Sequencer makes a noticeable break during playback

[Jay C]

Hi 🙂

 

I made this user-sequence for a song that I want to play with the sequencer playing drums, bass and chords.

This setup i've made makes use of all 8 patterns in the sequencer.

 

However, when I play the sequencer patterns, or to be more precise, when I jump from pattern 4 to pattern 3 

there's a noticeable break or pause in the playback of the sequencer data. It annoys me greatly since I spent

quiete some time to program the sequencer. Why is it doing that?

 

It seems that on it does this break everytime it reaches the third beat of the measure in pattern 3.

 

I tried to disconnect any other instruments/devices connected through MIDI, in order to see whether that made

a difference. It does not.

 

So I looked closer at the trackparameters of pattern 3. The different parts have different values in the parameters step size and notelength:

 

Drum parts 1 and 2 have 1/16 note values as step size and a 99% notelength. Drum part 3 and 4 have 1/4 note values and a 99% notelength.

Drum part 5 has 1/16 note values as step size and a 99% notelength. Bass part 6 has 1/16 note values and a 100% notelength

 

In pattern 4 the bass part 6, for instance, have 1/4 note values and a 50% notelength  - for the sake of programming the bass line.

 

So, I guess my question is - does these varying notevalues (stepsize) and notelengths in a pattern and between to patterns

create this break in the playback that should not be there?

 

Thanks!

 

Jay🎹

 

 

 

[Jokeyman123]

OK-took me awhile to even attempt to answer this one! You might want to study this old group of posts from me, Brad, AlenK and others-might help you understand what is happening, and whether this is a glitch, or the way you recorded the tracks. You can create loops that sound longer or shorter than the default 16 steps. Might be what you did, even with one track, it could sound like it is stoppng between parts.

 

https://www.casiomusicforums.com/index.php?/topic/4633-linear-recording-with-xw-p1/

 

I can't tell unless I saw exactly how you recorded these tracks-did you do this live from the keyboard or did you program it manually in step mode, or just using the sequencer buttons? Each method can give you different results-using the step sequencer buttons are easiest IMO because if you miss a 16th note step, you will see it right away. A little harder from the keyboard live.

 

This is unfortunately one of the more complex sequencers to wrap one's head around, IMO if you want to go past the simplest 4 on the floor beats, which should not cause any break between different parts. It is possible, for example though-that the quarter note tracks somehow are recorded as a longer or shorter loop than the 16ths and something is missing in the number of beats. This sequencer can be "fooled" into sounding like it is playing in a different time signature when in reality it is always using the 16-step foundation for everything. This might cause a pause between parts although I've not noticed it with any parts I've created.

 

I will try to duplicate something similar to what you've done and see what happens. Even if you have separate loops of different note durations-everything should sync anyway-but the different note resolutions could be screwing this up somehow.  I'll post back if i come up with anything that might make more sense! 

Edited July 11, 2021 by Jokeyman123

[Jay C]

On 7/11/2021 at 4:30 AM, Jokeyman123 said:

OK-took me awhile to even attempt to answer this one! You might want to study this old group of posts from me, Brad, AlenK and others-might help you understand what is happening, and whether this is a glitch, or the way you recorded the tracks. You can create loops that sound longer or shorter than the default 16 steps. Might be what you did, even with one track, it could sound like it is stoppng between parts.

 

https://www.casiomusicforums.com/index.php?/topic/4633-linear-recording-with-xw-p1/

 

I can't tell unless I saw exactly how you recorded these tracks-did you do this live from the keyboard or did you program it manually in step mode, or just using the sequencer buttons? Each method can give you different results-using the step sequencer buttons are easiest IMO because if you miss a 16th note step, you will see it right away. A little harder from the keyboard live.

 

This is unfortunately one of the more complex sequencers to wrap one's head around, IMO if you want to go past the simplest 4 on the floor beats, which should not cause any break between different parts. It is possible, for example though-that the quarter note tracks somehow are recorded as a longer or shorter loop than the 16ths and something is missing in the number of beats. This sequencer can be "fooled" into sounding like it is playing in a different time signature when in reality it is always using the 16-step foundation for everything. This might cause a pause between parts although I've not noticed it with any parts I've created.

 

I will try to duplicate something similar to what you've done and see what happens. Even if you have separate loops of different note durations-everything should sync anyway-but the different note resolutions could be screwing this up somehow.  I'll post back if i come up with anything that might make more sense! 

Hi Jokeyman123

 

Thank you for your elaborate answer to my question. 🙂

 

In short,  as I recall it, I mainly programmed the sequencer in step edit-mode (via the edit menu), although some chords I think I input via the keyboard (but still in step edit-mode, I think).

 

This problem with the sequencer pausing  or halting playback long enough to be noticeable - and then resuming playback seems to come and go.

I wanted to document this in some way, since it's a bit hard to explain what actually happens.

 

So I rigged up a video camera to record the display since it shows infomation like tempe, measure and beat + the graphical output of the synthesizer's level meter, that shows which parts are playing. In addition to this, I hooked up the audio output from the synthesizer to the mic input of the video camera to have audio as well. 

 

The first attempts to do this were futile as it seemed the synthesizer played it's sequencer parts flawlessly. 

However, I now have a short video where this error occurs. First the playback is normal and then the un-called for pausing commences.

And It is a preset performance I am using. Also, I had the arpeggio activated during this recording, and it too halts or pauses.

It is like the instrument stumbles over its own feet - if you can follow me?

 

I am considering uploading the video somewhere online on a platform that allows for private sharing for the purpose of solving this mystery.

 

In conclusion, I think the problem is caused by a technical failure of the instrument, not a matter of how the sequencer is programmed. 🙁

 

Jay 🎹

 

 

 

 

[Jokeyman123]

You can always send me a PM if you don't want to post publically, whatever the reason no problem.

 

Now that you've described the glitch this way-I think you have stumbled upon possibly-the limits of the CPU in the XW. This would explain why it sometimes does this and sometimes doesn't. There wouldn't be much we can do about this-since we cannot upgrade internal meory for example, or slim out the Windows OS like I've done to reduce Windows CPU overhead for use in older computers. Especially since you've noticed this with the arpeggiator, its obviously not something you have done to cause it. as I said, even if you did all kinds of different note durations, loops-it should still stay synched. I can tell you I have noticed this problem with the PX560 on rare occasions-but only when I am creating a large multi-track composition with the midi song recorder-or doing alot of editing at once with the hex layers, so there is probably a limit to what these CPU's can do-although i have not hit that wall with the XW, its certainly possible. The 560 sometimes locks up for ome seconds, then resumes where it left off-whatever i was doing before. I wish I knew the answer-how to avoid this glitch in the sequencer-I haven't come across it myself yet, I've created simple parts though so far-3-4 tracks plus drum tracks. I'll push it and see what happens.

[Jay C]

Ok, ok ok ok. 

 

I think I found a solution (cross my fingers and hope it lasts*).

 

I took the plunge (so to speak) and updated the firmware (whoo-hooo) 🤭. No problems occured during the update proces, it was quite un-dramatic. 😉

 

While the updater-application run I noticed it deleted or erased the XW-G1s memory. So maybe this erase-thing did it, huh?

 

With this update accomplished I started testing. And lo-and-behold, the sequencer now plays in time, no breaks or break-downs 🥴.

No lagging CPU, seemingly.

 

After re-installing the ALL-DATA file it still plays as it should.

 

 

Jay🎹

 

*No guarantees given 

[Brad Saucier]

Moral of the story, make sure your firmware is up to date. 

[Jokeyman123]

Good. If it happens again-aside from re-flashing the memory-I would hope you can "flush" the memory again somehow. I suspected it might be a matter of reaching the limits of the CPU or the internal memory-or both. I don't know the particulars of the Casio CPU in these-but since the operating system can be flashed-at least part of the memory might reside on the cpu which would explain how your doing this would restore this function in the sequencer. And unless Casio could tell us, or if many others have the same problem, I don't think I could call it a "bug" or flaw. I'm surprised then-that a factory restore did not fix this. And it it happens again-can you re-flash the same version of the firmware to fix it again? Maybe Brad or Mike Martin would know.

[AlenK]

There isn't necessarily a mystery here. All software updates fix bugs as well as (sometimes) adding features. This timing problem could have been caused by one of those bugs. Casio didn't list anything that sounds like it might cause this problem in the release notes but that doesn't mean anything, not with "Improved certain operation" in the list of both updates to the G1 and P1 to date. 

[Jay C]

11 hours ago, AlenK said:

There isn't necessarily a mystery here. All software updates fix bugs as well as (sometimes) adding features. This timing problem could have been caused by one of those bugs. Casio didn't list anything that sounds like it might cause this problem in the release notes but that doesn't mean anything, not with "Improved certain operation" in the list of both updates to the G1 and P1 to date. 

You may be right about that, AlenK.

 

I am just happy that the sequencer now plays in time, every time (again). 😄

 

This problem was not present when I bought the instrument some months ago. It was preowned when I got it but seemed to work fine - just as the ad promised. The problem of the sequencer going out of sync came later on. Some days it would play it's notes perfectly, some days it just messed up in an unpredictable manner. 

 

Jay 🎹

[AlenK]

Software is complex. And hence, devilishly difficult to fully test and debug. It's possible that this particular bug of the older software version doesn't rear its ugly head until the Flash memory is stuffed with user content to a certain amount.

 

As a design engineer in my old job, I used to write embedded code for microcontrollers (on circuit boards that I had also designed). I spent many late nights debugging my own code. It was amazing just how many dIfferent ways I could screw it up!

[Jay C]

Oh, no! 

 

it.still.does.that 🤧

 

[I really wanted to end this thread on a happy note (no pun intended) and have big fat SOLVED added to the title. However, the problem still persists]

[Jokeyman123]

I didn't think the firmware update would solve the problem. The CPU is probably reaching the limits of what it can do-why, i have no idea. A timing problem-usually a CPU that can't keep up. I wish I knew how to solve this problem. If it were a memory constraint-factory restore should clean the memory registers, why I think it is a CPU. I've never had this problem with the XW-P1-possibly it has something to do with the sampler functions needing CPU cycles? maybe AlenK would know better.

[Mike Martin]

A change in effects change could be a cause. Are you sequencing the solo synth?  

[Jay C]

On 7/27/2021 at 1:51 AM, Mike Martin said:

A change in effects change could be a cause. Are you sequencing the solo synth?  

 

OK. Solely looking at the user-sequence I am talking about in my original post in this thread - if I go to this sequence (that is a part of a user-performance) and press Edit and successively press Mixer I get the following information from the display:

 

Part 01

Tone P:084 Whistler DSP

 

since I wanted to play the melody using this tone. To by best knowledge that is a solo-synth tone. 

 

So, could that be the culprit?

 

None of the other parts in the mixer have solo-synth tones applied to them.

 

I am going to replace the tone for part 01  with another that isn't a solo-synth tone to see if it makes a difference. And let you know of the results.

 

Whether I find a solution to this problem or not, I wanna thank everyone for their contributions in the effort to find the root cause

of the sequencer playing out-of-time intermittently. 😀

 

 

Jay 🎹

 

 

 

 

[Mike Martin]

I'll do my best to get my XW-P1 out tomorrow and do some digging for you. 

You absolutely CAN sequence the Solo Synth but I want to double check a couple things.