Riding the storm out intro on PSR-E453

[hlawhon]

Hi all.   I am new to the forum and new to the PSR-E453.  I am mainly a guitar player but picked up this keyboard to add some variety.   One of the things I wanted to try to do was to replicate the minimoog opening to Ridin' The Storm Out by R.E.O. Speedwagon.   Does anyone have any suggestions on how to do that with this keyboard?

Thank you !

Henry

[SeaGtGruff]

Welcome to our cozy corner of the internet, Henry! I imagine that Bob (SciNote) might be able to suggest something, otherwise I'll give a listen to that song to refresh my memory about the Minimoog sound in it.

One thing I've found is that you can get some interesting results from voices you might not expect by editing the filter cutoff/resonance and envelope attack/release settings, so it can pay to look beyond the usual synth voices and tinker with the other instrument voices.

[hlawhon]

Thanks Michael !    I literally got it today (thank you UPS) and have been tinkering a little but really have not looked to deeply.   Thank you for the suggestions !    I am looking forward to getting to know what this beauty can do. 

Henry

[SeaGtGruff]

Okay, I listened to the beginning of the song on YouTube to refresh my memory. The PSR-E models don't have portamento, so you'll need to use the pitch bend wheel to imitate the portamento. And the best way to get a smooth portamento effect with pitch bend might be to create a short MIDI file that you can play. It's fortunate that the Minimoog is used to start off the song, because that makes it easier to kick off the song with a recorded passage and then let the band jump in and take over. I'll have to listen more carefully later tonight to make sure the Minimoog isn't used again later in the song.

[SciNote]

I wasn't familiar with that song by the title, so I checked it out on YouTube to hear the sound.  Check these out! ===>

https://app.box.com/s/ll8n5e79g2kmnqzwrd47ubedjcqln4z9

https://app.box.com/s/yrmbadwt8nxks5ynd0rwi99sazvyxc8v


The first sample was done "live", whereas the second sample was made by piecing together two separate recordings using my DAW (Audacity).  Here's why.  To get the sliding/portamento type effect, as Michael said, required using the pitch-bend wheel, so I set the pitch-bend range to 12 (1 octave).  Unfortunately, at least on the E433, it turns out that the pitch-bend is really made up of a fairly small number of discreet steps.  With small pitch-bend ranges, such as a whole tone or a minor third, for example, those discreet steps are close enough that you get a smooth pitch-bend/slide effect, especially when the bend is done quickly.  But on this song, the tone starts out with a "C", then slides up to the "C" one octave up, and then slowly slides down a minor third to "A" below the "C".  The trip up from one "C" to the next "C" isn't too bad, but to use the pitch-bend wheel to slowly slide down to the "A", I could hear the discrete steps so clearly that it did not even sound like a pitch bend -- it just sounded like playing notes on the keyboard -- except that the steps/notes were not exact chromatic notes, and I could not use the wheel to land exactly on the "A".  So, what you are actually hearing on the first sample is me using the wheel to slide the note from one "C" to the one an octave up, and then me actually just playing the "B", "Bb", and "A" keys in quick succession to go down to the "A".

With the second sample, it starts out the same way, using the pitch-bend wheel to go from one "C" to the one an octave up.  But then I recorded a separate track of me using the pitch-bend wheel to slide from the higher "C" down to the "A", but the difference is that I set the pitch-bend value to 3 (a minor-third) for that recording, so that the slide from "C" to "A" was much smoother than when the pitch-bend value was set to 12 (an octave).  But of course, trying to change the pitch-bend value "live" while playing that tone would be quite cumbersome.  So, I made the two separate recordings (the slide up with the pitch-bend set to 12, and the slide down with the pitch-bend set to 3), and then just blended them together with Audacity.

So, if you're playing live, you'd need to do something like I did in the first sample.  But if you're making a recording, you can use the technique of the second sample.  Or, like Michael said, if you're playing in a band (and therefore don't need to use the styles on the keyboard), you should be able to make something similar to the second sample using two tracks of the onboard sequencer and just play that onboard "song" when needed.

I came up with this sound pretty quickly, so you may be able to improve on it.  Note that I did this on a PSR-E433, so the voice numbers are likely different, but the sounds should be available on the E453.  Here are the basics of the patch:

I first hit the "Portable Grand" button to essentially reset all of the main parameters, then I selected the following settings...

Main Voice: 80s Brass (Sound 140 on PSR-E433)
Main vol: 90
Main oct: 0

Dual Voice: Fifth Ld (Sound 561 on PSR-E433)
Dual vol: 110
Dual oct: 0

Reverb: Hall3
Chorus: Chorus1

All of the other parameters of the voices, such as filter and envelope, I left as default.  However, after setting up the sounds, I used the live-control knobs to crank up the reverb and chorus to their maximum settings, as in the song itself, it sounds like there's a little bit of echo going on with the sound.

Note that for the dual voice, it is important to use the "fifth lead" sound in the XGLite collection of sounds, because, at least on my E433, most (if not all) of the "fifth lead" type sounds in the main "panel voices" automatically use the dual voice to make the complete sound.  Therefore, these sounds would not sound as intended when used as the dual voice, and if used as a main voice, you could not add a dual voice of your own without altering the original sound.  But the XGLite voices all seem to only use one voice, not two, and therefore can be combined with another voice.

[hlawhon]

Wow.  Thanks guys for the great info.   I will definitely be giving it a try.   I was thinking I had seen that the PSR-E's did have the portamento.
Guess that explains why I couldn't find it anywhere !!!   Thanks again and happy New Year fellas !

Henry

[SeaGtGruff]

Regarding pitch bend, there are a few things to keep in mind which can be helpful:

(1) The Pitch Bend MIDI message has two data bytes, MSB and LSB, so in theory there are 128 * 128 = 16384 steps from the wheel's lowest value of 0 (pulled all the way forward) to its highest value of 16383 (pushed all the way back). However, in practice a lot of MIDI hardware and software ignores the LSB value, using only the MSB value, reducing the number of steps to 128, hence each step is 128 times larger than it "should" be. I'm not sure which Yamaha keyboards fall into that category, but the PSR-E models certainly do; if you record a song using the pitch bend wheel and examine the data in a MIDI editor, you can see that all of the Pitch Bend messages have a value of 0 for the LSB-- that is, the possible values are 0, 128, 256, 384, etc., up to 16256, rather than 0, 1, 2, 3, etc., up to 16383.

(2) The Pitch Bend Range is a plus-or-minus value, so setting it to 12 semitones actually gives you a range of two octaves, or -12 semitones to +12 semitones. There are always 16384 steps-- or 128 steps if the LSB is being ignored-- so the size of each step is obviously determined by the Pitch Bend Range, and can be calculated by the formula SIZE = 2 * RANGE / STEPS. Thus, setting the range to 12 gives SIZE = 2 * 12 / 128 = 0.1875 semitones per step. That might seem small, but it means there are only 5.3333 steps from one note to the next, making the steps easy to hear if you're trying to make the change slowly-- especially because it can be difficult to move the wheel smoothly, so the change in the pitch bend value is likely to jump by some multiple of 128 instead of "only" 128.

(3) The Pitch Bend message is a channel message, and you can set the Pitch Bend Range to different values for different channels.

Taking those things into consideration, the way I'd do it is as follows:

(1) Set up the voices and effects to get a good approximation of the sound you're wanting. I'd try the settings Bob suggested; the voices numbers might not be the same as on the PSR-E433, but their names should be the same.

(2) Set the Pitch Bend Range to 6 semitones, which will give you a range of one octave, or -6 semitones to +6 semitones.

(3) Turn on the song recorder, but before you play any notes pull the pitch bend wheel all the way forward with your left hand-- and keep it pulled forward-- so any notes you play will be lowered by 6 semitones.

(4) Instead of playing a C, play the F# that's halfway between the two Cs you want to begin and end the slide with.

(5) While holding the F# note with your right hand, use your left hand to move the pitch bend wheel from all the way forward to all the way back, as smoothly as you can and at the desired rate.

(6) Keep the bent note held for a while, then release the F# key, stop the recording, and wait for it to finish writing the data for track 1.

(7) Unfortunately, there's no Pitch Bend Range of plus-or-minus 1.5 semitones, and if we were to set it to plus-or-minus 2 semitones it would be tricky to slide down 3 semitones from C to A, so let's just change the Pitch Bend Range to 3 semitones.

( 8 ) Push the REC and TRACK 2 buttons at the same time so you can record just track 2 by itself. The little "1" indicator in the bottom right corner of on the LCD screen (for track 1) should stay steady, whereas the little "2" indicator should start blinking to signal that track 2 is ready to be recorded.

(9) Press the START button to begin the recording, but don't play anything yet-- just listen to track 1 playing back.

(10) Sometime before track 1 completes its slide from C to C, push the pitch bend wheel all the way back (so it will be at +3 semitones) and hold it there.

(11) When track 1 finishes sliding up to the higher C note, play and hold the A key that you want to end on, which will play as a C and blend with track 1.

(12) Wait until you're sure the C note on track 1 has stopped, then slowly move the pitch bend wheel back to its centered position, which will slide from C down to A, and hold the A note as long as desired.

(13) Then stop the recording and wait for it to write the data.

When you press START to play back the recording, you should hear the lower C slide up an octave to the higher C, then slowly slide down a minor third to A.

To achieve the best results, what you can do then is use the FILE CONTROL button to save the song recording to SMF on a USB flash drive. Then you can load the SMF (or standard MIDI file) into a DAW, view the Pitch Bend events on the MIDI channels as automation lanes, and use the DAW's functions to edit the Pitch Bend events so their values form a smooth slope. There should be four MIDI channels in all-- (1) the Main Voice sliding up from C to C, (2) the Dual Voice sliding up from C to C, (3) the Main Voice sliding down from C to A, and (4) the Dual Voice sliding down from C to A. After you've smoothed the Pitch Bend events, save the changes back to the MIDI file. You might want to rename the file so you know what it is, such as "RidnStrm.MID" (only the first 8 characters will be displayed on the keyboard's screen, so you might as well abbreviate the name of the song file). Then you can either put the USB flash drive back in the keyboard-- or even better, you can use the Yamaha Musicsoft Downloader to transfer the MIDI song file from your computer to the keyboard's memory, so you won't need the USB flash drive to be plugged in whenever you want to play the song. You can also erase the original User Song recording, since you won't need it anymore.

I don't know how much experience you have with keyboards in general, or with MIDI editing in DAWs, but since you're just getting up to speed on this keyboard then I don't necessarily expect you to be able to do all of the above right away. Instead, I've done the first part for you, and I'll post the MIDI file in this thread as soon as I've finished the second part. But I'd encourage you to try it on your own as well, hence the lengthy explanation!

[SciNote]

Michael, I'm not sure about the 128 steps with the pitch-bend wheel.  Maybe the data/software is set up to be able to handle that resolution, but I'm thinking that the wheel itself only supports a much lower number of steps.  When I had it set to 12 (and you are correct, in that gives it a total range of 2 octaves from -1 octave to +1 octave), I heard what sounded like a LOT less than 128 steps as I was recording my samples.  It was almost chromatic (and 2 octaves chromatic would be 25 individual steps) as it almost sounded like playing the individual notes on the keyboard, but not quite, as I could not exactly hit the "A" on the way down after sliding up to the "C".  I'm guessing there are actually maybe 20-32 steps -- but I'll have to check it again.  I'm at my Dad's home right now and do not currently have my keyboard with me.  I'm wondering if the E443 and E453 have a higher resolution for the pitch-bend wheel than the E433.

The idea to set the range to 6 and to hit an F# and go from the lowest range to the highest range of the wheel is interesting -- I never would've come up with that -- but that does give you a higher resolution with more, smaller steps from the low note to the high note when doing a one octave bend.

The details you provided concerning recording the effect using two tracks of the onboard sequencer is pretty much exactly what I was thinking when I wrote, "...you should be able to make something similar to the second sample using two tracks of the onboard sequencer and just play that onboard 'song' when needed."  I haven't tried it, but I don't see why it wouldn't work, and it would be a good way to get the effect right on the keyboard without having to use a DAW.  It's kind of like what I posted a couple months ago about using the sequencer tracks like multi-pads, but of course, like I realized after I posted that, you cannot use that technique and use the styles at the same time, because when you play back a recorded sequencer song, it will only play whatever style (if any) you recorded with the sequencer, and only for as long as the length of your sequencer recording.

[SeaGtGruff]

Michael, I'm not sure about the 128 steps with the pitch-bend wheel.  Maybe the data/software is set up to be able to handle that resolution, but I'm thinking that the wheel itself only supports a much lower number of steps.  When I had it set to 12 (and you are correct, in that gives it a total range of 2 octaves from -1 octave to +1 octave), I heard what sounded like a LOT less than 128 steps as I was recording my samples.  It was almost chromatic (and 2 octaves chromatic would be 25 individual steps) as it almost sounded like playing the individual notes on the keyboard, but not quite, as I could not exactly hit the "A" on the way down after sliding up to the "C".  I'm guessing there are actually maybe 20-32 steps -- but I'll have to check it again.  I'm at my Dad's home right now and do not currently have my keyboard with me.  I'm wondering if the E443 and E453 have a higher resolution for the pitch-bend wheel than the E433.

No, I'm pretty sure they're the same resolution as the PSR-E433. And you're probably right about the resolution that the wheel itself can actually produce versus the resolution that the keyboard can handle.

As it happens, I already tried something similar to what I described just last week, because I wanted to see if I could create a pulsating wah-wah synth sound that slowly swept up an octave with a combination of turning the filter cutoff knob back and forth continuously for the pulsating wah-wah and using the pitch bend wheel for the slow octave sweep. I did the octave sweep exactly as described in my post-- set the pitch bend range to 6, then played the note halfway between the low and high ends of the sweep (in this case, I was going from D to D, so the tritone note was G# rather than F#). It was awkward to repeatedly turn the knob with my right hand while playing the G# with my left hand and simultaneously using my left pinkie finger to operate the pitch bend wheel, so you can imagine that the recording was far from perfect, but I imported the recording into Cubase so I could even out the slope of the pitch bend. I was also going to edit the cutoff events to a consistent sine wave shape, but I haven't done that part yet.

Anyway, in looking at my original recording in a MIDI event list, I can see how the data values for the Pitch Bend events are always multiples of 128 (since the LSB is fixed at 0), but the changes aren't consistent multiples of 128-- e.g., it might jump by 512 from one value to the next, or by 384, or by 256, or by 128. I'd previously noticed that there's also an issue with the values produced by the Live Control knobs, in that the change in values appears to be greatest when the knob is turned to the far left or far right, and least when the knob is near the middle. I'm guessing the situation may be similar with the pitch bend wheel. I'm not trained or edgy-kated in electronics, but I imagine it's something to do with potentiometers and how voltage levels are translated into digital values.

In any case, I'm pretty sure the instrument will respond to all possible Pitch Bend MSB values, even if its physical controls has trouble generating them, hence the benefit (one might even say "essentialness") of importing the recording into a DAW to even out the change in values.

you cannot use that technique and use the styles at the same time, because when you play back a recorded sequencer song, it will only play whatever style (if any) you recorded with the sequencer, and only for as long as the length of your sequencer recording.

That's true, and is a major drawback of trying to use a song file while performing if you also wish to use a style. But one possible workaround for the first part of that-- "it will only play whatever style (if any) you recorded with the sequencer"-- would be to use one of the User Styles, because when you record a song that includes the accompaniment track it doesn't record the actual style, just the style number, along with the changes in the key, chord type, and style section. Thus, if you load a different external style file into whichever User Style slot you'd used when making the recording, the song will play back with the new style-- the same key changes, chord changes, and section changes, but with a different style.

[SciNote]

I had a chance to check my keyboard today.  It is hard to tell exactly how many steps the wheel allows, as there are more steps than I originally thought, and some of them are pretty close together, but it seems like there are about 43 steps, which I know is a strange number for a digital device, but it seems to be at least in that ballpark.  Additionally, it seems like the steps closer to the neutral middle part of the wheel's position (the position it springs back to when not being used) are closer together in pitch then the steps near the upper and lower edges of the wheel's travel.