Help in recreating this sound?

[Parduz]

It looks like all the effects are created using a BBD

Probably it's a simple chorus.

The chorus is a "global effects" that acts on all sections (bass - chords - solo) while the Celeste effect is a mono effect that affect only the Solo (monophonic) voices.

ps-55 solo section.jpg

I don't know how to describe it, but i plugged an oscilloscope to the output of the keyboard, 25 years ago: once the sound is stable (so, after all the envelope(s) movement, i was able to trigger the scope in order to see a single wave cycle, and the Celeste was like a thin triangle put in the middle of the cycle, and the vertice was moving up and down like if it was driven by a slow sine LFO.
Sorry, i have no better way to describe it.

[Ah_Dziz]

I would try a simple delay plugin with some modulation. Use zero feedback and start at 50% dry/wet. There are tons of delays that will do this. You'll want the delay set close to 0ms. You'll be able to get close. With most of these things you can get very close but getting it "exactly the same" is a ton of work and you may never get it. By figuring out how to get close you'll learn a bunch about sound design. If you absolutely need that sound, just get the Yamaha.

[Uncle E]

The schematic shows the Celeste switch going to the BBD. So while they may have implemented it in an unusual and cool way, it's still just a short modulated analog delay at its core.

[Parduz]

It's literally just 2 saw waves an octave apart. The actual hard part you could spend days on would be to get the exact filter settings and detuning/unison/pwm/ everything there is to the upper mid grit/movement.

Additional tip: Everything above 11k is brickwall-filtered off in the original sample (probably due to file conversions / hardware limitations), so you should try to achieve the same.

EDIT: Proof below (2nd sound is the original sample), feel free to check in spectrum analyzer:

https://www.dropbox.com/scl/fi/f67w0ot5 ... rb42e&dl=0

I admit my ignorance, or my dumbness:
i drawed two saw waves in Vital (the init saw goes on the wrong direction), put one of them one octave up, both on the left channel, and put in the right channel the sample.
Then, using SPAN to look at the result, i started drawing up and down the frequencies bars in the editor of Vital, until everything matched at best.

But the result is nowhere near to the sample, nor to your example: it sound too thin.

May i ask to look at the patch of your example? I'm pretty sure that, at this early stage, i am missing something which is both basic and obvious, but i can't grasp what i am doing wrong.

[Agg]

It's literally just 2 saw waves an octave apart. The actual hard part you could spend days on would be to get the exact filter settings and detuning/unison/pwm/ everything there is to the upper mid grit/movement.

Additional tip: Everything above 11k is brickwall-filtered off in the original sample (probably due to file conversions / hardware limitations), so you should try to achieve the same.

EDIT: Proof below (2nd sound is the original sample), feel free to check in spectrum analyzer:

https://www.dropbox.com/scl/fi/f67w0ot5 ... rb42e&dl=0

I admit my ignorance, or my dumbness:
i drawed two saw waves in Vital (the init saw goes on the wrong direction), put one of them one octave up, both on the left channel, and put in the right channel the sample.
Then, using SPAN to look at the result, i started drawing up and down the frequencies bars in the editor of Vital, until everything matched at best.

But the result is nowhere near to the sample, nor to your example: it sound too thin.

May i ask to look at the patch of your example? I'm pretty sure that, at this early stage, i am missing something which is both basic and obvious, but i can't grasp what i am doing wrong.

I created it in MPowerSynth, but it really doesn't matter. The original sound was made with basic oscillator functions, signal routings, filters & filter envelopes, and fx. Mine was just 2 saws, with pwm and mono unison, through 2 filters (lowpass and highpass), into compressor, and like an inch of distortion from a waveshaper.

Both compression and distortion were just for character/trying to make the thing sound less boring, they are not needed. In fact I did another one in Vital just now without any compression/distortion, you get the idea:

https://www.dropbox.com/scl/fi/2y2m08gx ... 1liv8&dl=0

Patch (I'm not too familiar with Vital, hope importing works): https://www.dropbox.com/scl/fi/k2vntsjz ... z1ne7&dl=0

Some screenshots:

[Agg]

Also:

i drawed two saw waves in Vital (the init saw goes on the wrong direction)

Not sure exactly what you mean, but nothing wrong with Vital's saws. They could have opposing polarity tho, or the graphical elements in the synth might be different.

[richtrelo]

PCM wavetable synthesizer. You can see in the schematic (attached) that it uses the iG09510 for the solo and YM2022 for the orchestra, each with its own ROM. The YM2023 used for the rhythm is strictly PCM playback.

oh! this are great infos. Thanks a lot!!!

Yeah. Moogish bass with a tiny loop after the filter envelope makes it to its sustain level.

So is it doable by Vital?

You could definitely do this in Vital.