Headphone correction

[Alan Valdez]

I own Waves Nx and copied the eq to use it with MFreeformEqualizer it works great and doesn't consume cpu like the Nx, would be nice to have something like MPhones meanwhile.... I will keep mixing with my MFreeformEqualizer

[Opal Two]

Is waves NX mix room seriously just an EQ curve? I thought it at least simulated crosstalk.

Anyway, AirWindows Monitoring is a free tool that does this too, and I imagine anyone could whip up an MXXX preset to do it. It's not a dark secret.

I'd rather see more cool unique tools from Melda than this personally I love that their plugins are so different and customizable, I'm not sure what you could customize in a room sim/headphone corrector (that you can't do with other tools, probably even DAW stock plugins.)

[vectorwarrior]

I agree, there's no benefit to the user of Melda making this plugin over anyone else. I personally prefer when melda work on unique tools or things that can integrate into their suite of tools better. For instance, something that adds additional use or functionality to MXXX or MSoundFactory. As a headphone correction plug is put at the end of the chain, it doesnt really aid melda's larger plans.

[Burillo]

Nx is not an EQ curve, it's a spatial reverb simulator (so it does have cross-talk by virtue of modeling your head being in a room in front of the monitors). it does have headphone correction module (and an EQ curve for that), but its main purpose is not that.

[Alan Valdez]

Waves NX its an eq correction for the headphones, it has a button where you can disable the room there's where I copied the eq curve, have been mixing with it a lot. It's a cool idea to do this on Mxxx now the question is... How can I simulate speakers in front of me with Mxxx? Convolution?

[ElVincente]

I copied the sonarworks correction for my headphones with it too. I prefer to use this as I can only use sonarworks with one computer... Works great, although copied room correction seems not to work so well. Maybe something sonarworks is doing with signal's phase can't be replicated with MFreeFormEq, I don't know.

[Burillo]

if the reverb they use is non-linear, you aren't going to be able to capture it with an IR, but you still should be able to get very close with one. i suspect you're doing something wrong.

[JeffLearman]

if the reverb they use is non-linear, you aren't going to be able to capture it with an IR, but you still should be able to get very close with one. i suspect you're doing something wrong.

Right! And I'd be surprised to learn that it needed to be nonlinear. Convolutions perfectly imitate what sound sources do in a passive room, including phase. Frankly, you should be able to capture an impulse response from NX and use that in any convolution plugin. Reduce the CPU usage by choosing the right length of impulse response; for this application a pretty short one should be fine.

The length of the impulse response is the length of the reverb tails. To model a room that's good for mixing in, there shouldn't be long reverb tails! Figure one foot per msec; do you expect to get reflections from a wall more than 10 feet from you and the virtual monitors? If not, 20 msec should be plenty. (This is based on theory. I haven't tried it so take it with a grain of salt, but the theory is solid.)

[Alan Valdez]

I copied the sonarworks correction for my headphones with it too. I prefer to use this as I can only use sonarworks with one computer... Works great, although copied room correction seems not to work so well. Maybe something sonarworks is doing with signal's phase can't be replicated with MFreeFormEq, I don't know.

Me too! I prefer to use MFreeformEqualizer, the room simulation is better with the normal NX than the Abbey Road Studio 3 this one includes a weird feeling to me, like isn't normal or natural, also use it at the end of any of my mixes to check in "other room" hehe!

[Alan Valdez]

if the reverb they use is non-linear, you aren't going to be able to capture it with an IR, but you still should be able to get very close with one. i suspect you're doing something wrong.

Right! And I'd be surprised to learn that it needed to be nonlinear. Convolutions perfectly imitate what sound sources do in a passive room, including phase. Frankly, you should be able to capture an impulse response from NX and use that in any convolution plugin. Reduce the CPU usage by choosing the right length of impulse response; for this application a pretty short one should be fine.

The length of the impulse response is the length of the reverb tails. To model a room that's good for mixing in, there shouldn't be long reverb tails! Figure one foot per msec; do you expect to get reflections from a wall more than 10 feet from you and the virtual monitors? If not, 20 msec should be plenty. (This is based on theory. I haven't tried it so take it with a grain of salt, but the theory is solid.)

Thank you! that helps a lot! going to try that! probably the nonlinear can be emulated with some LFO doing small movements in the background!

[MeldaProduction]

NX is just an convolver, but it tracks the head and selects the IR based on the location. Somehow I bought it and never really used it though... will try again...

[Starship Krupa]

If I may interject here: we have the Preset Exchange, and if headphone EQ curves are easily implemented in existing MeldaProduction plug-ins like MFreeFormEqualizer, I strongly encourage everyone to upload their curves to the Exchange.

I've experimented with copying Sonarworks' and Toneboosters' curves into the EQ's I have, like MEqualizer and MAutoDynamicEQ, but it's not yet close enough to upload a preset. I need to get better with MAnalyzer and copying and pasting the analysis. Even with the fixed number of bands I like the results so far.

[MeldaProduction]

Well, you can certainly do that, but there are certain ethical and legal factors...

[Starship Krupa]

Well, you can certainly do that, but there are certain ethical and legal factors...

I thought for a while about the ethics matter before I posted. None of the products mentioned use only EQ curves to achieve their results (at least they claim so). Analysing just the EQ curve from one plug-in's preset and recreating it in another company's preset doesn't seem too terrible in these days of amp (and everything else) modeling.

None of the EQ curve cloning I've tried has resulted in my cans sounding the same as they do with Sonarworks or ToneBoosters' excellent products.

Understood that it's your job to raise this issue.

[MeldaProduction]

Actually I think it is pretty much only the EQ curve. But who knows... Everytime I tried, a simple IR indeed did the trick.