> dropout occurs in the same place all the time. MIDI data is
> definitely not present during those dropouts.
>
> That's excellent news that a floppy should be able to hold about 2
> hours. I think we're going to use that as a backup. If we find that
> we're experiencing dropouts again, we'll immediately check the floppy
> to see if it also occurred on that.
>
> Again, thanks for your help.
>
> Tobin
>
> --- In disklavier@yahoogroups.com, "George F. Litterst"
> <PianoBench@...> wrote:
> >
> > Good afternoon, everyone.
> >
> > I am a little confused, Tobin. On the one hand you indicated that
> the
> > dropouts on playback do not always occur in the same places each
> time
> > you play back a file. That tells me that all of the MIDI data is
> > present in the file. However, you also said that you checked the
> file
> > and saw no MIDI data in the event list where dropouts occur.
> >
> > The first and most important thing is to figure out if data is
> being
> > continuously recorded during the performance. The event list
> should
> > confirm this. If the data for the entire piece is present, you
> should
> > be able to listen to the entire piece without dropouts by
> directing
> > the MIDI output of the sequencer to an internal tone generator on
> your
> > computer (QuickTime GM Synth on the Mac or Microsoft GS Wavetable
> SW
> > Synth on Windows). Playback by those 2 soft synths may be muddy
> > (because they don't understand incremental pedaling), but you will
> be
> > able to confirm aurally that all of the notes are there.
> >
> > Assuming that all of the notes are there, there should be no
> droupouts
> > on playback by the piano. If there are, the most likely culprits
> are:
> >
> > --sequencer
> > --the MIDI cables or interface
> > --the Disklavier
> >
> > In the case of the Disklavier, overheating can cause playback
> > dropouts. However, consistent overheating usually results in an
> error
> > message on the control unit.
> >
> > If you choose to record to high density floppy disks (formatted as
> > HD), you should be able to get 2 hours of normal piano playing on
> a
> > floppy (assuming that the piano is not a version called the Pro).
> >
> > Regards,
> > PianoBench
> >
> >
> > On Aug 26, 2008, at 11:26 AM, slothmachine wrote:
> >
> > > Athomik & George,
> > >
> > > Thank you very much for replying with such good suggestions.
> > > Athomik, I think your suggestion is best to actually record on the
> > > piano itself, in addition to sending it through the MIDI outputs.
> I
> > > think the reason we didn't do that the first time is that those
> > > floppy disks don't hold that much information, so we would have
> ended
> > > up continually replacing them. We could be wrong though about how
> > > much info they hold. About how much time does one of those disks
> > > hold? (I know that with MIDI it's not time that takes the space,
> but
> > > the number of notes played, but just assuming that he's playing
> some
> > > normal piano music, how much time...)
> > >
> > > George, to answer your questions, the dropouts don't always occur
> in
> > > the same place, but they seem to almost always last for about 33
> > > seconds. This is kind of what made me rule out faulty cables,
> > > because it would just be too much of a coincidence that cables
> would
> > > fail, several times, each lasting 30 seconds. The behaviour would
> > > seem to me that it's some device along the way, whether it's the
> > > piano sending to the DKC100R, the DKC100R sending to the MIDI
> > > interface into the computer, or the MIDI interface into the
> computer,
> > > that somehow gets overloaded with info and takes 30 seconds
> > > to "restart" itself at which point it continues to spit out MIDI
> info.
> > >
> > > The MIDI and audio both get recorded simultaneously into the
> > > sequencer REAPER, and yes, we're certain that for those dropouts
> > > there is no MIDI at all. You just have to look at an event list to
> > > see that there's nothing there.
> > >
> > > Also George, as per your suggestion, I will instruct the engineer
> to
> > > pay close attention to the flashing lights on the MIDI interface
> > > while he records to make sure that that is being recorded
> correctly.
> > >
> > > Thanks again for your help, I now have some very useful
> information
> > > for when we go into the studio next.
> > >
> > > Sincerely,
> > > Tobin
> > >
> > > --- In disklavier@yahoogroups.com, "George F. Litterst"
> > > <PianoBench@> wrote:
> > > >
> > > > Good morning, everyone.
> > > >
> > > > I recommend looking at the MIDI data carefully on the computer
> and
> > > > determine with certainty whether that MIDI data is present.
> > > >
> > > > Assuming that the MIDI data is present, I think that it is worth
> > > > determining:
> > > >
> > > > --whether the dropouts always occur at the same places on
> playback
> > > >
> > > > --determine what happens if you start playback at a location
> where
> > > > there is a dropout
> > > >
> > > > --look at the MIDI status LEDs on the MIDI interface (if there
> are
> > > > any) during playback and see if the interface shows MIDI data
> going
> > > to
> > > > the piano during the dropout
> > > >
> > > > How are you synchronizing the audio and the data? Are they both
> > > being
> > > > played by the same sequencer?
> > > >
> > > > Regards,
> > > > PianoBench
> > > >
> > > > On Aug 25, 2008, at 7:00 PM, slothmachine wrote:
> > > >
> > > > > Hi all,
> > > > >
> > > > > I recently had a recording session that used a Disklavier. We
> > > needed
> > > > > to record the actual piano sounds as well as get the
> corresponding
> > > > > MIDI. For the most part, it worked well. Our resulting audio
> files
> > > > > are great, and the corresponding MIDI almost always matches
> > > > > perfectly. There were a few exceptions though. In a few of the
> > > > > files, the MIDI would cut out for almost exactly 30 seconds,
> then
> > > > > after that silence, it would return, still in sync with the
> audio.
> > > > >
> > > > > There are a few possible explanations for this. It's certainly
> > > > > possible that the MIDI interface going into the computer was
> > > faulty
> > > > > in some way (audio and MIDI both recorded into Reaper). Me
> feeling
> > > > > though is that it was probably something getting stuck on the
> > > > > Disklavier's end. The exact devices were Disklavier II, and
> the
> > > > > controller that spit out the MIDI was a DKC100R.
> > > > >
> > > > > Has anyone else had similar issues? Any suggestions for how
> to fix
> > > > > this? We're planning on going back to the same studio, with
> the
> > > same
> > > > > setup, in a few days, so we were hoping to have something new
> to
> > > try.
> > > > >
> > > > > I did try searching previous posts for "lost MIDI", but wasn't
> > > able
> > > > > to find anything pertinent.
> > > > >
> > > > > Any help would be greatly appreciated. Thanks very much!

Hi George,

Yes, we recorded again, and this time the problem went away for the 
most part.  We didn't really do anything different, but it just 
seemed to be behaving itself this time.  In about 3 hours of recorded 
material, the MIDI only dropped out once, and I can live with that...

Thanks again for everyone's help with this issue.

Tobin



--- In disklavier@yahoogroups.com, "George F. Litterst" 
<PianoBench@...> wrote:
>
> Good afternoon, everyone.
> 
> Tobin, just checking: has everything been working correctly?
> 
> Regards,
> PianoBench
> 
> 
> On Aug 26, 2008, at 5:02 PM, slothmachine wrote:
> 
> > Hi George,
> >
> > A little confusion there: We recorded about 30 separate files. The
> > MIDI dropout occurs in about 15% of those files at different 
places
> > in each file (in other words, in one file it may occur 15 seconds
> > into the file, in another file it may occur 5 minutes into the 
file)
> > but in every case, the dropout lasts for 30 seconds. In any one
> > given file where the dropout occurs, however, yes, obviously the
> > dropout occurs in the same place all the time. MIDI data is
> > definitely not present during those dropouts.
> >
> > That's excellent news that a floppy should be able to hold about 2
> > hours. I think we're going to use that as a backup. If we find 
that
> > we're experiencing dropouts again, we'll immediately check the 
floppy
> > to see if it also occurred on that.
> >
> > Again, thanks for your help.
> >
> > Tobin
> >
> > --- In disklavier@yahoogroups.com, "George F. Litterst"
> > <PianoBench@> wrote:
> > >
> > > Good afternoon, everyone.
> > >
> > > I am a little confused, Tobin. On the one hand you indicated 
that
> > the
> > > dropouts on playback do not always occur in the same places each
> > time
> > > you play back a file. That tells me that all of the MIDI data is
> > > present in the file. However, you also said that you checked the
> > file
> > > and saw no MIDI data in the event list where dropouts occur.
> > >
> > > The first and most important thing is to figure out if data is
> > being
> > > continuously recorded during the performance. The event list
> > should
> > > confirm this. If the data for the entire piece is present, you
> > should
> > > be able to listen to the entire piece without dropouts by
> > directing
> > > the MIDI output of the sequencer to an internal tone generator 
on
> > your
> > > computer (QuickTime GM Synth on the Mac or Microsoft GS 
Wavetable
> > SW
> > > Synth on Windows). Playback by those 2 soft synths may be muddy
> > > (because they don't understand incremental pedaling), but you 
will
> > be
> > > able to confirm aurally that all of the notes are there.
> > >
> > > Assuming that all of the notes are there, there should be no
> > droupouts
> > > on playback by the piano. If there are, the most likely culprits
> > are:
> > >
> > > --sequencer
> > > --the MIDI cables or interface
> > > --the Disklavier
> > >
> > > In the case of the Disklavier, overheating can cause playback
> > > dropouts. However, consistent overheating usually results in an
> > error
> > > message on the control unit.
> > >
> > > If you choose to record to high density floppy disks (formatted 
as
> > > HD), you should be able to get 2 hours of normal piano playing 
on
> > a
> > > floppy (assuming that the piano is not a version called the 
Pro).
> > >
> > > Regards,
> > > PianoBench
> > >
> > >
> > > On Aug 26, 2008, at 11:26 AM, slothmachine wrote:
> > >
> > > > Athomik & George,
> > > >
> > > > Thank you very much for replying with such good suggestions.
> > > > Athomik, I think your suggestion is best to actually record 
on the
> > > > piano itself, in addition to sending it through the MIDI 
outputs.
> > I
> > > > think the reason we didn't do that the first time is that 
those
> > > > floppy disks don't hold that much information, so we would 
have
> > ended
> > > > up continually replacing them. We could be wrong though about 
how
> > > > much info they hold. About how much time does one of those 
disks
> > > > hold? (I know that with MIDI it's not time that takes the 
space,
> > but
> > > > the number of notes played, but just assuming that he's 
playing
> > some
> > > > normal piano music, how much time...)
> > > >
> > > > George, to answer your questions, the dropouts don't always 
occur
> > in
> > > > the same place, but they seem to almost always last for about 
33
> > > > seconds. This is kind of what made me rule out faulty cables,
> > > > because it would just be too much of a coincidence that cables
> > would
> > > > fail, several times, each lasting 30 seconds. The behaviour 
would
> > > > seem to me that it's some device along the way, whether it's 
the
> > > > piano sending to the DKC100R, the DKC100R sending to the MIDI
> > > > interface into the computer, or the MIDI interface into the
> > computer,
> > > > that somehow gets overloaded with info and takes 30 seconds
> > > > to "restart" itself at which point it continues to spit out 
MIDI
> > info.
> > > >
> > > > The MIDI and audio both get recorded simultaneously into the
> > > > sequencer REAPER, and yes, we're certain that for those 
dropouts
> > > > there is no MIDI at all. You just have to look at an event 
list to
> > > > see that there's nothing there.
> > > >
> > > > Also George, as per your suggestion, I will instruct the 
engineer
> > to
> > > > pay close attention to the flashing lights on the MIDI 
interface
> > > > while he records to make sure that that is being recorded
> > correctly.
> > > >
> > > > Thanks again for your help, I now have some very useful
> > information
> > > > for when we go into the studio next.
> > > >
> > > > Sincerely,
> > > > Tobin
> > > >
> > > > --- In disklavier@yahoogroups.com, "George F. Litterst"
> > > > <PianoBench@> wrote:
> > > > >
> > > > > Good morning, everyone.
> > > > >
> > > > > I recommend looking at the MIDI data carefully on the 
computer
> > and
> > > > > determine with certainty whether that MIDI data is present.
> > > > >
> > > > > Assuming that the MIDI data is present, I think that it is 
worth
> > > > > determining:
> > > > >
> > > > > --whether the dropouts always occur at the same places on
> > playback
> > > > >
> > > > > --determine what happens if you start playback at a location
> > where
> > > > > there is a dropout
> > > > >
> > > > > --look at the MIDI status LEDs on the MIDI interface (if 
there
> > are
> > > > > any) during playback and see if the interface shows MIDI 
data
> > going
> > > > to
> > > > > the piano during the dropout
> > > > >
> > > > > How are you synchronizing the audio and the data? Are they 
both
> > > > being
> > > > > played by the same sequencer?
> > > > >
> > > > > Regards,
> > > > > PianoBench
> > > > >
> > > > > On Aug 25, 2008, at 7:00 PM, slothmachine wrote:
> > > > >
> > > > > > Hi all,
> > > > > >
> > > > > > I recently had a recording session that used a 
Disklavier. We
> > > > needed
> > > > > > to record the actual piano sounds as well as get the
> > corresponding
> > > > > > MIDI. For the most part, it worked well. Our resulting 
audio
> > files
> > > > > > are great, and the corresponding MIDI almost always 
matches
> > > > > > perfectly. There were a few exceptions though. In a few 
of the
> > > > > > files, the MIDI would cut out for almost exactly 30 
seconds,
> > then
> > > > > > after that silence, it would return, still in sync with 
the
> > audio.
> > > > > >
> > > > > > There are a few possible explanations for this. It's 
certainly
> > > > > > possible that the MIDI interface going into the computer 
was
> > > > faulty
> > > > > > in some way (audio and MIDI both recorded into Reaper). Me
> > feeling
> > > > > > though is that it was probably something getting stuck on 
the
> > > > > > Disklavier's end. The exact devices were Disklavier II, 
and
> > the
> > > > > > controller that spit out the MIDI was a DKC100R.
> > > > > >
> > > > > > Has anyone else had similar issues? Any suggestions for 
how
> > to fix
> > > > > > this? We're planning on going back to the same studio, 
with
> > the
> > > > same
> > > > > > setup, in a few days, so we were hoping to have something 
new
> > to
> > > > try.
> > > > > >
> > > > > > I did try searching previous posts for "lost MIDI", but 
wasn't

> > > > able
> > > > > > to find anything pertinent.
> > > > > >
> > > > > > Any help would be greatly appreciated. Thanks very much!
>