Recording Tips from Bad Mister: How to increase the GAIN of your MOXF PARTS prior to recording AUDIO to your DAW
Here are some things that can help you with becoming a MO-XF "recording engineer" (the person responsible for capturing proper record levels when recording audio signals to a DAW like Cubase AI). Built-in to your instrument is the ability to communicate with a computer, both MIDI data and AUDIO data (bi-directionally) via a single USB cable. As an example, you can record to the MO-XF internal sequencer (MIDI data), taking advantage of the many features like "Direct PERFORMANCE RECORD"
and the thousands of arpeggio Phrase Types. With its powerful editing and remixing tools, the sequencer of the MO-XF can be very useful for music composing. Once you are happy with your tracks you can embark on recording your composition to your DAW as audio. After all, audio is the finished product. It finalizes your work in a format that can be played by and distributed to the general public (your friends and fans around the world). The world of MIDI sequencing has one set of rules and the world of AUDIO recording has another. That is what we will be dealing with here.
This workflow, recording MIDI tracks to the MOXF's internal sequencer and then rendering audio tracks, as necessary, becomes a very appealing method. While your project is MIDI data, you have a full complement of editing tools to not only perfect your performance but to assemble your musical arrangement. You also get to study and modify "the sound"
you are generating, per instrument, prior to committing to it. When you render audio you have committed the musical performance to a level of permanency.
The dual stereo outputs and the ability to isolate a Part or Parts to a discreet audio pair allows you to "Multi-track Record" (track-by-track) your audio to your DAW taking advantage of all of your MOXF's advanced modeling Effects, Equalizers, and Phrase creation features.
Of course, transferring or recording MIDI Tracks in your favorite DAW is another appealing workflow that opens a different set of features, functions and opportunities. It is really a personal preference. We highly recommend you spend a bit of time investigating different workflows, because only you can decide what works and inspires you. And being able to command several methods is an important part of true creativity. You will find yourself using each method for its advantages and that decision comes quite naturally when you have experience working several different ways.
By default, the 16 internal instrument Parts are assigned to USB 3/4 (when 4CH mode is selected). You can isolate any internal Part for discreet audio transfer by setting the Output Select parameter for that PART to USB 1/2. Literally, you reroute the Part by assigning it to ride the USB 1/2 bus to your computer DAW. Multiple Parts can ride a bus to the DAW Track destination. You isolate tracks in this manner when you need to process them separate or further enhance the signal within your project. MIDI causes the MOXF tone generator to output audio, we route that audio signal and capture it on a DAW audio track. You would/could then mute the MIDI event stream, keeping it as a backup, in case you want to redo the performance documented in the audio track. The MIDI data gives you an ultimate level of "Undo" in that it will allow you to un-commit to the audio data because you can go back and make further adjustments.Working with MIDI
Recording data as MIDI EVENTS
does not involve you monitoring any kind of metered levels, however, when recording audio you will be concerned with the record level as indicated on a meter. This is because with AUDIO data there are bad, bad things that can happen to your signal if you overdrive the system. Audio data can distort - MIDI data cannot be heard, it only 'represents' the music so there is no distorting MIDI events by sending too much signal in, as there is with AUDIO signals. Knowing how and, more specifically *where* to address output levels of your MOXF Parts is what this article will address. Understanding what LOUD
is versus what GAIN
is, is where we will start our discussion. We will use an analogy of the recording studio MIXER with faders, the MO-XF PARTS are your musicians. The musicians, are actually the VOICEs that you select to play via the PARTS 1-16. A Voice is placed in a PART, like a musician is connected to a channel of the studio mixer. Each VOICE outputs signal at a particular volume - the signal arrives at the MO-XF mixer where you have another control over the volume (the mixer's PART fader) before you send the signal to the audio output.
We will learn how to adjust the volume of the VOICE, and learn not
to rely just on the mixer's (PART VOLUME) channel fader. If you have ever turned up the PART VOLUME and and found you still are not getting enough output - this article is FOR YOU!!! There are two separate controls that will affect the overall OUTPUT LEVEL of each MOXF sound used in your mix: The VOICE VOLUME and the PART VOLUME. It is the same as each musician has a control of their Volume before
their signal arrives in the studio's mixer.Volume is different from Gain
First, let's define the word "LOUD" as an analog word expressing mucho volume. It is a separate thing from the record LEVEL (record level, we'll refer to as "GAIN"). The main VOLUME slider is an analog control for listening VOLUME and has no bearing on your audio's RECORD GAIN. You can set the main VOLUME slider to 0 (minimum) and audio will still be sent to the USB1/2 and/or USB3/4 digital audio outputs. The digital (output level) equivalent for the main output is found as follows:
Press [UTILITY] > [F1] GENERAL > [SF1] TG (Tone Generator) > VOLUME = 127
(127 is the default setting) and this should be left here. While we are talking about the digital output:
Press [SF2] OUTPUT
Make sure the USB1/2 and USB3/4 are set to +6dB... (Also the default).Factors that affect Level
Second, we will need to grasp the role of "velocity" – or how strongly you are striking the keys. Velocity has an influence on the resulting audio signal - Of course, I'm not telling you anything new here. Depending on the sound that you are recording and depending on how many signals are being recorded - all these things affect the results you get.
For example, if you are sequencing 16 PARTS simultaneously, they will have a cumulative output that is greater than any single PART. To fully understand how sound level increases requires an understanding of logarithms. (In other words, 1 sound playing is not doubled in level by adding a second sound playing the same thing - it would basically take 10 sounds playing the same thing to double the perceived volume) basically and simply put: sound increase is not linear. Far from it.
If you have recorded your tracks as MIDI data to the MOXF sequencer, you can view your Velocity output on the TRACK "Event List" - from the main [SONG] or main [PATTERN] screen: Press [EDIT] and select a Track -. Velocity is 0-127. If your music tends to be in the 40-60 range, well, anticipate low audio levels as a result of these tracks when transferred to your DAW as audio. This is also pretty obvious. And on the other hand if you have too many 127's you probably are not going to enjoy your mix either. You want to be able to play your keyboard and have a fair idea of what MIDI velocity you are sending. If in the past you never paid any attention to your Velocity values, you should begin to recognize that they can greatly influence the results you get when you attempt to record/render audio based on what you have played.
We should mention, changing Velocity must not be done without considering that it can dramatically affect the resulting performance. It is not an "engineering" decision... It is a "musical" performance decision. This is because it changes the emotional impact of the musical output. Altering Velocity can change what Element (Oscillator) sounds. Imagine a slap bass or other articulation that can be determined by velocity. So a change (edit) to the documented Velocity value will affect loudness and record level, it additionally affects the musical result as well. Velocity change can cause a drum phrase to add a a Crash cymbal where one did not appear before. Edits to velocity should always be a musical decision. While velocity is a contributing factor to record level it is not strictly speaking an engineering decision, and therefore you should always consult your musical side before altering Velocity. Gain Staging your Audio
Third thing, we recommend and this is the critical step that most new MOXF "recording engineers" will miss. When you are in a recording studio (real world now) the engineer has several tasks to perform - among them:
1) Ensure that the recording device is properly calibrated for incoming signal
2) Ensure that the musician's signal is arriving at the mixing console with enough output so that it can be recorded properly
3) Know what to do when it is not.
4) Make sure the drummer knows where the bathroom is located! :-)
How then, on the MOXF, can you ensure that the individual musician's output level is enough to be recorded properly as audio?
Before we answer the question, let's define why this step is vital. When the recording engineer in the real world studio puts a microphone out to record something or connects a direct signal to the console, one of the most important tests that the engineer does is the "run through" - they will ask the musicians to play what they are going to be performing - the engineer's ears are glued to the sound quality, the engineer's eyes are glued to the meters. What they are looking for is an indication of the DYNAMIC RANGE
of the performance. That is, they are listening/looking for the average level, the lowest level and the highest level that this performance reaches for each Track. Checking the lowest-and-highest audio level points will allow the engineer to ensure that the highest level does not distort, and then they can determine if this makes the lowest level too low to be heard in the context of the particular composition.
But during record, typically, it is the highest gain (peak) that is of MOST CONCERN. DISTORTION is not allowed, period. You want to give your music plenty of headroom - that is, you do not have to maximize everything to the ultimate amount. You must leave room for transient peaks - those quick percussive strikes that cause levels to spike. What you want to do is have a nice full dynamic range - that is, the soft to loud ratio should be used and its impact felt. This allows the music to move the listener. The softest moments need to be far from the loudest moments - much like the darker black dots (pixels) in a high definition television increase the clarity of the color picture - it is the contrast that is stunning to the eye... A similar thing happens with sound versus silence. Peaks need room to breath - do not slam all your audio to maximum. Learn by what sounds best, not what sounds loudest! Recording without leaving some headroom can result in a harsh sounding digital recording.
If the recording engineer adjusts the input GAIN, and still is not getting enough LEVEL on the meter (GAIN) from the channel, they will ask the musician to increase the OUTPUT at the SOURCE. Picture this situation in a real studio, if you are not getting enough signal you would ask the musician to increase the output gain at their instrument. You have a microphone in front of an amplifier and you cannot get enough level at the mixer, you ask the player to increase the gain at the amplifier! The mixer level is not the only thing influencing the record gain... The sound source itself can contribute and does contribute its own output level which you can control separately.
Here is how you do the equivalent in the MOXF (and this is important to learn how to do when recording audio with your MOXF):To increase Audio Output at the Source: Voice Volume vs. Part Volume
From the SONG or PATTERN mode
Press [F1] VOL/PAN (Volume/Pan)
Here you can see the PART's channel Fader (this is the equivalent to the studio's mixing console channel fader) This is considered PART VOLUME. Notice all channels initially are set to 100 on a scale of 0-127. Faders here, as on a real console, are for artisitc changes in the mix, during record they simply pass signal. And just like on a mixing console, they control the channel's volume relative to the adjacent channels.
However, if we are to change the volume of the instrument at the source we must go into EDIT on the Voice data itself and find the parameter controlling the individual VOICE VOLUME before
it arrives at the mixer. Here's how:
Using the TRACK SELECT buttons, -, select the VOICE (instrument) you want to edit
Press [F5] VCE ED > (VOICE EDIT)
This is a shortcut that can be used on any NORMAL (non-Drum Kit) VOICE to edit the original VOICE. Editing the actual voice will allow us to 'turn the signal up at the source'. If we asked the guitar player for more output signal they would simply turn the knob on their guitar or increase the output on their amplifier. This is exactly the same thing. We are going to edit the instrument.
Press [COMMON] if not already selected
Press [F2] OUTPUT
This OUTPUT parameter is the Volume of the source. In many (most) cases Yamaha has left this level at a conservative output on the scale of 1-127. So in most cases you have plenty of room to increase the output level of the VOICE (at the source). And this is the first place you should increase the individual signal. (Do not use the input gain of your computer program - that is about as noisy a gain as you can imagine and would be poor choice in increasing gain).SIGNAL FLOW
- if you follow the signal from source to destination you can follow the rule: Send sufficient signal at each stage so that the next (receiving) stage is 'happy' with the signal's level. The less 'work' each stage has to do, the better. So send enough from the source so that you can effectively mix it with other signals.
When you have increased the VOLUME of this VOICE, press [STORE]
You will be offered a MIXING VOICE
location to store your newly edited VOICE.
Press [ENTER] and store the MIXING VOICE. The new VOICE will be added.
You will be returned to your MIXING setup - Remember to also [STORE] the entire MIX as well, as now this new MIXING VOICE
has been added to your current SONG or PATTERN. This newly edited Voice will be stored with the Song or Pattern you created it for and will be automatically saved and reloaded anytime you load this Song/Pattern. It does not overwrite an internal User location, rather it is stored locally in the composition in which you made this adjustment.
Anticipated question: Why did Yamaha program most voices with such conservative levels? This is purposeful. The 'noise floor' is so very far down now in the digital domain, it is no longer an issue. If all sounds were stored at 127 it would make combining 16 sounds initially impossible.
We have big confidence that musicians understand how to TURN things UP, but little confidence that they understand that FADERS (they are called FADERS because they) are used to turn things down!!!
I'm kiddin' - it's funny but too true! It is proper to adjust the gain (GAIN is to increase) at the source, and then turn down at the mixer. The theory is: send in enough clean gain so that you can use just enough of it to make the mix work!
If you are attempting to record a MOXF sequence that is a string quartet (four parts: violin, violin, viola, cello) by knowing how to turn the instruments UP at the source, you can get plenty of GAIN without having to increase velocity or use external gain control devices or anything unusual. Simply increase the audio output at the instrument (Source): VOICE VOLUME.