Pedal Problem

[pandrews]

I'm using the new Machfive with DP and a Yamaha P-120 controler. Uploaded the new German Grand Piano. Started playing a realized that the sustain pedal keeps all the notes playing even when I let go of the pedal. I have to play the sustained notes again to shut them off.

I tried using the pedal with all the plug-ins I have and they all work fine with the pedal except for Plugsound Pro and Machfive.

I tried using different pedals, but got the same results.

Is there a fix for this? Has anyone else had this problem?

Thanks.

[Dwetmaster]

the sustain might be responding to another ctrl or on inverted polarity. Is there a setting fo it?

[pandrews]

I recorded the pedal and saw that there was no other information present. I tested this with the quicktime synth and I was able to use the pedal with no problem. MX4 worked fine too. All the DP instruments worked fine too.

I'm thinking Machfive reads pedals as all the way "on" and "off". the P-120 does have "on" and "off" (when looking at the event list), but there are numeric values in between.

[Kurt Cowling]

I'm having the same problem with M5v2. I use a Yahama S90-ES that has a continuous pedal for hold so that you can do half-damper effects. Depressing the pedal holds the notes, but releasing the pedal doesn't stop the notes. I've confirmed that the last value sent by the pedal is zero (off). This pedal works fine in MachFive1 and with all my other hardware synths and VIs. A definite bug.

--Kurt

[bkshepard]

If you manually insert a Controller #64 with a value of 0 into the track, does that work?

[Kurt Cowling]

If you manually insert a Controller #64 with a value of 0 into the track, does that work?

Yes it does. In fact, when I release the pedal I get a quick series of events going from 127 down to 0. If I delete all but the last event (the zero) it works as it should. It's apparently the rapid succession of events that seems to confuse M5v2. (Although it reads the depression of the pedal fine, which is also a rapid succession of events, just in reverse, 0-127.)

[pandrews]

So, do you think there's some kind of setting that would correct this problem?

Thanks.

[Kurt Cowling]

So, do you think there's some kind of setting that would correct this problem?

I don't think so. I emailed MOTU tech support and they confirmed they can reproduce the problem. They are looking into a solution.

[Shooshie]

I'm just curious. First of all, are the pedals sending CC#64 events?
Secondly, are all the pedals that are having trouble sending continuous data as opposed to ON/OFF?
Thirdly, Are the offending pedals all by Yamaha?

If the answer to all those questions is yes, then check the CC data closely and see if you are getting a smooth stream of numbers in the data, or if there are some events that appear out of place. The reason I ask is because the Yamaha Disklavier Grand, which should be a whole different beast, used to send very screwed up pedal data. If they are using any of the same programming from the Disklavier pedals, there may be a problem with it. The Disklavier Grand would send out a smooth stream of data, except for two points, close to the beginning of a stream, and close to the end of the stream. (We're talking about one action -- depression and release of the sustain pedal). The continuous data was all CC#64, and it all streamed as you would expect it to, but at those two points -- near the beginning and the end -- it would send the switched version of that data -- on and off. In other words, it would send out a CC64 "ON" command in the midst of the stream, and then a CC64 OFF command near the end of the stream. On and Off are not the same as 127 and 0, though they are supposed to have the same effect. 64 is a "switched" controller which can be viewed as continuous data. There may be some programming kludge involved in it.

On the Disklavier, which of course is moving a heavy pedal with a huge solenoid powering it, it would cause a "clunk" that made your pedals noisy and uneven. I was the first to figure out the problem and make Yamaha aware of it. I do not know if they fixed it in subsequent models, and it's just possible that it's the same programming they are using in the pedals you're talking about. Come to think of it, their Unacorda pedal did the same thing -- CC#67.

Sorry for the detour if this has nothing to do with the problem, but Yamaha DOES have a history of funky pedal data. You might check the values carefully and see if they are all sequential and correct.

I've tried this again and again with my current setup. My Kurzweil pedal is merely a switched CC64 controller, and sends only ON and Off. For whatever reason, it has not failed on any sample in Mach Five 2.

Shooshie

[grifmx]

switched sustain pedals with another one. no difference.

FATAR footswitch 93 & large Roland piano-like sustain pedal.

next thing to try tomorrow is different controller I guess.

[3over3]

Pedal working fine in v2 here - it's a switch though, not a continuous controller.

[Kurt Cowling]

I'm having stuck notes. My pedal is a Yamaha continuous pedal. The actual data from the pedal looks fine. It's all CC64 and there are no spurious bits of data that seem out of line. Smooth curve up when depressed, smooth curve down when released. The pedal works fine in M5v1 and all other synths, both hardware and VI.

After a little experimenting I've found this:

Any CC64 value 64 or higher is considered "ON".
Any CC64 value 63 or lower is considered "OFF".

****If there is any CC64 controller event with a value of 64 or higher between the note off and the CC64 Off (Off being 63 or lower), then the note will stick.

--Kurt

[Kurt Cowling]

On and Off are not the same as 127 and 0, though they are supposed to have the same effect.

Yes, they are the same. View CC64 as continuous in DP and type a value of zero; DP will change it to "Off". Same thing with 127 and "On". Values in between may be interepreted as on/off or continuous, depending on the receiving device, but 0 is Off and 127 is On.

The continuous data was all CC#64, and it all streamed as you would expect it to, but at those two points -- near the beginning and the end -- it would send the switched version of that data -- on and off.

This sentence doesn't make sense to me. There isn't a switched version of CC64 and a non-switched version of CC64 in terms of the actual data. The difference is in how the receiving device interprets it (and how you choose to have it displayed in DP).

--Kurt

[Shooshie]

Maybe it doesn't make sense, but that's exactly how it was displayed in DP. Maybe it's a DP/Yamaha thing. The data would be going along like this:

99
98
97
96
95
94
OFF
93
92
91
90
etc.


Or:
1
2
3
4
5
6
ON
7
8
9
10
11

When you converted it to a graph, it would be a smooth curve except where the OFF or ON was, and that would be a full-off 00 or full-bore 127, making a huge spike in the curve, either way you went, but I repeat; there was a difference in the way DP displayed the switched data vs. the continuous data. The CC had a maximum of 127, and did not say "ON." Likewise, the minimum was zero, and did not say "OFF." The fact that the ON/OFF were mixed with the CC numbers tells me there was a problem at work.

Incidentally, I could edit out the ONs and OFFs, leaving only the CC numbers, and everything worked fine on the Disklavier, giving a much smoother operation of the pedals. The ON/OFF only slipped in there when recording the pedals.

I don't have enough information to work with here, but something tells me these problems are related. This may end up being a MOTU thing, and not a Yamaha thing. Wow, that would be a hoot if we uncovered this bug which is now somewhere between 15 and 20 years old!

Shooshie

[Shooshie]

Please don't remind me that anything above 64 is "ON" and 63 or below is "OFF." I've actually told you the way it displayed. ON and OFF were not the same as 127 and 0, even though their numeric values were probably transmitted the same. They were displayed differently when using Switched method, while when using the continuous method they said 0 and 127. The fact that the two were actually mixed together is what was strange about that, and it may be the clue to solving this problem in MachFive. (or it may have absolutely nothing to do with it) I'm just saying that we have a history of strange problems in DP (with the Yamaha Disklavier, at least) when using continuous pedals.

Shooshie