Native Instruments Massive X Synth - Sequel to Massive (Out Now!)

[Kian Russell]

With Saturn in low quality mode on it?

Hey we should talk more about massive x
Sadowick saved me money I was going to buy it, now I am better off.

[mladi]

PS. We learned Saturn is shitty as well.

No Saturn is actually great, when you use it for things it's intended to. See Dan Worrall's video about it!

Of course it is. I forgot to add a or maybe or

Hope it's clear now.

[BlitBit]

OK I must say I am confused now, Sadowick. I just set the Sin-Tri-Saw-SQ wavetable direct to output, WT position at max (so it's a square wave), bypassed filter and all, then did a 96 semitone pitch bend on it (because MX is crazy like that it allows you such huge bend ranges, pretty cool) and recorded the output. Here's the result:

https://app.box.com/s/z1vrk9p037ggh8nwlejz2iy6ir6ef0pt


It's a WAV file. Can ANYONE hear any aliasing here? Even SPAN doesn't show anything that's not at around -130 dB or so.

I really hate to say this. I have a huge amount of respect for you, but I definitely hear what sounds like "heterodyning" or "aliasing" in that WAV file. After the main frequency reaches a certain pitch, I can then hear spurious pitches for the rest of the frequency sweep. It's easy for me to hear on my modest studio speakers. It's not a tiny detail I need precise isolated headphones and careful listening to detect.

That said, I still like Massive X.

By the way, does Massive X have any sort of "HQ" or "perfect" mode that it uses when not rendering in real time?

Did you download the file and play it in your DAW (without resampling) or a good player (like VLC)? As already mentioned above the web player of that download site introduces artifacts.

[pixel85]

200!

[BDeep]

You have to be careful to make sure that your observations aren't an artifact of your testing scenario. When I just run the sweep through iZotope insight on 44k the alisasing appears to be about 90 dB down from the fundamental, however, if I switch my system to 96k then they appear to be about 120 dB down. The wav file isn't changing, it's fixed at 48k so what's being observed is either a side effect of the measurement or dithering, or both. Without more testing, I can't know if even those results are also a side effect of measurement and that the aliasing is actually lower than that.

This, the wav is in 48kHz. I tested it in a 48 kHz environment, and you'll hear nothing (by ear anyway). You're right, people are hearing the SR conversion.

Jeez

[McLilith]

200?

Missed it.

Some of us had a goal of this thread reaching 200 pages, preferably before Massive X was released.

[fisherKing]

200?

Missed it.

Some of us had a goal of this thread reaching 200 pages, preferably before Massive X was released.

hmmm. 200 pages. 200 dollars. maybe the goal should have been 79 pages??

[Tj Shredder]

OK I must say I am confused now, Sadowick. I just set the Sin-Tri-Saw-SQ wavetable direct to output, WT position at max (so it's a square wave), bypassed filter and all, then did a 96 semitone pitch bend on it (because MX is crazy like that it allows you such huge bend ranges, pretty cool) and recorded the output. Here's the result:

https://app.box.com/s/z1vrk9p037ggh8nwlejz2iy6ir6ef0pt


It's a WAV file. Can ANYONE hear any aliasing here? Even SPAN doesn't show anything that's not at around -130 dB or so.

Actually I do hear aliasing (very soft and only when really focusing on it), but in a range where I do not hear the original pitch anymore. It must be way above any pitch you would want to compose as fundamental. But this was playing it in the browser, playing it back after a download I don't hear any aliasing... (SR mismatch and bad sample rate conversion in the browser...)

[BDeep]

200!

Cake! Where's cake? NI should bring us all cake. And I want to eat it too!

[McLilith]

OK I must say I am confused now, Sadowick. I just set the Sin-Tri-Saw-SQ wavetable direct to output, WT position at max (so it's a square wave), bypassed filter and all, then did a 96 semitone pitch bend on it (because MX is crazy like that it allows you such huge bend ranges, pretty cool) and recorded the output. Here's the result:

https://app.box.com/s/z1vrk9p037ggh8nwlejz2iy6ir6ef0pt


It's a WAV file. Can ANYONE hear any aliasing here? Even SPAN doesn't show anything that's not at around -130 dB or so.

I really hate to say this. I have a huge amount of respect for you, but I definitely hear what sounds like "heterodyning" or "aliasing" in that WAV file. After the main frequency reaches a certain pitch, I can then hear spurious pitches for the rest of the frequency sweep. It's easy for me to hear on my modest studio speakers. It's not a tiny detail I need precise isolated headphones and careful listening to detect.

That said, I still like Massive X.

By the way, does Massive X have any sort of "HQ" or "perfect" mode that it uses when not rendering in real time?

Did you download the file and play it in your DAW (without resampling) or a good player (like VLC)? As already mentioned above the web player of that download site introduces artifacts.

Ok. I just used the web player. I'll give it another listen in a bit, after downloading the WAV file, and using better software to play it.

[EvilDragon]

Yeah, it's a 48k file. You want to avoid any SRC when playing it back, so set your audio interface to 48k as well.

[Functional]

OK I must say I am confused now, Sadowick. I just set the Sin-Tri-Saw-SQ wavetable direct to output, WT position at max (so it's a square wave), bypassed filter and all, then did a 96 semitone pitch bend on it (because MX is crazy like that it allows you such huge bend ranges, pretty cool) and recorded the output. Here's the result:

https://app.box.com/s/z1vrk9p037ggh8nwlejz2iy6ir6ef0pt


It's a WAV file. Can ANYONE hear any aliasing here? Even SPAN doesn't show anything that's not at around -130 dB or so.

Because with processing these artifacts are accentuated. EQ, compression, saturation, ect bring these out.

Demonstration just for you using your audio sweep:


These things matter

You're starting to lose your credibility if you're using third party plugins to add something that isn't there in order to prove it. Up until this point the proposition "I like NI, I just think their standard should be higher" was reasonable although I'd call it still BS. But now, beyond any doubt, you're making a fool out of yourself with such intellectual dishonesty and at this point you're just proving that you have a mission which is to prove that Massive X has supposedly some design flaws just because you don't like the synth.

It's alright to not like it, but if you're going to lose every bit of integrity in order to prove that there supposedly is an objective reason (there is not), then you're at trouble.

Also, why don't read for shit here? For example the audio rate lfo thing - don't you think it's pretty relevant that there exists practical reasons to not implement anti-aliasing measures to audio rate LFO's? Yet you have still a video up where you don't even mention any of these practical reasons nor do you respond here when you have been asked twice to give us an usecase for audio rate LFO's going at 9C or above.

Seeing that plugin reviews are large part of your channel, I'm actually now concerned how easily someone can get 156k subs when you make content with such colossal flaws. Haven't you seriously considered that at some point you actually have a responsibility to at least try and be correct about info that you put out?

You could have attempted to reach out to NI first before making the video and ask for clarifications. I can't say whenever they would have responded, but then you could have said that they didn't respond to you or something. You could have consulted at least someone else than only yourself on this topic, who could have pointed out few quite integral aspects. You could have also mentioned that aliasing at that is below -100db is practically inaudible. We're talking about dithering levels here in some cases and dithering truly is meant to be inaudible noise.

[zerocrossing]

Here are some things that I think could immediately help the interface with little effort by NI.

• Underline to indicate a button isn’t enough, especially when color is so subtle. Put a slight gradient underneath that can still terminate with a line if you want. Come up with a similar way to indicate the difference between a number that can be modified vs. a number that can’t.
  
  Implement a little tooltip system, like in So much in the UI isn’t labeled!
  
  Perhaps label things.
  
  Double clicking on an element in a menu should select it and close the menu window.
  
  Implement scroll ring interaction. (You’d think this would be standard. :facepalm:)
  
  Design a good preset browser. (hint: we all know how to use a Windows/Mac style browser window!)
  
  Decoupling the drag element graphic from the modulator that’s being dragged is... :facepalm: Put them in the window. (Upper right looks fine)
  
  Mod target windows too subtle. A little gradient to make them look a bit concave would be nice to know something goes into it.
  
  Why are Pitch Bend, Mod wheel and aftertouch (which should respond to polyphonic aftertouch ) separate from the rest of the modulators?
  
  Some modulation animation would be nice. Doesn’t have to to be extensive... maybe have it only work on mouse over to save CPU.
  
  Shocked that ADSR curves aren’t shown. Really? That’s kind of an epic fail in this day and age, especially with a complex envelope.
  
  They started slight differentiations in color in the noise module... and stopped. It doesn’t have to look like a Moog Matriarch, but something to denote the different sections would be nice.
  
  Add a modulation matrix page. Dragging around the little icon is nice... but sometimes a plain list like in Omnisphere is really a lot better, especially in a complex patch.

That’s all I can quickly think of off the top of my head. If one of my designers handed this UI to me, those would be the beginning of a punch list of things to take care of first.

The sound, is excellent, so I really have no complaints there. Love that you can route things pretty easily. I love routing an oscillator past a filter while the other goes through.

Filters sound great. Dune 3 and Massive X have really raised the bar here, IMO. That comb filter sounds phat as hell. I don’t quite understand how FM works, though. I’ve used the routing to put osc 2 into FM and nothing happens...

Effects sound good. Why no sync on effect mod rates? That seems like an omission.

The presets are very good. They seem to be a good blend of impressive but unusable and usable patches.

Anyway, despite all my criticism, I really like it a lot. Much better than Pigments, to me. Pigments seems to want to burn my CPU to the ground for sound that I don’t feel is any better. I’m going to really dig this one.

[pixel85]

200!

Cake! Where's cake? NI should bring us all cake. And I want to eat it too!

They will but instead of flour they will use aliasing as an ingredient

[BlitBit]

OK I must say I am confused now, Sadowick. I just set the Sin-Tri-Saw-SQ wavetable direct to output, WT position at max (so it's a square wave), bypassed filter and all, then did a 96 semitone pitch bend on it (because MX is crazy like that it allows you such huge bend ranges, pretty cool) and recorded the output. Here's the result:

https://app.box.com/s/z1vrk9p037ggh8nwlejz2iy6ir6ef0pt


It's a WAV file. Can ANYONE hear any aliasing here? Even SPAN doesn't show anything that's not at around -130 dB or so.

By the way, this file has a sample rate of 48kHz. If the sampling rate of your DAW, audio player or operating system is set to something different the any of these components will have to resample the file to play it back. This will - you guessed it - introduce artifacts.

So: download the file, make sure that the whole chain is set to 48kHz and then play the file. You won't hear anything.