I’ve always been unsatisfied with the Motif’s portamento/glide implementation. However, since the technical resources and will don’t exist to do anything about it, I’ve stopped bothering to complain. One feature doesn’t break a deal, after all. On the XF, though, I’ve noticed a new terrible failing of the portamento function that wasn’t present on the XS. This one isn’t as easy to ignore.

Here is some background regarding how portamento works on the Motif. If you’re playing a note (like C3), and, while still holding the note down, press a second note, then the Motif acts as follows.

1.  The Motif determines the keybank that is used by the second key that you pressed.

2.  The Motif switches the voice elements that were playing the first notes sample to now play the sample that is associated with the keybank of the second key, but continues to play that new sample at the pitch of the first key. So, the Motif has switched the playing sample to the one needed by the new key (the portamento target).

3.  Based on the settings of the portamento function, the Motif begins to glide the pitch of the new sample up or down until it reaches the pitch of the second key that you played.

My old gripe is that switching between samples always produces a click, and the Motif always switches samples when you press that second key. Even if both keys belong to the same keybank, and use the same sample, the Motif will retrigger the sample from the beginning. This always results in a click, as there is a momentary jump in the phase of the waveform. This means that analog patches will never smoothly glide from note to note, like on real analog gear (or even sample-based synth engines with advanced portamento behavior control like Roland and Korg gear from the last 10-15 years). However, Yamaha’s way has a great advantage when using portamento with samples of real instruments. If you’re using natural samples, and you glide from one note to another without switching samples, then you’ll be listening to a sample that is playing at the right pitch for the second key that you pressed, but the tambour will be wrong. It is like older samplers that stretched just a few samples all of the way across the keyboard. Beyond a certain range, it is obvious that you’re listening to a sample that has been speeded up or slowed down, rather than an actual sample of the instrument playing the exact note that you played. By switching samples when you press the second key in a glide, Yamaha makes it so that, when the glide finishes at the second key, you’ll be listening to the sample that was originally intended to be played when that key was pressed, not a sped up or slowed down version of the sample on the first key.

Okay. Enough of the background. Here is what is wrong with the XF. Gliding down is broken. I’m first going to show you how to experience this for yourself. I know that this example isn’t something that you’ll run in to every day. It isn’t supposed to be. The example is only so that you can hear what I mean. Later, I’ll tell you how this is a problem in practical situations.

Select voice Pre1 A01. Go to the F2 screen, and turn on portamento at the top. Now, when you’re holding one note, and press the second, the pitch will glide, based on the rules that I explained above. However, you’ll see that the Motif completely ignores the rules when gliding down.

I’m using a Motif XF6. Start by placing your left hand in position to hit lowest C. Now, play a slow alternating pattern, with your left hand playing lowest C, and your right hand moving up the keyboard, from lowest C, about a 4th or a 5th each time. Play like this: C1, G1, C1, C2, C1, G2, C1, C3, etc. Remember, each time, to press the second note before you let go of the first. You’ll notice that when the pitch glides up to the note played by your right hand, the resulting sample when the glide stops has a natural tambour. However, when you next play the lowest C again, and the glide finishes, the tambour will sound increasingly worse, the greater the distance between the notes. As your right hand moves further up the keyboard, you’ll clearly hear a huge amount of aliasing in the sample played by your left hand’s C.

For another vivid way to hear this, switch to any of the string ensemble voices, turn on portamento, and repeat the same playing behavior.

As you’re playing in this way, it is important to note that the second time that you play the lowest C in a row, the Motif will switch to playing the correct sample for the key. So, if you play C1, G3, C1, the second C1 in this series will sound heavily aliased. However, if you play C1, G3, C1, C1, the last C1 here (the second time in a row that you pressed the key), will sound correctly.
Many voices behave like this, but others don’t. I haven’t found the pattern yet. Most real world instruments act like this. I think that switching between samples in different keybanks is what triggers this behavior. On the pianos, for example, I only hear a change in tambour after my right hand passes a keybank boundary. Not all of the synth voices do this, though, and some of the ones that don’t still have waveforms with at least a few keybanks. On the synth voices that do, though, playing arpeggios with wide octave ranges results in the lower notes sounding muffled.

Try it yourself. What do you think?
If people don’t believe me and/or can’t hear it, then I’ll upload an audio clip, but I really don’t want to waste 30 minutes with all of that if everyone else can reproduce this.

This has been my gripe with Yamaha’s portamento, glide since the days of the SY85. A lot of keyboard companies don’t get this. The only keyboard companies that have done it right is Kruzweil and Ensoniq. Roland has it partially correct, but Yamaha never quite got it right. If I set my portamento I should be able to take a sine wave and slide my fingers across the keyboard and NEVER EVER hear a re-trigerring of the key.

Check out this video of the Yamaha EX5: http://www.youtube.com/watch?v=p-HCyA-fDss. DON’T BE FOOLED, you can only do this with the AN (Analogue modeling option), but you can’t do this with regular samples. The answer I got from Yamaha some years back, was that this was only possible with Analog modeling, which is completely false. Emu, Ensoniq, and Kurzweil can do this same type of slide glide effect with regular samples....

Yamaha does not realize how critical this feature is for sounds. I guess what you and I are asking is for a GLIDE function that really works. All the software synths have a glide function for SAMPLES that allow one sample to glide in to the other without the retrigerring of samples.

You’re very right.

On Roland synths, for example, there is a voice parameter that changes how retriggering of samples during portamento works. You can either turn retriggering on or off. Retriggering is unquestionably better when using samples of real world sounds. For synth sounds, though, retriggering is terrible. With retriggering always on, analog sounds don’t slide correctly, and playing trills and short arps plays a click at every step.

However, as crappy as that behavior has always been on the Motif, the current problem is beyond that. When gliding down, they have the worst of both worlds. The Motif XF not only retriggers (introducing the click), but it doesn’t even retrigger the correct new sample. Try it.

here’s the reason why it is bad not to innovate when a new keyboard is produced, yamaha, even if a lot of people enjoy paying so much money and defend it against unsatisfied customers.

Not to “defend” (anyone can think whatever they like), just an important point of order;

No discussion of Portamento is complete without a discussion of the parameters (all the parameters of the portamento) Please before you conclude anything consider how it actually works.
Go to VOICE
Press [EDIT]
Press [COMMON EDIT]
Press [F1] GENERAL
Press [SF2] PLAY MODE

Your full compliment of Portamento parameters are found here. In the current discussion you have not figured out how it works… in particularly why it works differently in some Voices than in others. Portamento on a piano (well, we will not even talk about that decision) but perhaps what you expereince on the other Voices is a function of how the following Portamento parameters are set. Maybe the following will help (if you need Portamento on that piano you may be able to adjust the response more to your liking). We think it is at least worth a try.

Portamento Switch*
Determines whether Portamento is applied to the current Voice or not.
Settings: on, off

Portamento Time*
Determines the pitch transition time when Portamento is applied. The effect of the parameter differs depending on the settings of the Portamento Time Mode. Higher values result in a longer pitch change time.

Settings: 0 – 127

Portamento Mode
Determines how Portamento is applied to your keyboard performance.

Settings: fingered, fulltime

fingered
Portamento is only applied when you play legato (playing the next note before releasing the previous one).

fulltime
Portamento is applied to all notes.

Portamento Time Mode
Determines how the pitch changes in time.
Settings: rate1, time1, rate2, time2

rate1
Pitch changes at the specified rate.

time1
Pitch changes in the specified time.

rate2
Pitch changes at the specified rate within an octave.

time2
Pitch changes in the specified time within an octave.

Portamento Legato Slope
When the Mono/Poly is set to “mono,” legato playing may produce an unnatural attack depending on the waveform assigned to the selected Voice. To solve such a problem, you can use this parameter to adjust the attack of the Voice. Normally, this should be set to a low value for waveforms with short attack times, and should be set to a high value for waveforms with long attack times.

Settings: 0 – 7

Bad Mister, I appreciate you posting. If anyone has the technical answer on here, it’s usually you. However, if there is one part of the Motif that I feel that I understand in profound detail, it is the portamento behavior. I turned it inside and out for years trying to figure out any way that I may have overlooked that would get rid of that freaking retrigger click. I’m not concerned with the retrigger click in this thread. I’m not even concerned with putting portamento on pianos. I just used the piano voice as an example where people could hear the effect. I’m very clear about how portamento works. The only part that I’m not clear about is the cases when the Motif selects the incorrect sample during downward glides *ONLY*.

Just so people won’t be picking at my methodology, instead of addressing the problem, lets forget the piano, and talk about a solo violin voice. Go back and do the exact same steps that I said for the piano, but use a violin instead. If people can’t imagine why portamento would be useful on a violin, I might as well give up. I’m aware that the XA function is used to trigger alternate legato samples for most of the violins, but use one without the XA setup. I’m trying to demonstrate behavior only.

Since a violin is exactly the type of voice that *should* use retriggering, all of my retriggering gripes aren’t even on the table in this case. My one and only concern, in this case, is why the Motif retriggers the incorrect sample when gliding down to a note over a glide distance that spans keybanks.

I can adjust thos portamento parameters, but none of them affect, in any way, how the Motif selects samples to play. Portamento switch on/off: well, it must be on, or else we have nothing to discuss. Portamento time and portamento time mode will simply affect the speed of the glide, but the Motif’s sample selection behavior is identical, whether the speed is fast or slow. Portamento mode: again, doesn’t affect sample selection, though I suggest that you set fulltime in order to more easily hear the results of the repro steps that I posted.

Regarding Portamento Legato Slope, I’ve never been able to satisfactorily hear this parameter demonstrated. All of the way back to the ES, when I’ve changed this setting on an element playing either a short synth or long natural sample, with the voice in mono mode, I’ve never heard any difference. Could this fix the problem? I don’t really think so, but, hey, I’m open minded, so would love to hear from anyone with any theories regarding this parameter.

However, that brings us to the end of portamento parameters. For the record, I suggest trying my test with portamento switch on, rate time mode set to rate1, rate set to any large value that you desire, and portamento mode set to fulltime. Try with a solo violin voice. After you’re gliding for more than an octave or so, you’ll clearly hear aliasing each time you glide down, though the second time that you play the lowest note the Motif will select the correct sample on the key, and the aliasing will be gone.

I suspect that this is going to be like 2007, when it took pages of a thread to prove the whole retriggering of samples actually existed. I had to make an audio demo with the Motif XS playing virtually the same simple synth voice that I’d played from the Korg TR rack and the Roland XV5080, before anyone would believe me that the retriggering was a fact.

So, once again, here is another audio demo, demonstrating that this is not in my head.

http://bryansmart.com/files/temp/xf_broken_porta.mp3

I demonstrate with both a sax and a pad voice.

We’ve covered the settings. We’ve heard the results. If I’m missing something, what is it?

Of course, we could reduce the impact of this problem by not gliding down via portamento for more than a 4th or 5th at a time, though there would still be a change in tambour. Beyond that, I guess we should avoid using arpeggios that span more than an octave or two, at least if we wish to use the voice in mono mode. I guess TB303 style porta arps are out.

I don’t care to bash the Motif or degrade anyone’s work. This just seems like a small software oversight. The right sample is selected when gliding up, just not when gliding down. Some Yamaha programmer probably missed a conditional branch somewhere. No big deal. Would just like it acknowledged and resolved.

On the other hand, if you, or anyone, can still see how I’ve managed to overlook some setting that is preventing the Motif from selecting the correct sample, please let me know.

Also, I’m almost certain that the XS didn’t do this. However, when making a claim about an issue like this, I hate to claim certainty about something unless I can verify it, as, if I’m wrong, that will just be fuel to dismiss me. I played with this a lot on the XS, and don’t remember this being the behavior. However, I no longer have an XS, so can’t verify it. I’d really appreciate it if someone on the forum with an XS would doublecheck this for me. You have the same presets that I use in the demo. You should be able to follow along, make the same adjustments, and not experience the same problem.

Sorry. Accidentally double-posted.

Hello, Bad_Mister, I happened upon this post in search of a solution to a potentially related problem.
You may recall (though you have helped thousands of others, this week alone, perhaps) I use an S30 and a Motif xf7. There are a couple voices I prefer on the S30 over the Motif; among these voices are the piano/strings and a few of the guitar leads. I have recently been using the sequencer on the Motif to rework many of my projects, formerly done in Cakewalk, using some of the Motif voices that are superior to the original voices I used from the S30. Additionally, the the 76-key bed suits me better than the 61.
Anyway, when I play the S30 via midi from the Motif, the piano/string voice is perfect in every respect. There is no way to tell I am not playing it on the S30, save the extended keyrange. However, a voice called VeloChunk will not play correctly. On the S30, I can hold down two or more notes to make a distorted chord-like sound, but still rip out a searing lead and fake some harmonics now and again.
On the other hand, when I play that voice on the Motif, whether I am in song or voice mode, there is a distinct interruption from one key to the next (no balanced flow), even when the sustain peddal is used. Furthermore, not just any two notes can be held simultaneously, as if the polyphony had been exceeded.

Therefore, my question: If I am using the Motif XF7 as the master keyboard, are there internal settings I have to make on the Motif in order to ensure that the voices play back exactly as they would if they were played from the S30’s keybed?

Thank you for your time and consideration.

Hey Bsmart,

You’re not crazy man! I’ve been fighting this issue forever. This simple implementation would allow the Motif to really, really sound like an analogue synth. This is why I don’t even bother making synth basses with the Motif, because it makes absolutely no sense to have a synth bass without the correct portamento/glide implementation.

I shot a quick video with my iPad and iPhone to show how the portamento/fingered/glide should really sound like:

See Video: http://youtu.be/Fkt5tEHZVig

OMG, I just ordered a Motif XF7 and this might be a deal breaker. Even a Garage Band instrument on my iPad 2 does the correct portamento. So does my 20 year old Kurzweil K2000.

Portamento Glide is important.