Plogue OPS7 (bit-accurate DX7 emulation!)

[NikkiA]

When 10,000 users are happy with something so just 'STFU', then the 1 single user who isn't interested in it remaining working like a DX7 becomes the only voice in the room.

That's why I'm not STFUing about your absurd 'make it act like other synths' BS, just buy FM8, it's $10 atm and has the envelopes like you'd prefer.

edit: it's not a 'bug' btw, it's the DX7's keyboard level scaling and breakpoint interacting with the rate-level envelope system, almost certainly the same way it does on a real DX7

[Nokenoku]

@NikkiA

Are you stupid? Or are you trolling?
It has to be one of those 2, since as I already explained my bug-report and UI-suggestion has NOTHING to do with "isn't interested in it remaining working like a DX7".

[jbraner]

That's why I'm not STFUing

Just for the record, I'm pretty sure it should be S-ingTFU - shouldn't it?

OK, Carry on...

[davidv@plogue]

The non deterministic envelope is, in fact accurately modeled after the YM2129 EGS chip.
The internal clocks of the EGs are designed a certain way where certain levels are compared at the wrong "tick" hence you never get the same thing (and it depends on specific settings).

[MadBrain]

I have possibly found a bug with the envelopes.
(Not sure, if the original DX7 acts the same.)

With L2 being at maximum, the R2 timing is randomly changing from note to note. Also the envelope-curve-picture is not reacting to R2 in that case (it should show a plateau).
Lowering L2 a bit seems to fix that.

I made a short video to show the issue:
https://www.youtube.com/watch?v=wvoiIiLLSXQ
[...]

Not a bug. The real DX7 (mk1) totally does this and I just tested it and it does it indeed. IRL on all Yamaha FM chips, envelopes share a global clock, which makes every single note unique, depending on which precise sample the note starts on, and on the exact previous state of the envelope. On most settings the difference is very small, but you've found one configuration where the difference is much larger (and there's more than one configuration like this). Our goal is to emulate the DX7 mk1 100% with all its quirks, including this one.

The envelope display is just intended to give a visual preview of the envelope shape with more normal settings, no it doesn't handle all the fringe cases (but the actual sound emulation DOES), yes setting an envelope level to 50 gets you much less than 50% of the maximum volume on a real DX7 too (it's on an exponential scale). You should use your ears much more than your eyes to decide if you like the envelope. Hope this helps.

[Nokenoku]

Thx for letting me know.
Then of course it needs to stay this way as being able to accurately emulate the DX7 is top-priority for this instrument. I am not disagreeing about that!

Of course I use my ears to tweak the sound.
But the DX7 envelopes are rather different from what you know from other synths. And maybe it's just my bad memory, but always when I come back to it (or rather Dexed in the past), I had to relearn it.

And then of course you look at the graphs in expectation that they will help you, but all they do is to confuse you even more.
I know, there's no perfect solution for this, as you can't properly portray all the timing-proportions in such a small window without zooming, but I think there has to be a better method.

[gentleclockdivider]

Of course I use my ears to tweak the sound.
But the DX7 envelopes are rather different from what you know from other synths. And maybe it's just my bad memory, but always when I come back to it (or rather Dexed in the past), I had to relearn it.

And then of course you look at the graphs in expectation that they will help you, but all they do is to confuse you even more.
I know, there's no perfect solution for this, as you can't properly portray all the timing-proportions in such a small window without zooming, but I think there has to be a better method.

It's simple , a high rate amount (fully clockwise ) is instantly going to the level
The reason why the window does not make 100% sense is because all of these time parameters are trying to fit in the window .
If there would be a time indicator along the x=axis ,you would see an increase or decrease in time when you move the rates .
Maybe that is what is confusing you , when you adjust the rate to something very slow , another rate parameter is shortened in the window but this is becasue it's all trying to fit into that window
Tip , you can just use the envelopes as regular adsr envelopes
Rate 1 is attack
Level 1 set to max and leave them at that position
Set rate 2 and level 2 to max and leave them at that position
Set level 4 to min and leave at that position
Your adsr is now
Rate 1 =attack
Rate 3 =decay
Level3 =sustain
Rate 4 -release
Voila , a regular adsr

[stash98]

I love this synth. Have owned it since it came out.

Amazing that one guy is bumping multiple plague threads to complain about petty things.

[nominil]

I love this synth. Have owned it since it came out.

Amazing that one guy is bumping multiple plague threads to complain about petty things.

Such is the way of KVR.

OPS7 is a great synth. As close to perfect as they get, imo

[SebAV]

I love this synth. Have owned it since it came out.

Amazing that one guy is bumping multiple plague threads to complain about petty things.

Such is the way of KVR.

OPS7 is a great synth. As close to perfect as they get, imo

People have different knowledge and culture and sensitivity. One thing can be tough of being « petty » for some and be important for other. It’s not as if there’s only one dx7 emulation in VST world (but only knowledgeable people know how OPS7 is the one).

[gentleclockdivider]

Weird bug
initialise OPS 7 , set volume to max , don't play anything
Algoritm 16,17,18 produce background noise.
I reloaded the plugin multiple times to make absolutley sure (no other instrumets or effects loaded )
Issue only occurs on algo 16,17,and 18 and ONLY after initialising ops 7

[Tj Shredder]

The non deterministic envelope is, in fact accurately modeled after the YM2129 EGS chip.
The internal clocks of the EGs are designed a certain way where certain levels are compared at the wrong "tick" hence you never get the same thing (and it depends on specific settings).

Just curious, does the DX 7 II behave the same? I always preferred the second incarnation. The first was too noisy…

[davidv@plogue]

Its a hardware thing of the OPS7 when only a single carrier is active. Question is what about initial boot/load state, well there are too many factors influencing the EG and OP volumes at any certain point to match a HW that's been booted for a while free running. Anyway We had this report a few times and made sure it was the same. Not sure if the DX7II behaves the same on that end for the moment, we really concentrated on the EGS/OPS.

[gentleclockdivider]

An interesting observation
When ratel/level 2 is zero , only rate/level 1 and rate/level 4 have an effect , rate/level 3 has no effect at all .
This essentially makes the envelope a pure attack-sustain-release envelope , where the sustain phase is defined by the level 1
This makes me wonder , since rate 2 is zero and thus infinitely long is this the reason why R3-L3 is ommited ?



Edit , I think found a bug
Repeatedly play a note with a single operator enabled and set the envelope to the following settings

Result , most of the times it's short decay but sometimes there is an audible sustain phase before the decay (see screenshot)

.

[Robmobius]

I'll say it again - It's a great synth.