I sent my MUSE back to MOOG after a relentless tuning issue. It came back a couple of weeks later with
a new main board installed and firmware updated. I have had it back now for 5 days and still, the same tuning
problem. It’s not limited to one OSC. It’s both at the same time. It has gotten slightly better over the past 5 days but
is still unacceptable. There is no way I can play it live like this. During warm up, it drifts out of tune every 4 to 5 min. After about an hour, it drops out of tune every 10 to 15 minutes. Very frustrating to say the least! Other people are complaining
on FB of a similar tuning problem or other tuning problems. How does the mapping improve the tuning over time and how long will I have to fight this thing until it decides to stay in tune? I understand it has to warm up but still, I shouldn’t have to fight it this much. My OBX8 and Prophet 10 never have this problem. I don’t want to return it and if I did my store has a 15% restocking fee and a 2 week time limit! I love the sound, but this is ridiculous.
Update: Tuning is getting better after about 10 days but still needs to improve. Much better after bring on for a couple of hours. Still fighting it the first hour but not as much.
Do you have a case #?
I‘m also experiencing strange tuning issues.
From time to time, OSC 2 of some voices go out of tune. Not just a few cents but up to 5 or 6 whole notes. Quick tune sometimes fixes it, but sometimes the voices are off again but on different notes.
I would suggest reaching out directly to our support channel for assistance. They can help get you up and running!
https://moogmusic-help.freshdesk.com/en/support/tickets/new
Yeah I mean it’ll drift until the internal temperature stabilizes. It would be fun if they did something like slowly pulse the tune button led until the temperature rise plateaus or something like that. They’ve already got the code to determine this point because when you do the full calibration it waits for the voice card temperatures to become stable before proceeding.
I had this issue as well. At the time my room temperature was 80F. Muse temperature was 44.5C. Autotune helped.
I can only speculate that the at the factory, when calibrating it never got to this temperature? But then why did it not effect oscillator 1 in a similar way?
Some musings…
In the autocal-verbose.log file I found “file Beginning Autocal. Start Temperature: 42.1
Running temperature insensitive cals first…” It takes my unit about 45 minutes to get to 42.1C. Tuning gets better as time goes on, seems 42.9C or 43.6C is where tuning is better. I must confess I’ve run autotune many times. I’d like to run a full autotune but the manual states “The instrument is calibrated & tuned at the factory in a controlled environment, so do not run full TUNING or AUTOCAL unless there is a significant problem that cannot be solved by other means.” Although it doesn’t define “significant”. Before I get over 42.xC doing anything that requires finer tuning seems futile, as the muse temperature changes frequently. Once warmed up the oscillator drifting seems excellent, however the frequency difference between oscillators, at 440 HZ can be .75HZ (didn’t measure the mod osc yet). Haven’t found any specifications stated from Moog. I think the quick autotune function should be able to get the oscillators closer in tune than it does, considering the drifting seems low.
I’m having similar issues, but mostly at home where the temperature is a lot more variable than in my studio. Maybe a suggested update could be to have a menu screen available that displays current voice card temps? As mentioned here, it’s in the basic diagnostics for the tune/cal procedures, so shouldn’t be too difficult.
The voice temperature is already available at the end of the tuning menu.
I’m starting to wonder if there is a widespread issue with the first wave of these instruments with relation to tuning. I’ve seen so much on this. Youtube videos, forum posts, facebook comments etc etc
I run into tuning issues mith my muse every day. Cant say how widespread it actually is. But there are users here and there sharing this experience. I hope an update will fix it soon.
What I’m doing is letting it warm up 30 minutes then doing at least two “quick tunes”. That seems to lock it in. It’s almost like the turning curves (previous quick tune data points) are not saving between sessions.
Sorry about the formatting, forum doesn’t seem to allow to add a .txt file.
I’ve been having more tuning issues now that the season is cooling.
Perhaps Moog could provide some guidance on how to tune the muse or
expectations tuning accuracy, or a better “Start Tuning” and “Quick Tune” functions.
Rarely has this been in tune to my satisfaction, then is does sound great.
The oscillators sound so that they are beating against each other,
rather they are phasing against each other. Say within .25 HZ.
Almost identical issues as someone on reddit posted “Hard questions - Moog Muse tuning”.
At A440 a 1 HZ delta is about 6 cents. These results were for playing the A440 note.
My voice 7 oscillator1 was almost always at least .5 Hz and sometimes 1.5 HZ off of all the other osc1s.
The spread of the other osc1s are usually within .4 HZ of each other. Yes I know equal temperament is off. But
playing one note in a layer and the with osc1 voice1 at 440 HZ and osc1 voice7 playing 441.5 HZ, the beating is pronounced.
With this much slob in the tuning, the detune knob seems purposeless, and no it doesn’t tighten up my tuning issue.
So I held off, until today, running “Start Tuning”,
as the manual states “…should not be run unless there are significant tuning issues.”
Start “Start Tuning” at 4:45, room temp 72F. Muse had been off hours.
I saw this in original factory autocal-verbose.log file “Start Temperature: 42.1”, which is ~108F.
This indicates to me the factory temperature was warmer than 72F, the manual states “at ~70 degrees F”.
At ~6:00 “Start Tuning” completed, last temp I saw was 39C.
As I noticed later the muse continued to warm for at least the next hour.
Test measurements with INITIAL PATCH modified by setting osc2 volume up.
First results
6:12 40.6C
voice osc1 osc2
1 440.58 439.45
2 440.59 439.43
3 440.50 439.82
4 440.99 439.86
5 440.66 439.47
6 440.57 439.55
7 440.65 439.63
8 440.76 439.39
Largest diff between osc1s is .45 HZ, beat frequency ~2.2 second.
Largest diff between osc2s is .47 HZ, beat frequency ~2.1 second.
But the following are not acceptable.
Largest diff between osc1s and osc2s is 1.60 HZ.
All osc1s are over A440 by between .50 and .99 HZ.
All osc2s are under 440 by between .18 and .61 HZ.
All osc1s are around 1 HZ over osc2s.
What went wrong? These are all speculative.
- OS has timing issues, preventing measuring frequency accuracy.
- Unit temperature has not stabilized (enough), tuning is extrapolated as synth warms.
- Hardware is farther out of tolerance than the software can correct for.
- Software needs to provide handling greater hardware intolerance.
- The software was hastily shoved out the door, without testing on enough units.
Could osc1 be manually tuned down .5 HZ and osc2 up .5 HZ, yes.
But as time continued the muse continued to warmup.
6:46 40.6C 6:55 40.6C
voice osc1 osc2 osc1 osc2
1 440.56 439.45 440.47 439.46
2 440.56 439.56 440.64 439.55
3 440.54 439.95 440.61 440.01
4 441.15 439.92 441.11 439.94
5 440.72 439.48 440.78 439.53
6 440.65 439.60 440.73 439.60
7 440.66 439.68 440.64 439.70
8 440.82 439.38 440.79 439.39
Notice the .4C temperature jump.
7:01 41.0C 7:21 41.0C
voice osc1 osc2 osc1 osc2
1 440.51 439.50 440.44 439.56
2 440.63 439.60 440.60 439.58
3 440.63 440.01 440.52 440.01
4 441.10 439.94 441.03 439.87
5 440.83 439.53 440.80 439.53
6 440.75 439.65 440.73 439.65
7 440.60 439.72 440.63 439.69
8 440.89 439.37 440.82 439.41
Notice the .1C temperature drop.
Here I did a “Quick Tune”
7:32 40.9C 7:42 40.9C 7:52 40.9C “another Quick Tune”
voice osc1 osc2 osc1 osc2 osc1 osc2
1 439.52 439.36 439.49 439.38 439.39 439.37
2 439.51 439.49 439.48 439.53 439.37 439.44
3 439.58 439.26 439.48 439.25 439.42 439.22
4 439.78 439.59 439.77 439.57 439.63 439.62
5 439.37 439.38 439.45 439.43 439.33 439.41
6 439.47 439.28 439.56 439.28 439.38 439.33
7 439.85 439.48 439.87 439.43 439.85 439.43
8 439.56 439.64 439.61 439.65 439.50 439.65
Notice here is the first time after “Start Tuning” the osc1s and osc2s are closer,
however, all oscs are off A440 by about .5 HZ.
Notice the .1C temperature increase.
7:58 41.0C 8:10 41.0C 7:52 41.1C
Quick Tune
voice osc1 osc2 osc1 osc2 osc1 osc2
1 439.49 439.36 439.44 439.40 439.47 439.45
2 439.52 439.49 439.46 439.56 439.49 439.57
3 439.48 439.23 439.40 439.25 439.39 439.21
4 439.68 439.60 439.68 439.68 439.71 439.68
5 439.44 439.40 439.41 439.39 439.40 439.40
6 439.41 439.29 439.37 439.35 439.42 439.36
7 439.83 439.42 439.87 439.47 439.82 439.47
8 439.58 439.65 439.51 439.61 439.48 439.72
What can be concluded?
“Start Tuning” must be followed by a “Quick Tune”?
After “Quick Tune” voice7, osc1 is always the highest outlier.
After “Quick Tune” voice3, osc3 is always the lowest outlier.
Oscillators 1 and 2 must be manually tuned to A440.
I am not satisfied with the muse’s tuning.
I don’t think anyone is. Hopefully the next software update fixes this.
So I just got my replacement Muse, and while my hardware issues are now resolved (VCA bleed, USB midi not working, weird clicking and popping while swapping voices) the very annoying tuning issue still persists. Sometimes it will drift out of tune within minutes. Quick tune resolves the problem but if there is any kind of temperature change at all (ie, turn off the synth and turn it back on) it’s out of tune again. I’m starting to think this is a software bug because it wasn’t anywhere near this bad on 1.10. I updated to 1.20 and bam huge tuning problems.
Hi Moog and All, I have tuning issues. Read somewhere unconnecting everything during power on may help. It doe but only sometime. I warm up and quick tune too. Besides inconvenience sometime it simply does not work time after time. Is there a better remedy? Is this something that can be fixed at software level or I return Muse to store?
Try doing a full tune from a cold start. Like leave it off overnight and then in the morning go straight to the tuning menu and do a regular tune (not quick tune)
And see if that helps. Also I’ve noticed the tuning gets more stable over time as the oscs break in. Maybe I’m imagining it but it seems like after some oscillator exercise it got more stable. Perhaps it needs to spend a certain amount of time at it’s nominal temperature for values to stabilize on new components. Could be completely false but that’s my theory.
This is a guess but I believe each time you do a quick tune it saves the various offsets for whatever the current voice board temp is into a lookup table, which it then uses in real time to compensate for varying temperature. So, as you perform more quick tunes it saves more data points and so tracks better with temperature changes without user intervention.
To clarify, what I did was run a midi file of toccata and fugue into the synth, ran it for a few hours, and this stabilized the tuning. This was after doing a full tune (not a quick tune) that morning from it being cold overnight. The reason for this was I was tweaking my patch to try and get as close to a pipe organ as possible and then record the results but what I noticed was the tuning got better over time, and it stayed better. Even after a cold start.
So my supposition is that thermal expansion occurs on the components which deviates the values, and at a certain point no more expansion occurs as the components are “broken in” and then the tuning becomes more stable. Some evidence I noticed of this was some voices that were not as accessed in the round robin when I was playing other things were wildly out of tune but the more common ones were not. This midi file worked EVERY voice so maybe this helped with stability and consistency of tuning between voices.
Or it could be as you say and the software is adapting. But being that it can’t really even run the delay without glitching, and the cpu can’t really handle a lot of mod destinations without crashing I doubt it.
Also I kinda thing Quick Tune is worthless because it tunes for like 30 minutes and then drifts again.