Yahoo Groups archive

Disklavier

Index last updated: 2026-04-28 23:20 UTC

Message

Re: [disklavier] mx100a problems and questions/suggestions

2014-03-14 by Spencer Chase

good point about the virtual synth latency. all would have some but good ones(with a good sound card) should have minimal. there is a simple test (that i forget but i think you can find it on audacity forums etc) to measure latency of any audio or midi. it involves making a click track and ???

i think the test was mainly to quantify the latency of audacity itself in playback and record so if you are using audacity to record the tests, you may have to considers audacitie's latency itself too :) brain is not working well enough at this moment to figure it out.

on a computer, the latency of a virtual synth will vary depending on sound card properties so you can not assume anything but need to measure it.

On 3/14/2014 11:32 AM, George Frederick Litterst wrote:
\ufffd

Good afternoon, everyone.


Sam, if the piano is behind the bass and drums (which come from a virtual synth), then there is no extra latency in the virtual synth.

I suggest doing a simple test. Create a sequence where the piano plays a few simple chords and you have some synchronized drum sounds to be played by the virtual synth. Make absolutely certain that all MIDI notes in both tracks are quantized to the beat and then try playback again using the configuration that I suggested.

In my experience, that 500ms delay in the Disklavier is consistent. I am wondering if any of the notes of your sequence don't align to the beat. Using the configuration that I recommended, there should be no need to shift any tracks in your sequencer.

If the piano and drums were to be out of sync using the configuration that I have suggested, I would guess that the drums would be late, not the piano, because some virtual synths have latency.

Regards,
PianoBench

On Mar 14, 2014, at 11:03 AM, Skanter123 wrote:

\ufffd

1) In your computer sequence, do not offset the virtual synth tracks by 500ms.

(2) Set up your virtual synth to receive MIDI data from an external source.

(3) On your Mark II, set the PIANO RECEIVE channel to something that matches the piano tracks in your sequencer. (The simplest solution is to set the Piano to receive on channel 1 and set the piano tracks in the sequencer to send on channel 1.)

(4) On your Mark II, set MIDI OUT to DELAY OUT.

(5) On your Mark II, in the PIANO PART are under the ENSEMBLE heading, set PIANO PART MIDI OUT=OFF.

In this configuration, the Disklavier will receive data from your computer and do two things:

(a) Play the piano tracks on the piano.

(b) Delay the other tracks by 500ms and then send them to your virtual synth.

Be sure that your sequencer is not set to echo back any MIDI data that it receives.


George, i tried your configuration, but things still did not sync up. Piano was behind bass and drums, maybe less than 500ms.

Perhaps there is some latency in the virtual synth on my computer?

Sam\ufffd

On Mar 12, 2014, at 8:24 AM, George Frederick Litterst <PianoBench@...> wrote:

\ufffd

Good morning, everyone.


Sam, as I understand it, the Mark II-to-Mark IIXG upgrade kit included a new control unit of a different size than the original, a couple of circuit boards, and a new connecting cable. I don't recall the price, but I would not be surprised if it was similarly priced to the DKC-850 because it included a new control unit. In the case of a Mark II upright, it also included a replacement piece of metal for that area of the piano that holds the built-in Mark II controller.

I don't believe that these complete kits are still available in the US. However, I have heard that the circuit boards may still be available, thus enabling an upgrade to a Mark IIXG minus the control unit. At that stage, the piano could receive the DKC-850 as a replacement control unit.

As for your playback issue: If I understand it correctly, you want to play multitrack sequences from your computer \ufffdan d enjoy playback that includes the Disklavier for the piano track(s) and a virtual synth (running on the computer) for the other tracks. Correct? If so, try the following:

(1) In your computer sequence, do not offset the virtual synth tracks by 500ms.

(2) Set up your virtual synth to receive MIDI data from an external source.

(3) On your Mark II, set the PIANO RECEIVE channel to something that matches the piano tracks in your sequencer. (The simplest solution is to set the Piano to receive on channel 1 and set the piano tracks in the sequencer to send on channel 1.)

(4) On your Mark II, set MIDI OUT to DELAY OUT.

(5) On your Mark II, in the PIANO PART are under the ENSEMBLE heading, set PIANO PART MIDI OUT=OFF.

In this configuration, the Disklavier will receive data from your computer and do two things:

(a) Play the piano tracks on the piano.

(b) Delay the other tracks by 500ms and then send them to your virtual synth.

Be sure that your sequencer is not set to echo back any MIDI data that it receives.

Regards,
PianoBench




On Mar 11, 2014, at 8:48 PM, Sam Kanter wrote:

\ufffd

George, thanks for the excellent article on the history of the disklavier - very informative. I recommend that everyone hear reads it - much better than the Wikipedia article.

I was particularly interested to the Mark II to Mark II XG upgrade you mentioned. I wonder if there is still any possibility of that being available, and significantly less expensive than the DKC-850? At this point I cannot see a justification for upgrading to the DKC-850 with the limited improvement it would provide with the Mk II DKV I have, especially since I use it with a computer, not floppy. In general my Mark II unit performs beautifully, but always a few bugs and glitches to iron out. Still trying to sync MIDI playback with a virtual synth in my computer, difficult as I have to delay each MID sequence by 500ms. I do have a Yamaha MU-50, might be easier setting that up.

I will have to do some testing to see how much compression my piano is recording. I suppose a good test would be doing a record playback test with a MIDI monitor. I assume I could fix some of the compression using one of Spence's volume utilities?

Thanks for your help...

Sam Kanter
www.keyboardcollective.com
(212) 684-3304


On Tue, Mar 11, 2014 at 4:40 PM, George Frederick Litterst <PianoBench@...> wrote:
\ufffd

Good afternoon, everyone.


Sam, the Mark II was a big step up from the earlier instruments. However, there is some compression of the dynamic range on playback. You can easily do a recording and compare the playback.

The first Disklavier that captured a huge range was the Mark IIXG PRO.

For anyone who is interested in the history of the Disklavier, you might enjoy this article that I wrote for the Disklavier Education Network:


For those who are uncertain as to which model they have, you can figure it out from this chart that I created:


For those interested in upgrading to a DKC-850, there is another chart here:


Regards,
PianoBench


On Mar 11, 2014, at 2:06 PM, Sam Kanter wrote:

\ufffd

George, can I assume my Mark II does not have the compressed volume issues of the earlier models?

Sam

*(Sent from phone - please excuse brevity and typos.)

On Mar 11, 2014, at 12:39 PM, George Frederick Litterst <PianoBench@...> wrote:

\ufffd

Good afternoon, everyone.


Steno Jazz, here is a more complete explanation of several distinct issues that affect your MX100A (and similarly, other early model Disklaviers):

MIDI Out
The choices are Keyboard Out and Delay Out:

Keyboard Out sends out MIDI data that you play on the keys yourself. The Delay Out setting causes MIDI data that is being played from floppy disk to be sent out.

MIDI In
There are two MIDI In settings: Delay In and Real Time. You should almost always use Delay In.

The basic issue is this: When you send MIDI data over a MIDI cable (such as from a computer to your piano), you expect the tone generator that received the data to respond immediately. In the case of a digital piano, you'll hear the MIDI data instantly.

The Disklavier, however, is a mechanical (Message over 64 KB, truncated)

Attachments

Move to quarantaine

This moves the raw source file on disk only. The archive index is not changed automatically, so you still need to run a manual refresh afterward.