by BillB » Tue Jun 26, 2018 7:54 am
I'll shortly receive a Yamaha PSS-780 for very low cost. I am very interested in the options and limitations of two-operator FM synthesis. The FM synth parameters are:
Op1 (modulator): attack, decay, frequency, feedback.
Op2 (carrier): attack, decay, frequency, level.
Mod depth – Op1 to Op2.
There are numerous other fun such things as chorus, vibrato, portamento, and sustain/reverb which seem to be envelope effects.
I've viewed enough Youtube to obtain the concepts of modulator modifying carrier and therefore producing additional harmonics, and sweeps while using A/D parameters etc.
Things I do not understand is, because of the "home keyboard with presets" beginning point, how would be the initial preset sounds showed up at? Could it be likely that they're pre-configured FM waves to create the Piano, Brass, Flute, Chime, Drum, Voice etc presets, which could then be further modified through the on-board FM process? Someone (on GS I believe) stated that using the same FM parameters towards the "Sine" voice and also the "Harp" voice (for instance) doesn't increase the risk for same seem. That implies the preset sounds (and subsequent edits) might not be simply variants of two-op sinewave interaction.
Can anybody shed any light?
Incidentally, if anybody comes exactly the same way, the PSS-480, 580, 680, 780 would be the only PSS's with MIDI _AND_ FM synthesis. The 480/580 and 680/780 differ only within the paint jobs. All could be edited via MIDI, giving use of more FM parameters, via some old (Home windows) and newer (Java) editors. They likewise have 5 (FIVE!) memory slots which may be dumped to sysex storage. I'll be having fun with the program options in the end, but it might be best to comprehend the the beginning reason for the sounds.
by AndyHornBlower » Tue Jun 26, 2018 8:23 am
I've the PSS-580, with no, it's not only 2 sinewaves. You choose from the small group of other waveforms, like half sine (having a flat bit), sine squared (double hump), and so forth. So, it is a bit a lot more like a 2op form of a TX81Z than the usual 2op form of an Facebook-01 – however with less parameters.
There's some software that allows you to edit the 5 "bank" patches from the PC, known as PSSedit. That uncovers a couple of more parameters, but not the entire set that the built-in preset sounds derive from. I am unsure.
The waveforms derive from doing various things having a wavetable – the sine wave is within a wave table, possibly only a quarter cycle, that is all that you should rebuild a sine wave (see clearly backwards, invert it, and so forth). The additional waveforms do such things as departing area of the cycle flat, or mangling it in certain alternative way. It's fairly primitive, but helpful.
So far as I understand, the drums are PCM, i.e. built-in samples.
You may still download PDF manuals for that PSS-480/580, most likely others. E.g. try manualslib, or simply google on their behalf. That covers the parameters you are able to edit in the keyboard, which fits pretty much.
You might not require the manual for the actual model – the 780 is really just a long 480/580. The 480 and 580 are the identical, even lower towards the firmware, they simply have different coloured cases and buttons.
PSSedit was designed in VisualBasic, however it operates on Home windows 7, or XP (I have attempted it this way, like a virtual machine in "XP Mode", inside Win 7 Pro). If you are Mac bound, you'd must find an online machine that may run XP, or possibly Win98. On Linux, this could work via WINE.
by BillB » Tue Jun 26, 2018 10:04 am
I've the 680/780 manual and also have had phone parameters editable in PSS-Edit. I see that which you mean concerning the sinewave variants. Never happened in my experience that FM wave variations might be generated as simply as studying the wavetable diversely – cunning method to squeeze higher productivity of the nick – presumably exactly the same pertains to the second-gen 4-op synths (such as the TX81z you mention) that offer sinewave variants.
Possibly it is only individuals variations and also the full-range of parameters which take into account the 100 presets, that are then further editable.
A workout that we is going to do after i obtain the 780 would be to load say preset "Piano 1" into an edit bank, upload it to PSS Edit, then screenshot the parameters. Then load preset "Sinewave", change all of the parameters towards the same figures because the "Piano" – and find out whether it really sounds exactly the same. If that's the case, that will claim that the much deeper degree of parameter editing over MIDI can take into account the entire selection of presets, instead of (when i was secretly wishing) there can be another operator or more hiding away inside to create the presets.
Intriguing little machines for techno-nerds
by AndyHornBlower » Tue Jun 26, 2018 8:08 pm
Yes, that appears just like a good test.
There may be some parameters PSSedit can't touch, I guess, but you can figure that out, a minimum of.
the purpose – they may used a 4op FM nick, but made the consumer edits to sounds 2op only, for simplicity. Most likely the clue they did not, is based on because you reach edit a duplicate of individuals sounds, and you may begin by just altering them just a little.
I believe in addition likely would be that the nick could layer several 2op sounds together, however the interface does not let's do this. so far as I'm able to remember.
I additionally own the PSS-390, using the 8 step sliders (but no MIDI), which quite a bit of fun. That certain layers two sounds automatically – it begins using the seem set to 2 layered "pianos", halving the polyphony – however, you can change one of these off. You are able to change a seem using the sliders, however that only edits one layer. Another needs to be a preset seem. Pointless to state, the sliders don't cover all the possible parameters, which means you can't completely re-produce the preset sounds this way – you cannot change someone to be the same as another, you are able to just have to change so that it is diverse from it had been.
The PSS-x80s make good controller keyboards, if you want small keys, and do not need velocity.
As small keys go, they are quite comfortable to experience on, a minimum of for me personally. For controlling something similar to a mono synth that does not use velocity, or perhaps an organ seem on the ROMpler, a keyboard without velocity is perhaps better – it does not need to wait for a answer to achieve the underside, before deciding how quickly it arrived, to calculate the rate.
by BillB » Tue Jun 26, 2018 9:40 pm
I bring your point, but I believed the other way round. While in "seem source" mode, they react to velocity messages over MIDI, so possibly become a bit more significant. Unsure whether the Mod depth etc reacts to velocity, or if it's only note amplitude. Will be a shame if perhaps that, however, these were never intended to be "pro" – just "fun".
I must say I'm over-looking forward to getting this toy. I've bought a lot more costly synths previously, but always understood virtually what to anticipate when it comes to seem capacity. Using the 780, I do not. The interesting harmonic changes aren't especially well-symbolized on YT, but there's a couple of hints that could be rather interesting. Another aspects (sequencer, drums etc) also look interesting, provide expectations aren't set excessive!
Again, if other people has an interest, there's a great review of the PSS range here:
The X80's would be the only ones with FM and MIDI, which built them into the best choice for me personally. That, and somebody selling a "Yamaha organ" (780) on gumtree for £15.
by AndyHornBlower » Tue Jun 26, 2018 10:00 pm
Awesome. Yes, I have observed that page too. It is a good resource.
I have attempted playing the 580 from another keyboard, and I am fairly certain it's really a alternation in volume, with velocity, however the 780 may be different, I'm not sure.
It's surprising what you can do with 2op FM, simply sticking with sine waves. It's fairly simple to get a handle on it either. More operators enables you to perform a little more, however it will get harder to know what is going to happen whenever you change something, or things to try altering, to obtain an effect you are targeting.
I have viewed a number of youtube videos that "explain" FM, but many of them get so far as two operators then stop. I believe I have only seen one which were built with a stab at covering exactly what a third it's possible to do.
You will find 4op and 6op algorithms where it is just mixing multiple two operator stages, however when likely to operator modulating another, modulating another, and so forth, It really just comes lower to experimenting to determine what goes on.