Plogue OPS7 (bit-accurate DX7 emulation!)

[MadBrain]

Not sure if this isa bug but I mention it anyway
I can only compare to dexed and my tg77 ( which is not a dx7 )
Even when key sync is on ( instant attack and fast decay to zero for modulator ) , it sounds less clicky compared to dexed ..iow a softer attack .
Plogue's s a short linear fade in ramp , dexed has a short exp.fade in and a sudden drop responsible for the click
Anyone with a real dx7 that can jump in ?
It's most noticeable with low ratio operaotrs 0.5 going into 0.5
[picture]

Yeah, the fastest rate still takes about 40~50-ish samples to ramp up on the original dx7...

Later FM chips for home consoles and computers (Genesis, OPL2 on soundblasters etc) have a special case that instantly goes to the maximum level at the fastest setting, they might have added this special case to the TG-77 as well.

[gentleclockdivider]

In the example I gave, I tested 4 synth: MODX, OPS7, DX7 V and FM8. They were in that order as well.

Anyway, OPS7 is very near to MODX. It's almost identical while DX7 V and FM8 are lacking some
details and character. DX7 V is slightly better than FM8.
I have no idea how near the MODX is from the original DX7 though.

I don't think the modx uses oldskool sine lookuptables ( I am almost sure
That being said , every fm synth will sound close to each other since it's phase modualton after all .
I am not sure if the latest yamaha fm synths uses calculated sines or high res sine samples , but they sure don't use the old lowres lookup table method
Scroll down to YM 21280
https://ajxs.me/blog/Yamaha_DX7_Technical_Analysis.html

[EvilDragon]

BTW - I'm one of the "five people" that watched David's DX7 engineering videos, and thoroughly enjoyed them. I love people's passion for doing these slightly crazy things that interest them, and love this kind of reverse-engineering work to understand quite the work and engineering skill of creating the original instruments.

I think I'm the sixth!

I absolutely loved the nerdy video. David, when is part two coming?!

[Teksonik]

It's not just a dx7 , it's actually two
+ Every operator has extended parameters , feedback , noise , waveforms and then there's the modmatrix with stepsequencers that can modulate almost everything .
Pretty advanced stuff but if you didn't like the dx7 in the first place I gues it's not for you
p.s. I never cared about presets , I am just an fm nut ..I luvit

I owned a DX7 and TX7 combo from new back the 80's so am no stranger to layering FM patches. I also got to experience a TX816 with Bo Tomlyn at the helm.

I loved the DX sound back then but that was 30 years ago and the FM sound has grown a bit stale for me.

Although the Tubular Bell sound not only defines the DX for me but also an era in my life so it along with a few more still hold a place in my heart but rarely if ever find a place in my compositions.

I only managed to save a few of the patches I made for the DX back in the day and would love to hear how this emulation imports them as others have had difficulty with some of them.

I might be a buyer of this synth if I can demo it first but like I said it's going to have to blow me away.

[steb_osc]

In the example I gave, I tested 4 synth: MODX, OPS7, DX7 V and FM8. They were in that order as well.

Anyway, OPS7 is very near to MODX. It's almost identical while DX7 V and FM8 are lacking some
details and character. DX7 V is slightly better than FM8.
I have no idea how near the MODX is from the original DX7 though.

Thanks. This is exactly why I'm here on this forum. I love comparison wavs

For me, FM8 was the weakest sound, but I have only listened on a micro speaker so far.

[pdxindy]

This is obviously THE DX7 emulation!

[cantaloupe]

Brilliant work. Thank you Plogue.
Have been waiting years to have a real-sounding microtonal DX in software. (And even with MTS-ESP!!!)

Great having someone on the scene who shows this amount of attention to detail - that is what it's all about. This feels like my hardware DX.

Bug: MTS-ESP setting isn't permanent when saved as default, like the other parameters. It always return to 'disabled' everytime i add a new instance.

Only criticism i have of this masterpiece is that i think the GUI could need a tiny bit more polish, mostly regarding font use, padding/margin and positioning/grouping. The "Patch" area feels particularly haphazard visually. And small details, like the three operator knobs not being centered to the operator window with unequal margins on left and right. That said, it does feel cool in a utilitarian way, and the palette is nice.

[davidv@plogue]

I think I'm the sixth!

I absolutely loved the nerdy video. David, when is part two coming?!

Thank you!
If I don't finish it soon, I never will, so I'll see if I can get it out worst case in two weeks.
I naively thought I could finish it all for the hard release date we targeted, and started reducing the quality at the end. Decided that was wrong and cut at the end of the computer stuff to give the audio part justice it deserved. There is a lot of interesting audio trivia left (I think).

[davidv@plogue]

Bug: MTS-ESP setting isn't permanent when saved as default, like the other parameters. It always return to 'disabled' everytime i add a new instance.

Confirmed. Thank for spotting this. MTS-ESP was added very late in beta.

Only criticism i have of this masterpiece is that i think the GUI could need a tiny bit more polish, mostly regarding font use, padding/margin and positioning/grouping. The "Patch" area feels particularly haphazard visually. And small details, like the three operator knobs not being centered to the operator window with unequal margins on left and right. That said, it does feel cool in a utilitarian way, and the palette is nice.

I hear ya (and others too). The UI needs a few tweaks. Knowing me I always rather spend time on the DSP side in priority. I know that doesn't always equate to a better product ha.

[olepro]

BTW - I'm one of the "five people" that watched David's DX7 engineering videos, and thoroughly enjoyed them. I love people's passion for doing these slightly crazy things that interest them, and love this kind of reverse-engineering work to understand quite the work and engineering skill of creating the original instruments.

I think I'm the sixth!

I absolutely loved the nerdy video. David, when is part two coming?!

Seven here
I think this is what Siegfried Kullmann did with SQ8L, and Rainer Buchty had already done the reverse engineering

[Caine123]

hey guys i dont get it, is there no ADSR for all controls ? so you can set those for all operators?

[fmr]

hey guys i dont get it, is there no ADSR for all controls ? so you can set those for all operators?

This is an emulation of the DX7...

https://usa.yamaha.com/files/download/o ... /DX7E1.pdf

[Caine123]

hey guys i dont get it, is there no ADSR for all controls ? so you can set those for all operators?

This is an emulation of the DX7...

ok i dont know the DX7 so well, so i assume i need to adjust each operator's release? :/

[fmr]

hey guys i dont get it, is there no ADSR for all controls ? so you can set those for all operators?

This is an emulation of the DX7...

ok i dont know the DX7 so well, so i assume i need to adjust each operator's release? :/

You have to be aware that only the carriers "play". The modulators act on them by producing partials (this way changing the waveform). The more modulation, the more partials that are created, the richer and more complex the resulting waveform will be. Modulator envelopes control that (roughly speaking).

[beely]

ok i dont know the DX7 so well, so i assume i need to adjust each operator's release? :/

Welcome to Yamaha's implementation of FM from the early 80s - enjoy the ride!