I've owned a Reface CS since January. I absolutely love it, but within a day or two I started to notice some bugs. (It either it came with Firmware 1.30 or I updated it myself the very first hour I got it.)
For example there's this one patch that keeps changing over time. It's like there's a super slow LFO going, while the FX and LFO are turned off. The rate of change is also much slower then the LFO is capable of. Today I had another patch that did the same thing but a lot more noticable and I made a recording: https://www.dropbox.com/s/nu2peh1x9eip8h3/wtf3.mp3?dl=0
As you can hear there is a Huge change in the timbre while I'm not doing anything. I'm feeding it Midi notes through Reaper and I'm also not convinced the timing is correct. It seems like there's some minor variations. Reaper and SoundMondo do not seem to play nice together. But don't get me started on SoundMondo bugs...
I also very quickly noticed problems with the looper. If I turn on Distortion and move the Depth and Rate sliders quickly it affects the speed of the looper. I assumed this is just the DSP engine not being able to cope with everything the designers wanted it to do and that's okay. But I've noticed more problems with the looper.
Here I'm playing a simple 4 note sequence and again I'm not touching ánything.
https://www.dropbox.com/s/gv4r30bm8gueb90/wtf.wav?dl=0
Yet every time it loops through I get different results. Notes are missing. This seems to happen a lot with some of my patches with a long attack. I can't see why after recording (and quantizing) the loop shouldn't sound the same every time.
https://soundmondo.yamahasynth.com/voices/9004
This is the patch I used. I've asked the Synthesizer subreddit if they could reproduce it as there are a lot of Reface users there, but I didn't get a response. I've also asked the Reface Group on Facebook if they could reproduce the changing timbre problem, but no response to that either.
Again I love my CS, I'd happily buy a second one if I'd ever lost it or something, but how hard it can be to get perfect timing out of it when sending it midi data, how the sound just won't stay the same sometimes and how it's missing notes makes it very hard to use. Especially live, but also in the studio this can be very time consuming. I'm really hoping there's an easy fix, because I don't want to replace it. I'm not interested in the popular alternatives that only have 4 note polyphony and cost way more.
What do you guys recommend? Should I send it back? Should I try and re-install the 1.30 Firmware? I guess if it's already installed it just won't do anything?
(Also making a long post while only being able to see the last 1.75 lines is a real pain. And why can't I choose a username? Or atleast change it after registereing in my profile settings? Is this Forum as bugged as SoundMondo or is it just me?)
I saved the constantly evolving patch. https://soundmondo.yamahasynth.com/voices/22592
If only it were intentional and controllable it could be an interesting feature. I'd love to have a slower second LFO for drones.
I’m not near my studio right now (about 2500 miles away) we will look into this and get back to you. Thanks for the report.
“Long attacks” says to me ‘danger, danger Will Robinson’ (forgive the Lost in Space reference) but long slow attacks used with a looper Can by itself cause issues as Note-on events simply take longer to sound, once you exceed eight note-ons it’s highly likely you will be stepping on the toes of notes that are attempting to begin while others are sounding, it is inevitable you will run into the eight note polyphony wall. Try reducing the length of the Attack. If the problem disappears, then slowly raise the Attack slider, noting where the issue begins... then slowly reduce the tempo, if the problem goes away, you can be pretty sure it is the “long Attack” time that is your issue. It’s a ...Patient: “doctor, my arm hurts when I bend my elbow the in the opposite direction...” Doctor: “Don’t do that...”
See if the above experiment fixes/breaks/fixes the Looped data. (When I get back to my studio I’ll load up your sound and try it out)... until then
That's the first thing I thought too.
So first I sped it up. It helped somehow.
I slowed it back down to somewhere in the middle, it got worse again. I slowed it down all the way and it got better.
Then I set the octave slider 1 or 2 settings lower and I played along. It got better again. Quite strange.
I played along 16 notes, I stopped and after 2 notes it started to skip a few again.
Sometimes when I stop and play it again it does get worse when I speed it up though. It's almost as if the Release time is a little different every time I press play. Or maybe it just changes over time while it's playing.
Lowering the Attack or the Release time fixes the problem completely.
However, I don't believe it's the only problem. The inconsistency is quite strange. Especially the fact that I can play along, so it has to play double the notes not causing more notes to be skipped is odd to me.
I suspect that it just can't hold the settings correctly. Or it detects slider movement where there is none.
This reminds me of another problem that started occuring yesterday. Sometimes when I record something in Reaper and I've had Reaper and SoundMondo open at the same time it sends Midi CC data to the Synth. Even though I have turned off Midi CC Send and Clock Send in Reaper. I only requested it to send Midi Note data (and Velocity and note length of course), but the first note makes it switch to the Saw Wave.
There seems to be a theme here. In the first recording I posted in the original post I was mentioning a changing timbre over time, I suspect the Release time varying greatly being the problem of the long-attack-patch-problem and sometimes sending midi notes changes a parameter too.
I hope any of this information is helpful. It's hard to get something concrete as results are different every time I try something.
https://www.dropbox.com/s/1c4zj6ojn84z0pm/wtf4.wav?dl=0
I made a quick recording.
Here I'm starting the looper slow, I speed it up until it reaches the speed it usually gives me problems.
Then I go beyond that speed and it doesn't fix it. But while it's at a high speed I stop it and I play again and everything works fine. Then I speed up to about max speed and everything works fine.
Then I slow down to the problematic speed again, and once it gets close enough it starts skipping notes.
I hope this is more clear then the explanation in my previous post as you can just hear what's happening.
I just realized that I use soundmondo every time I test this.
Immediately after the last post I set the sliders as closely as Soundmondo will let me, tested it's still a problem, turned off the Reface and PC, turned on the Reface and it was much harder to reproduce. And if after minutes I finally managed to get it to skip notes I can just fix it by stopping and playing.
Which makes me think this is yet another Soundmondo / Midi CC data being sent for no reason problem...
I just put the 4 note sequence in a midi item in Reaper and let it loop through (16th notes at 120 bpm).
At first it bugged out, but after a few times starting and stopping I got it to work correctly for 5 minutes.
When it was working fine I made the Release as long as possible, no problems. I also added a lot more attack and again no problems at all. It played flawlessly for minutes on end.
I also tried letting Reaper play the notes twice as long so 2 notes are playing at all times and ramping up the speed until I hit the polyphony limit. I also lengthened the Release, but I can't hear any notes skipping untill it turns into a sea of sound and individual notes are indistinguishable.
Then I set the note lengths back to normal, 1 note playing at a time, and I sent Midi Clock out to the Reface from Reaper. While it was looping in Reaper I also set the Reface Looper to record, so it took over the exact same sequence.
I stopped the Reaper Loop and instantly I could hear notes skipping when it was all up to the Reface looper.
It does seem that it's worse in the Reface Looper then with External Midi input.
However that's not perfect either. Just like sometimes when Playing it manually notes are skipping too (although it's less common) and sometimes notes are skipping when I haven't used SoundMondo yet (also less common) it does also happen when sending notes through Midi. But again it's not as bad as using the Reface Looper.
So now I do know it's definitely not the patch or polyphony.
However I'm not sure if this is good or bad news. I was hoping Sequencing it externally would fix the problem completely. But I can't get rid of this bug.
I'm also not entirely sure anymore the problem is random Midi CC data being sent, because it doesn't seem like I can break it by adding Release Time.
(Although the randomly changing timbre is still a problem, so I do think it does send random Midi CC Data. I just don't think that's the cause of the Looper problem.)
Any thoughts yet?
Or should I just return it and get a Ultranova? It's also a Audio Interface, so I assume that one can actually handle being hooked up to a PC and it doesn't require something like SoundMondo to save patches. I really like the sound character of the CS, but this is getting ridiculous. I only have so much time to work on my Music and music production is time consuming enough as it is without these problems. In the last week I've spent twice as much time with these bugs as I have actually making music and I'm completely done with this.
(And it's not like the Reface Series or SoundMondo is in early stages of development either.)
Unfortunately my Reface CS has a problem with randomly skipping notes too, basicly they are not skipping, they are get triggered delayed, sometimes, for e.g. on sequences, arps or fast playing, it seems that they are skipping.
I will return my Reface CS because this is a major issue but it seems that Yamaha don't care about that. "Don't exceed the polyphony" is how they argue but that is just a cheap excuse, because when exceeding the polyphony limit on a synthesizer, new voices should be stolen from decaying voices and not delaying new voices.
Good point, I didn't think of that for some reason.
Yeah I think they'll be getting 2 CS's back soon.
It's a real shame, because I love the filter, simplicity and keybed. But I just don't have the room or money to hold on to gear that's too broken to use live.
Just curious, did you contact them about the problem too and get a similar response?
Or are you not content with the reply in this specific topic? Because it's possible they'll get back to us with a different reply, he was traveling when he posted and didn't have a Reface with him.
yet another broken patch
https://soundmondo.yamahasynth.com/voices/23855
it's being played quite slowly, at most 3 notes are sounding at the same time
New Bug:
Sometimes when I load a patch all I get is silence.
It's not entirely new, it wasn't caused by last nights update, I just hadn't reported it before and I hadn't found a Patch that did it more then once.
But this patch gave me silence two or three times now:
https://soundmondo.yamahasynth.com/voices/22592
(Yes it's the same patch I posted before that keeps changing)
So is there any chance you're going to look in to these issues?
RefaceCS ASDR
Sorry, for the delay... I lost the thread, fortunately, when that does occur, they “magically” arrive in my email (thanks to the behind the scenes site magicians!) I thank them, because I remember I promised to look into it but ... life interruptus...
It’s interesting but let me tell you what I’ve found thus far... and if I’m missing something important please let me know...
I loaded several of the sounds, the description: “For example there's this one patch that keeps changing over time. It's like there's a super slow LFO going, while the FX and LFO are turned off.” gives the biggest clue, IMHO as to what is happening.
Envelopes can close down in legato play situations....and behave exactly like a big slow one-time LFO. That’s exactly what an envelope can be described as... the LFO automatically repeats where the envelope is strictly, in this case, a “one-shot” movement. The sound is allowed as long as the envelope supports sound, once it completes its tour, no sound can happen until a new envelope is triggered, by a NEW Note-on event.
The slider FEG/AEG is the critical Depth setting that deepens/lessens the behavior. It’s like an envelope multiplier... It is possible to create an AEG where as long as the envelope is ‘traveling’ (that is, you have pressed a key... which “begins” the envelope, and a note-on is maintained, that envelope continues, it only restarts if all keys are released and the Release Time is engaged ....
Several of the sounds are mono ... eventually the time function of the envelope runs out. Now you windup with no AEG to support sound... until you trigger a new envelope.
In each case, that I was able to load, the issue is easily adjusted by changing the FEG/AEG slider. As this slider is lowered you are increasing the AEG DEPTH... This tends to increase/intensify the current set AEG (ADSR) settings.
Setup one of your mystery sounds... adjust the FEG/AEG slider until you reach the exact point where the AEG causes the sound to stop.
On a mono sound hold a key, trill an adjacent key... the AEG will eventually shut off amplitude, even though you continue to trigger notes... it’s not polyphony, you have an envelope that has “closed” the sound down... it has timed out. You must release and trigger a new envelope. You can just hold a key you don’t have to play anything (in fact, it makes the whole issue clearer if you don’t play anything... just listen to the envelope).
You can also setup a Filter EG that when the envelope closes, this will stop sound, as well. If the FEG is not allowed to be open enough to support a tone as the envelope reaches completion of its tour, sound stops. And you must retrigger.
You can, if you have the patience, play a single held note, wait to hear your entire envelope. You may, if you do not wait long enough, conclude that the sound will just simply keep holding... and that is when you have wrongly concluded something is amiss. Envelopes can be quite long, indeed.
In general, a sound will sustain forever when the key is held and the S (sustain) exceeds the D (initial decay). How long it takes for the sound to travel from Attack through the Decay stages to get to the Sustain portion can be surprisingly long when you have significant AEG DEPTH.
Forgive me for stating the basics, but perhaps this review will help the discussion....
Synths are always generating signal. It is, for all intents and purposes, an electric buzzer... connected it makes noise The keyboard is a fancy set on on/off switch which act to ‘gate’ the buzzer. Only when you press a key does the gate open and allow sound through. The sound is further shaped by the Amplitude Envelope Generator which controls what happens to the signal while the gate is open.
try this experiment...
LFO = OFF not needed
Select a Sawtooth Wave (bottom)
Set Texture and Mod to minimum
Set Cutoff fully up - filter wide open
Set Resonance fully down - eliminating any freq boost
Set the FEG/AEG slider all the way up - which is minimum AEG Depth
Start with ADSR all at minimum - in this experiment we will raise each one (slowly) in turn to observe how the affect the sound.
You should hear only a click as you trigger a key-on. Attack (Time) is immediate.
To understand the difference between Decay and Sustain... raise the Decay slider slightly until the “click” becomes a “plink” (a plink has a musical tone about it). If Attack is the time it takes for the ‘voltage’ to rush through when the Key opens the ‘gate’, the Decay is the level that voltage peak drops to long enough for us to identify a musical tone in the event. Playing with Attack all the way down, slowly raise the Decay slider noticing how it definitely affects level but you have to retrigger the key in order to determine what it has changed.... and very significantly it determines how quickly the sound returns to silence! In other words when all the way down, the return to silence is very immediate. The higher you raise Decay, the longer it takes to return to silence. The longer you must hold down a key for the envelope to complete its travels.
We will learn next the S is also going to affect level, but it is actually the level to which the Decay parameter returns the signal. When we only manipulated Decay, we noticed the higher we raised it, the longer it took the level to return to silence - in other words, you needed to hold the key down longer before it returned to silence. We now can state that Decay influences the return to the level set by the Sustain parameter. If Sustain is not zero, this means that as long as the ‘gate’ is open (key is pressed uninterrupted) the sound will continue (at the level set by the S).
With all sliders at minimum, start to just raise the Sustain slider as you play. Notice that you don’t have to retrigger a Key to find out what this slider is doing (as you did with Decay)... the Sustain slider determines the level of the signal when a key is held.
But significantly it must wait for the Decay stage to be finished with the signal... the greater the distance between Decay and Sustain, the longer the envelope will take to complete. If Decay is higher than Sustain the sound will die out. If Sustain is zero, the sound will die out even if a key is held. If Sustain is not zero, the sound will eventually reach that level and remain there until the key is released.
When Sustain is higher than Decay, by definition, there was no decay in signal. So important in setting up any ADSR is the relationship between the D-S. You often hear a description of an “organ envelope” as one that has a fairly immediate Attack, and a fairly immediate Release. There is no Decay parameter and Sustain is for as long as the gate is open. On some analog synths you will find the very simple AEG with just AR (Attack and Release). You can understand when describing musical instrument’s Amplitude behavior having the (initial) Decay parameter and the Sustain (level) parameter help to better detail how the loudness changes during a musical sound.
A plucked string is a fairly immediate Attack, there is definitely a Decay in Level before the body of the note sustains, and the Sustain Level will definitely be zero (all hammered, struck or plucked envelopes are percussive... and that means vibration stops eventually on its own), eventually.
The Release parameter, like Attack, is a Time parameter, and it takes over at anytime when the ‘gate’ closes (the key is released). If the sound has already reached zero energy caused by either Decay, Sustain, or both, the Release goes unheard. Even if the sound has failed to complete a previous stage, it does not matter, as soon as the gate closes, Release forces the final return to silence.
Envelopes can close down in legato play situations....and behave exactly like a big slow one-time LFO. That’s exactly what an envelope can be described as... the LFO automatically repeats where the envelope is strictly, in this case, a “one-shot” movement. The sound is allowed as long as the envelope supports sound, once it completes its tour, no sound can happen until a new envelope is triggered, by a NEW Note-on event.
Do you mean that this changing sound over time is because I'm playing Legato and the Envelope doesn't have enough time to reset? I have thought about that too, but I'm actually playing Staccato in the recording I posted before. Here's the Midi Track: https://imgur.com/ooHh7NT
The Release time of this patch is actually very short and when I stop the Midi track and give it 15 seconds to 'rest' and start playing it again it starts off playing with the strange changed sound again. Because it didn't go back to what it should sound like it took me a little time to figure out the sound had actually changed. Only when I went back and listened to the recording I realized I wasn't mistaking. I could not get it to sound how it was supposed to in any way. (Except for rebooting the Synthesizer)
I also have this changing sound problem with a polyphonic bass patch that I use when playing some Deep House song (by hand). There are plenty of rests in the bassline. And sometimes I don't play for a few seconds, then I play again and still the sound has changed. Just like with the other patch it's not because the Envelope doesn't have time to reset completely.
Here's the changing polyphonic bass patch: https://soundmondo.yamahasynth.com/voices/19922
The strange thing with this patch is usually if I wait long enough it goes back to normal. But that's at least a minute.
I appreciate you went through the trouble explaining how envelopes works, it's a great explanation and I wish somebody would have explained it to me that well a year ago when I didn't fully understand EG's, but nothing in there was new to me. I had experience with synthesizers (mostly VST's and VCV Rack) before I bought this Synthesizer. I already knew that playing Staccato, changing to Polyponic or shortening the A/D and R times can help when playing quickly gives me problems. 9 times out of 10 I just set it to Poly as that means I don't have to change the sound.
I really hope I'm understanding you correctly, I think we're both not Native English speakers and sometimes translating something twice can cause some issues.
By the way I don't think you ever addressed the Long-Attack / Skipping Notes problem in this last post.
Neither did you address the fact that this problem also happens to other people and that it only started to happen since the 1.30 firmware was installed.
(This was mentioned in the other topic about this problem that you stopped responding to.)