- lets you create a custom preset based on your audio: true, after analysis, if I like the result and think I can use it elsewhere, I can save it as a preset
- get you to a faster starting point - true, with one click and a 10 second wait, I get a pretty usable starting point that provides some nice "polish" to the track as a good starting point
- different instrument tracks are automatically detected - true, it works
- applies the spectral shaping technology to provide subtle clarity and balance to each of your tracks - true, it reliably provides clarity and balance to my tracks
- EQ nodes are intelligently placed where Track Assistant thinks you need them - true, when I exagerrate the nodes I can hear that usually there is some kind of resonance where they are placed that sound better when cut
- speeding up your workflow - true, because of the above
- eliminating much of the EQ guesswork - true, it can find those resonances and covolutions so I don't have to go searching for them
- recommendations for optimal starting points across all of Neutron's modules - true
- offer ‘infinite’ presets that are custom tailored to your sound - sort of true but they aren't really presets.
- allowing full control over all your mix decisions - true because all the parameters are editable
Izotope releases Neutron: "A smarter way to mix"
- KVRAF
- 6287 posts since 8 Jul, 2009
Here's my take:

#NONFR Check out my music at Bandcamp
Free Streaming!
Free music with your support on Patreon | Youtube: Music of Plexus Videos (music videos) | Youtube: Plexus Productions (audio related) Stop whining. Make music.
Free music with your support on Patreon | Youtube: Music of Plexus Videos (music videos) | Youtube: Plexus Productions (audio related) Stop whining. Make music.
- KVRAF
- 6287 posts since 8 Jul, 2009
I think there is further enhancements they can make to Track Assistant. For example, the ability to select subjective characteristics such as "warm, tight, crisp" and have it automate parameters to meet these. And in general make the process better.garryknight wrote:A few things occurred to me while using Neutron:
1) iZotope are going to have their work cut out for them, IMO. Where do they go next? What makes most sense to me, and I'm sure other people are going to think of this, is if the next version of Neutron were to listen to all of the tracks when making decisions about any particular track. Just as when we make manual changes to any of our DAW's channels, we listen to those changes in the context of the whole mix, perhaps Neutron ought to be doing the same.
Try setting the drop down to "Aggressive".garryknight wrote:2) I was one of the people who, when first using the standalone Neutrino, was able to hear the changes even though they were subtle. When I use Neutron's Track Assistant, I'm not so easily able to hear those changes that the built-in Neutrino makes, even if I A/B them. Anyone else noticed this?
I agree. I find the features do what you suggest and get me to a better sounding track faster by helping me find some of the problems automatically.garryknight wrote:3) For me, learning to mix audio has been a process of 'homing in'. When I started, I'd EQ a track and wouldn't really know what to listen for, so I wouldn't realise how gross my changes were. Likewise with compression. The more I've listened, the more I now know what to expect, so I spot the problem areas more easily and home in on them faster. It seems to me that this is one area where Neutron's Track Assistant, the Learn functions, and the Masking feature can help most. By noticing and really listening to what Neutron does, people starting out in mixing might progress pretty quickly towards 'homing in' on where adjustments need to be made, even when they're not using Neutron. Probably stating the obvious here.
Try it. I did this, set all the instances to Aggressive and the character to what I felt was suitable to the track. then I made very minor tweeks. The result was pretty nice sounding and fixed some problems in my own non-Neutron mix. I posted those in this thread (track name on soundcloud is Planck's Little Constant.garryknight wrote:4) On the other hand, some people might just put Neutron on every channel and only use Track Assistant (and maybe Masking on kick and bass channels), then think that that's good enough. Maybe it will be good enough. Maybe we could call those people 'Neutron stars'.
Yep! It's a really great tool. not only does it show you in the frequency response where the problems are but it uses the psychoacoustic concept of Masking and visually shows you this. The whole histogram thing is really helpful. all in all providing a very effective mix of interactive visual tools aimed at efficiency.garryknight wrote:5) When using Masking, for me it's never enough to only check kick against bass, vocals against guitar and so on. There are enough EQ nodes available to check most channels against each other and get everything nicely separated out. Anyone else found that?
#NONFR Check out my music at Bandcamp
Free Streaming!
Free music with your support on Patreon | Youtube: Music of Plexus Videos (music videos) | Youtube: Plexus Productions (audio related) Stop whining. Make music.
Free music with your support on Patreon | Youtube: Music of Plexus Videos (music videos) | Youtube: Plexus Productions (audio related) Stop whining. Make music.
- KVRAF
- 4085 posts since 28 Jan, 2011 from MEXICO
It is working for me,ñ and I like it but if its really true that track assitant is not really doing anything then even if the results are good I would feel scammed.
What I would like to see is some kind of learning algorythm which you can feed with music you like or maybe that it could learn from you so it can just make the things you do but faster.
What I would like to see is some kind of learning algorythm which you can feed with music you like or maybe that it could learn from you so it can just make the things you do but faster.
dedication to flying
- KVRAF
- 4085 posts since 28 Jan, 2011 from MEXICO
It would also be Nice to choosed the time you want the track assitant to "listen" so it could produce more unique and usefull presets.
dedication to flying
-
- KVRian
- 714 posts since 7 Sep, 2012
plexuss, I'm afraid you're not following me but I really can't make it any clearer than I managed (or not) to do so far.
I find it to be very misleading.
I find it to be very misleading.
Last edited by nirm123 on Thu Oct 13, 2016 8:48 pm, edited 2 times in total.
- KVRAF
- 6287 posts since 8 Jul, 2009
I know you do. What I am trying to point out is that your and my perspectives on Neutron are very different - it comes down to opinion, not fact. For you, iZotope did not accurately describe Neutron and it does not meet your expectations. For me, iZotope accurately described it and meets my expectations. You might want to stop trying to find facts in all this because there are none - it's subjective. The only facts are, it doesn't work for you. so uninstall it and get back to making great music!nirm123 wrote:plexuss, I'm afraid you're not following me but I really can't make it any clearer than I managed (or not) to do so far.
I find it to be very misleading.
#NONFR Check out my music at Bandcamp
Free Streaming!
Free music with your support on Patreon | Youtube: Music of Plexus Videos (music videos) | Youtube: Plexus Productions (audio related) Stop whining. Make music.
Free music with your support on Patreon | Youtube: Music of Plexus Videos (music videos) | Youtube: Plexus Productions (audio related) Stop whining. Make music.
-
- KVRian
- 714 posts since 7 Sep, 2012
this is not subjective, you ignore facts which leads you to think otherwise, doesn't mean it's subjective.
I'm suggesting simple testing which results can be measured, objective tests and welcome more people to do it.
subjective conclusion isn't relevant to here and wouldn't make me write there is anything misleading in here.
throw in some vastly different vocal and guitar audio tracks and see the exact similar settings, if you understood from the manual and the videos that this is the claimed behavior...I would find it hard to argue with it.
edit EQ in an extreme ridiculous way and then apply the plugin, see how it will still suggest the exact setting.
throw in some instances in series and see that it reapplies exciter and compression again and again and similar EQ configuration.
I call it facts, you call it subjects feeling, let's see what other people will find as we obviously can't achieve a progress here
I'm suggesting simple testing which results can be measured, objective tests and welcome more people to do it.
subjective conclusion isn't relevant to here and wouldn't make me write there is anything misleading in here.
throw in some vastly different vocal and guitar audio tracks and see the exact similar settings, if you understood from the manual and the videos that this is the claimed behavior...I would find it hard to argue with it.
edit EQ in an extreme ridiculous way and then apply the plugin, see how it will still suggest the exact setting.
throw in some instances in series and see that it reapplies exciter and compression again and again and similar EQ configuration.
I call it facts, you call it subjects feeling, let's see what other people will find as we obviously can't achieve a progress here
- KVRAF
- 25049 posts since 12 Jul, 2003 from West Caprazumia
You are clearly mistaken - apparently it selects the same type of processor-presets and processor-chain according to the type of instrument and Assistant-preset and then adjusts the processor settings depending on the analysis.nirm123 wrote:
please correct me if I'm doing something wrong here, but for me it seems that it just analyzes the audio input only to understand what kind of instrument it's and from this point it will use the exact preset which doesn't assist much.
What I did was to use four instances of Neutron on an Addictive Drums bus, each with a very different MIDI pattern. The Chain was always the same:
EQ first: four bands of Proportional Q plus Baxandall low and high - no Dynamics
different: frequencies, Q, gain
Exciter next; full, three bands, all pure Tape, same drive and blend settings on all instances
different: crossover frequencies
Compressor 1 next: three bands, Peak, Auto Gain and Auto release, yadda yadda
different: crossover frequencies and threshold
Compressor 2 last: single band, Peak, Auto Gain and Auto release, yadda yadda
different: threshold
- KVRAF
- 25049 posts since 12 Jul, 2003 from West Caprazumia
For the fun of it I now tried it with the same drum-pattern but instead two different drumsets: same kind of different results on both instances as with different drum-patterns.
-
- KVRian
- 714 posts since 7 Sep, 2012
not really mistaken, I wrote :
"throw in some instances in series and see that it reapplies exciter and compression again and again and similar EQ configuration."
you wrote :
"What I did was to use four instances of Neutron on an Addictive Drums bus, each with a very different MIDI pattern. The Chain was always the same" (and then you showed quite similar configurations).
don't know about you, I would immediately fire an assistant like this...
"throw in some instances in series and see that it reapplies exciter and compression again and again and similar EQ configuration."
you wrote :
"What I did was to use four instances of Neutron on an Addictive Drums bus, each with a very different MIDI pattern. The Chain was always the same" (and then you showed quite similar configurations).
don't know about you, I would immediately fire an assistant like this...
- KVRAF
- 25049 posts since 12 Jul, 2003 from West Caprazumia
Still drums, but different Assistant-preset means still the same chain, but then some of the other parameters change.
-
- KVRian
- 714 posts since 7 Sep, 2012
Can you please try it on two absolutely different vocals ? I did and it generated the same results as well.
again, maybe I'm indeed mistaken
I'd love to be proved wrong here.
again, maybe I'm indeed mistaken
- KVRAF
- 25049 posts since 12 Jul, 2003 from West Caprazumia
You also wrote: "it just analyzes the audio input only to understand what kind of instrument it's and from this point it will use the exact preset" and I was answering to that - for reference see what part of your post I quoted. And what you wrote there is not correct. Whether some other parts of your post are true bears no revelance to this.nirm123 wrote:not really mistaken, I wrote :
"throw in some instances in series and see that it reapplies exciter and compression again and again and similar EQ configuration."
If you write:
"Two plus two is four.
The moon is bigger than the earth"
, then your claim that the moon is bigger than the earth does in no way become any more valid because you correctly predicted the outcome of adding two to two in the previous sentence.
-
- KVRian
- 714 posts since 7 Sep, 2012
I still think that what I wrote about analyzing data method is correct (and it is relevant to the conclusion which directly derives from it and other tests I performed and your tests results showed as well).
- KVRAF
- 25049 posts since 12 Jul, 2003 from West Caprazumia
Please explain in what way you meant something different than to say:
the only kind of analysis the Assistant does, is to predict the type of Instrument and then it only loads the preset for this type of Instrument and does nothing further than that.
I am very curious about other possible interpretations of that sentence of yours.
the only kind of analysis the Assistant does, is to predict the type of Instrument and then it only loads the preset for this type of Instrument and does nothing further than that.
I am very curious about other possible interpretations of that sentence of yours.
