Hi,

Sorry if this has been addressed, but I haven’t seen a fix.

I’m a keyboardist for a LONG time and a studio owner, and I have had all of the Motif’s in my studio.

The XF does NOT have a professional gain level, and I’m not sure why. It is set to +6 as a default and can go no higher in the Utility section, but there’s no other parts of the “gain tree” in the XF that I’m missing, and it does not compare to any of the other boards in my setup, which includes a Hammond, a Roland, a Dave Smith, an Andromeda, a minimoog, several guitar processors, and a Virus.

I just cannot get a pro level output from this thing no matter what i do without getting severely distorted because of “overgain” on the mixer side.

I really hope this is a firmware glitch that can be remedied, because otherwise it’s a pretty big dealbreaker as far as professional use goes.

OR, do I have a defective unit? I’ve never had this problem from any professional keyboard that I’ve ever used…

Any help?

thanks,
Jason

here we go again ;)
wait up while i get myself some popcorn

Are you using the Firewire port?  I am having the same problem when I use FW.  The recorded waveform in SONAR looks OK as I record it, but when I hit “Stop” and paly it back it is way over 0db and sounds like digital distortion.
Was wondering if my FW card was bad.

Question 1 - are you indeed talking about analog outputs or digital output, and if digital output, are you talking FireWire or spdif?

Question 2 - are you running into a mixer board first, or directly into
an audio interface?

Question 3 - are you using a DAW (Cubase, Pro Tools, etc.)?

While you may find the overall level lower than your other keyboards, have
you stopped to consider that the other keyboards you have are too hot in
level? I have a DSI Prophet 08 and that is an overly hot output - I
have to pad it down on my mixer inputs if its volume knob is set to full.
Korg M3 - same thing. Roland V-synth version 1 - same thing. For me,
the Motif XF actually has a reasonable level (plenty for my studio).

If you are recording digitally, your average level going in should be
around -12db, with peaks no hotter than -6db, from 0dbfs. At 24-bit, you
have a huge amount of headroom - -12db at 24bit is the equivalent of
full dynamic range at 16-bit. You simply don’t need to record hotter than -6db at 24-bit.

Anyway, if you have trims on a mixer, you can adjust your gain structure
that way to compensate for what you think is a low level.

DavePolich - 30 October 2010 03:18 PM

While you may find the overall level lower than your other keyboards, have
you stopped to consider that the other keyboards you have are too hot in
level?

ha ha ha . Now is Motif VS the rest of hardware synths ?. And Motif wins VS all ?.. That’s ridiculous. So after buying Axxe and Synth Extreme (demoed the demo songs from synth extreme).. i noticed that Dave Polich have a strange “style” to cut down the volume and the intensity of sound impact .. So the sound become somehow small ,weak, anemic ?… I don’t know exactly how to describe… Also a lot of Motif factory sounds (maybe 70%) are cuted down in all aspects.. so the factory waveforms sounds much better alone compared to those presets factory sounds

The XF does NOT have a professional gain level

Apparently ‘louder is better’ lives on.

And, CrocAUS, how’s the popcorn? :)

I checked out one of these units at guitar center.  I did not notice any significant different in output from the fantom G that was near it.  And this is with speakers and headphones.

So something seems wrong here.  If you are using the analog inputs and if it is not at a high volume, then you should be able to turn your mixer up without distortion.  It is only if your volume output is too high (hot) that you may get distortion from your mixer without an attenuator.

If you want to go super-professional - put it into a pair of DIs with volume controls.

So something is missing here.  Good luck.

I record piano solos mostly, using digital output, so I can understand from where you are coming.
If you mix up a full song, you will find the levels are correct; you typically will not need to amp them on the mixer end and may even get clipping before it even gets there.  Of course you can decrease the output level for this.

Increasing the output level is not as easy, but can be done.

#1:
If you are in Song Mode you can easily set the levels of each track to 127 (from the default 100), but when in Voice Mode I personally have not found a way to do this on the device.  The Cubase AI 5 VST Interface does, however have this knob, so you CAN increase the part volume even in Voice Mode, and it helps.

#2:
You can use the part equalizer to get an extra 12 decibals with the following values:

Low:  2.00k +12.0
Mid:  5.98k +0.38
High: 14.0k +12.0

This keeps the equalizer line flat and raises it 12 decibals across the whole range, with the result of simply increasing the volume of the part.

#3:
There is also a Master EQ which can be used for the same thing, but if you want to actually use your equalizers as they were meant (not just as a volume boost), you should use the Master EQ for your normal equalization (especially since it gives you more bands).

To be honest, a +12 output should be available because not everyone uses these keyboards for power songs, full mixes, etc.
Soft piano solos are quite common, and a +12 output signal would go a long way here.  Luckily, however, there are workarounds.
I’m not exactly sure why this output level is missing to be frank.  If it is just a matter of adding a new preset multiplier to the system and a menu item to select it, I don’t see it could be justified not to add it simply because “+6 is suitable for most situations.”

L. Spiro

motif8mine - 30 October 2010 08:38 PM

The XF does NOT have a professional gain level

Apparently ‘louder is better’ lives on.)

Apparently there is a confusion between gain level and sound pressure level.
Louder helps your music to be heard better.
Hotter helps you mix your music without noise.

In my mind “professional level” means hot enough signal that rising gain levels when mixing doesn’t bring too much noise above the hearing level. In some cases noise isn’t a problem at all. Fender Stratocaster guitars need some noise to sound like stratocasters.

I think he is trying to say that when you record from the device digitally (SPDIF) you get a signal that is so weak you have to amplify it externally, which causes a lot of distortion.

If you record a piano solo and amplify it with Audacity, you may need to amplify it by 13 or so dB, and after doing so you can hear very very clearly some static over the piano.  It’s normal for 16-bit signals; boosting a low-volume signal just makes the lossiness of 16-bit integer truncation easier to hear.

If you can boost the signal enough from the device so that it does not need to be amplified in Audacity at all, you get a crystal-clear piano recording that accurately reflects the quality that the device can deliver (which is the best-of-the-best).

Same reason it is always best to equalize from the device rather than to post-equalize in an audio editor.

Although I found solutions to this, it would certainly be more intuitive and easier if the device could simply deliver output signals at 12 and 24 dB gains, instead of just 0 and 6 dB gains.

I may not specialize in audio programming, but as a CTO by day (musician by night?) I am all-but-certain that the implementation behind the output gain is trivial enough not to warrant any excuses for its absence.

This is too much to ask?  Really?  (My algorithm may be off as I am a game programmer, not an audio programmer, but I do know that, whatever the actual equation is for output gain, it is nothing more than a matter of changing a single number for different gain levels.)
The fact that my EQ hotfix works proves that there is no hardware limitation here.  Someone simply decided that 12 dB and 24 dB gains would not be useful, and he or she was wrong.
I expect such gain features in the next OS update.

L. Spiro

I think he is trying to say…

I think what we are trying to say is that this question has been proffered dozens of times. Bad_Mister (and others) have set forth the reasons why the outputs are what they are. How to address them at voice level edit and part output, and other solutions.

In fact, he did it recently here: http://www.motifator.com/index.php/forum/viewthread/442540/

See also: http://www.motifator.com/index.php/support/view/midi_velocity_and_audio_record_levels

(First I must say) I am drunk now so forgive any ignorance (timezones in Japan and such, drowning away romantic blues, you get the idea).

I read the article(s).
But why so much for what could be done by just having an output lever that reaches 24 dB?
Being drunk, I may easily have missed some important information (if I did, no problem, my apologies), but what I saw just says that you have to edit voice parts and so forth to get higher output volumes.
Basically what I said myself earlier (I mentioned the EQ settings but I have personally also resorted to using the part volumes too).

I saw nothing there that did not say what I said already.
#1: Edit the voice part volume.  I don’t know how in Voice Mode on the device but I do know it is possible somehow since it is an option available in the Cubase VST 5 editor.

#2: Edit more things related to voices.  EQ, Master EQ, or the levels for each wave (of 8 maximum), etc.  These all come off as hacks.  I can come up with mathematical solutions because I am a more-than hardcore programmer but I suspect this is not so intuitive for most musicians.  Such as the original poster.  Dare I ask, why not provide a simple and intuitive way to increase the output levels?

No really, what reason is there for not including an output gain of 12 dB?  I already showed how simple it is from a coding standpoint!
And your article(s) solidify my point.  They claim the audio engineer wants 2 recordings, one at average and maximum levels.
If I am to provide someone with these two recordings, would it not be simpler math to scale my voice volume(s) for the intended level I want to provide given that my base output level is +12 dB?
Put another way, that is, a musician who has access to +12 dB output gain could more easily and intuitively provide his or her sound engineers with the recordings they desire by increasing the output gain and scaling down part levels by a fixed amount.

And as an exceptionally top-tier programmer myself, already a published author, the only thing holding back my (keep in mind I am romantically drunk now) rage is that I bear in mind my lack of specialized expertise in audio algorithms.
If I am not to expect Yamaha to provide us with a +12 dB output gain in the first OS update for the Yamaha MOTIF XF, please explain why it is so complicated to add such a feature.  Because with what audio-engineering knowledge I have (which even when drunk I am fully open to admitting is limited) I still assert that adding such functionality is nothing more than a matter of adding a multiplier and the menu interface for selecting that multiplier.

I could be totally off base in my drunken state right now, but all I have seen so far are hacks to get the desired result.  Even after your article(s) I do not see why simply adding a +12 output gain is less intuitive.

L. Spiro

L.Spiro, sorry to burst your bubble but you are so completely off-base about gain that I almost don’t know where to begin. But I’ll attempt to
begin anyway.

First, there is no need to provide +12db of gain or +24db of gain, period.
That’s like saying every guitar needs a set of dual hi-gain humbuckers because every guitar needs to be “louder”.

All the original post mentioned was that the Motif XF was not as hot in level in comparison to other gear. I replied that the other
keyboards are likely too hot in output, which is true. All one has to do is use
gain structuring to match levels - that is, turn the other things down
to where they match the apparent level of the Motif, then turn the overall
monitor level (speakers) up. Problem solved.

If you’re a “master programmer"and know as much about audio as you say
you do, then you should know that at 24bit resolution, you reach the maximum dynamic range of 16-bit when you hit -12db. You simply don’t need to record at peaks close to 0dbfs at 24bit at all, ever. I recommend you read Bob Katz’s excellent book Mastering Audio to gain a better perspective on this subject, as he goes into great detail about it.

I do a lot of recording myself. I don’t need more level in most cases, the exception being a single-coil Strat or Telecaster D.I.’d into the
board, then I might need a little bit of gain.
I need average levels of -12db from 0dbfs, with peaks hitting around
-6db. That’s plenty of level. For one thing, if you record at average
levels hotter than that, then compressors, limiters, even EQ’s have
nowhere to “go” with the signal. There’s no headroom, it’s all been used
up.

If something isn’t loud enough, then turn something else down. That’s a
quote from bad Mister himself, it’s one of my mantras, and Bad Mister is
no stranger to recording and audio.

I’ll give you a good example - when I was the programmer on the Michael
Jackson “This Is It’ gig, what was the number one request from the
front-of-house and monitoring engineers? “Please turn down the level of
the Motifs”. I set them to 0 db, instead of +6db, and that’s where they
remained. +6db was just too hot going into the FOH and monitor consoles.

The argument that there needs to be +12db of gain on the Motif XF, or
any other keyboard for that matter, is completely without merit. The solution is turn the other things down, and turn your listening level up.

It might be less inituitive or not, but that is not the issue. The issue is one of safety. Improper use of levels combined with the technically challenged and a pair of headphones could lead to permanent hearing loss or worse. So if you are inconvenienced by having to learn how to use your equipment (so be it). If you want more level learn how to get it, you are given more than enough - if you are plugging into pro-level gear you will have no problem.

The whole situation is really just a matter of where you are at with audio. Taking an example of the real world (still the best measuring stick). If you think all musical instruments have the same output level (in SPL, sound pressure levels) you should get out more. Take flute and an electric guitar, a violin versus a trumpet. The request to have the flute sound as loud as the trumpet is well bizarre. How does this translate to instruments that have been sampled?

To say the “gain level output of the XF NOT “up to pro level"?" is laughable. When is it not up to pro level? When you are playing softly, when you are playing 1 instrument, when you are playing 16 instruments, just when?

What’s so-called “pro level” - we’ll assume you mean +4dB output. The XF is +0dB or +6dB out… why are there these two settings because this is the typical-use output level. Not the every instance output. The quest for the “my mix must be the loudest” thing - is for the technical n00bies to persue. Get over it, the sooner you do, the better and more relaxed your life will be. If you are making good music and people want to hear it, they will turn it up. If you must thrust it in the face of your audience all the time perhaps it is some insecurity about the quality of what you are producing.

Check your velocity - recording with no knowledge of your velocity is to really go forward unprepared to deal with the results you get. Remember I listen to people’s sequences for a living. I get to hear more diversity from more people than most of you. Trust me when I say some people have their speakers up so loud they don’t realize that they play with low velocity output, then they wonder why when they transfer their data to audio it comes out so low in actually level.

If you don’t know physically how much energy it takes to reach 127, you are working with only a partial deck of cards. Arghhh, I’ve said this all before, where’s my popcorn...!!!

Curse my sleeping problem; even drunk I could only stay asleep from midnight to 4:00 AM.  Sigh.

DavePolich, I may have hinted at my programming experience, but even drunk I knew well enough to point out that I do not know such a great deal about audio and its algorithms.  I make game engines and the games on top of them (ever play Leisure Suit Larry?).  Never yet had the chance or need to study audio algorithms and I hope to garner some insight from threads such as these (even if I seem only to be arguing).

But to my audio-deficient mind my core complaint still seems unaddressed.
#1: You said I should turn other levels down instead of trying to turn the Yamaha MOTIF up.
I am not arguing this at all; this is completely fine if I were playing in an ensemble.
But this whole strategy does not apply to me.  I am recording piano solos primarily, so there are no other levels to turn down.

#2: Turn up my listening levels?  Not a problem if you are just talking about turning up my speakers.  But I don’t think that applies to recording from an S/PDIF line into Audacity.  If it does (meaning, if there is a setting in Audacity to add gain to its input while recording), how is it different from recording in Audacity at the regular output level of the keyboard and then using the Amplify tool to add the gain I need?

I record in Audacity using a floating-point internal format to avoid integer truncation as much as possible, which means the buzz created by heavily amplifying my piano solos is caused by integer truncation happening before the signal reaches Audacity—it seems to be sent over S/PDIF in either 16- or 24- bit integer format, and this is lossy particularly at low levels.

In the case where I am not recording and have the option to just turn up my speakers, the buzz is still there but just harder to hear, not causing much of a problem.

Okay, let me give you a real-world example, since that is still the best measuring stick.
I recorded this piano solo.  Digital out from the Yamaha MOTIF to my S/PDIF in on my ESI Juli@ professional audio card. http://ixbtlabs.com/articles2/esi-julia/

And the song I recorded:
Hungarian Rhapsody No. 2

I don’t think my velocities are a problem here; this song has a great range of dynamics and gets especially loud in the second half.

To get this volume for the final recording it was not enough to slam the keys on the keyboard.
It was not enough to turn up the track volume to 127 and use a +6 output gain.
No, even after all of this, the original recording in Audacity was still in need of an additional “Amplify” of about 4 dB.
Because the amplify is so small you won’t be able to hear the buzz it creates in this recording, but some of my softer songs needed to be amplified by around 13 dB and the buzz was simply unacceptable then.

Now, the advice given so far has been:
#1: Hit the keys harder.  I hope the song I chose above shows that this does not apply.
#2: Turn down other levels.  What other levels?  There is just a piano.
#3: Turn up the listening volume.  Fine if I am just wanting to hear myself play.  Not an option to my knowledge (and I would love to be wrong) when recording the way I do.

So you can see now, I hope, how I would feel that the core issue was not addressed, and why I would heavily appreciate a +12 output option.
To record my softer Gnossiennes 1 (Lent), I had to use my EQ hack to boost the piano signal before it left the device or else be stuck with a final product with the same volume but significantly more buzz.

If I am doing something wrong or missing something, do please tell.  As you can probably tell by my two pieces above I am just a hobbyist and there may be something I don’t about my device or my software.  I would love nothing more than to be rid of my ignorance.
An example of something I literally do not know: DavePolich mentioned some reasons why a +12 would leave no room for compressors and EQ’s etc.  If I just do a +12 output boost, I may suddenly get clipping and etc., but if then I turn down my master volume, part volumes, or whatever, does this “no headroom” point still apply?
As it is now, a +6 leaves me unable to reach the final levels I need.  A +12 may take me over the levels I need, but I also have the option to turn down some parts to get the level exactly where I want it.  Is this wrong or what?

Oh, also, Bad_Mister, I feel that only your first paragraph applies to me, as I agree with audio levels of other devices being different and I do not feel that the MOTIF XF has non-pro levels, as the topic creator does.

Regards,
L. Spiro