Great reminder that nothing is free. Trying to solve one problem introduces another and sometimes it’s not worth the trade off

@panorama_mastering

25 күн бұрын

Bang on.

@behtashmiri1283

24 күн бұрын

Sorry may I ask u would not recommend it even on stems? Because I use sound radix phase align on kick and bass guitar for more coherrance, is it going to ruin them? Tnx ​@@panorama_mastering

@DavidPeck

16 күн бұрын

@@behtashmiri1283phase alignment is not the same as phase rotation

1:50 minor correction, you say it’s nanoseconds but that’s actually the symbol for microseconds.

@panorama_mastering

25 күн бұрын

Yeap! You're right, I completely brain-farted when I saw that.

Perfect timing. I remember being so confused about this concept when that Adam Video came out back then. I tried it once on a mix back then, and found the unaltered version better, as I felt that it messed up the low end, and didn't ever consider this technique again. Then some weeks ago, I stumbled upon it again, and tried to comprehend what is going on with all the knowledge I gained over the years, and it still just didn't make any sense to me. So many people point that out in the comment section of that video too. Good to have your analysis of this now

@panorama_mastering

25 күн бұрын

Yeah; I wish when videos like that go out, actually in fact, when anyone puts out a video they take the time to pair it with some critical thinking and technical deep dive. It could have saved many people some pain.

I once had a problem with the low end being a mess, but it was a puzzle why because I was applying Shaperbox Volume to the kick and the sub and the bass. In Oszilloscope, the kick and the sub and the bass were all separated in time by Shaperbox Volume. In the end, I discovered that in my master chain I had a few plugins to boost the low end, and they were causing phase shifts and the collective effect of these plugins was actually moving the sub back in time so much that it was overlapping the next kick! This has led me to be more cautious about modifying the low end during mastering. I now try to boost the low end in the mixing stage so that I can see the effect of plugins in the time domain in Oszilloscope.

You should ask Daniel Robert Ford about it. I did some work with him years back and his master was insanely loud and sounded amazing. When I picked his brain about the loudness phase rotation was one of his secrets. I never brought it into my workflow either but whatever he is doing is definitely working maybe there is something there worth playing with

The "ringing" you are seeing in a spectrogram is actually below 20 Hz and above 22000 kHz. You can easily prove it by increasing the FFT size of spectrum analysis: the ringing will go lower and lower until it's below 20 Hz. I agree with the point, though: no process should be applied blindly, critical listening is key.

There’s no free lunch. Good insights as always man.

Another good rundown, please consider doing another version where you show the effects of the Adaptive Phase, while doing phase rotation

This always felt "instinctively" wrong to me so have never used this technique... couldn't explain why it felt like it could be wrong to use indiscriminately, so thanks for the investigatory work! The best "cheat" for getting free headroom is still clipping inaudible peaks, as you've mentioned many times before ;)

I have rotated the phase before because i like the sound. On 808s with snappy transients I think you can hear the difference

Phase rotation is best left to the broadcast processors of stations who want extra headroom during moments of speech and don't care if their music has bass that sounds boxy.

Thanks, best/Mathias

Finally, someone who took the time to look into this. I've seen this being applied as a "standard" but ... not knowing what's actually happening. Now, things start to make sense. Good job!

Hey! I loved running into this video, I was thinking about making a video of this kind seeing all the convo on facebook. On a small sidesubject, I feel like whenever someone is obsessing about peak and just measuring how big a number is, they are probably not listening yet to decay & spectral content! For me that's key, that's how we really separate transient from tonal information and that's a lot more relevant for M.E., besides being QC, imo we should be some kind of transient professionals. Learning how to listen to distortion on the slightest hint, looking for precise decay over the whole spectrum. I hope I can have you as a guest for future videos! Love the precision of your content. Stay cool ❣

@panorama_mastering

5 күн бұрын

Thanks for watching! I appreciate your observations too! Would love to jump on as a guest! Just let me know how!

Yea sir , thank you

It’s why experience engineers say use your ears 👂 😢. These things are not helpful because the results are different every time. You can gain headroom with a filter or loose it, you can get huge preringing from gentle curve mods or small from big curve mods. Phase rotations it self proves your video without a video, if one is paying attention, the curve shows its balance or it just sort of flips some what. More importantly, you know you’re showing its exact function or purpose? Or maybe I’m confused, what’s its purpose?

I’ve been using adaptive phase rotation on mono tracks (the whole track) for quite a while - when needed. Now I’m questioning everything!

@panorama_mastering

25 күн бұрын

There are use cases for this in mixing and broadcast for some signals! This assessment was purely through the lense of mastering.

You're doing God's work with debunking these myths and giving actual, detailed explanations! I actually tried and let adaptive frequency rotation go after I tried it on a full song and just noticed that snare hits don't only sound bad, they sound bad all over the place. It's a nice sound design technique for really specific, REALLY uneven waveforms. But just check the waveform of a bass guitar, if you have any at hand: I'd wager it won't be even across the negative and positive side. And there's nothing wrong with this.

I saw that Adam audio video. I was a bit skeptical about doing it as a step one application to slap onto a whole mix, as you should be with most things! Thanks Nicholas

I work nearly entirely digitally with synths and unless I use non-linear phase multiband processors or certain distortions, phase stays fairly symmetrical but it's something to keep an eye on as it will eat up headroom, especially if something wonky is happening to the sub waveform. I actually stopped using God Particle because it was offsetting the phase of my sub bass on the scope. That said, I've never thought to rotate the phase of the entire mix, I'll give it a shot because who knows, might sound interesting for some use cases or genres if it has a distinct "sound". That said, my target for most bass music is -3.5 through -5 LUFs and I get there with staged clipping and saturation in the mix so every little bit counts.

Interesting. I've been loving the phase rotation feature on the Fuser plugin from Mastering The Mix. Will definitely be going back and listening more closely to those transients now.

@akuratide3446

22 күн бұрын

Idd same thing here! Curious to examine it further with this new knowledge.

Yeah. Phase rotation is destructive. When I use it, usually I built a whole mastering chain around it to cover its caveat while keeping it. Sometimes Frankenstein comes out of it, but it serves its purpose. Usually I use it while overprocessing heavily compressed or low quality song, so it's not so harsh in the end. Thimeo Stereo Tool have much more complex options for phase rotation than RX I believe when mpre advanced options are turned on.

I saw that Adam video took! I talked about it to a greek audio engineer FB subgroup, consulting engineers that I value their opinion. They also told it was the rave in forums a few years back, as some sort of secret sauce, but they avoided because it causes smearing. I tried it in a master, found it made no difference - since it was 4 degrees or something. But, thing is, only with your video I was able to actually understand it. Thank you. These examples are to the point, reproducible, and once you can see the phenomenon in its simplest most visible form, then you are able to shift your focus to teach yourself how to listen to it. This was invaluable, thank you.

I would think your perception of transients in time would be frequency dependent, so that even if the phase rotation represents more time at low frequencies, you are also less capable of perceiving that difference. Maybe I'm wrong about that though.

It's great to know the why of this! Ta. I knew phase rotation was not a lossless process (for example if you rotate something 360deg it will not null, and as far as I am aware it will also not null once you re-rotate it back 360deg to it's "original" phase (at least from what I've seen when looking into watermarking audio using phase rotated watermarks). When you rotate back from even just 90deg there is a considerable amount of audio degradation, although it could be to do with other factors in the watermarking process, it would appear to be coming directly from the phase rotation as the more you rotate it (say 360deg in 4x90 deg steps instead of one step) before bringing it back to the "original" phase, the worse the degradation is.

To me THIS was the final tip to sound pro and I use it all the time now in RX though I avoid it if it doesn’t sound good on its own or doesn’t fit in the final mix. I find your tips interesting but find you sometimes dig too much into technical domain that I still don’t find relevant to music but still try to learn from you hoping I’ll get somewhere. If it sounds the same and gives more headroom I just do it

This pretty much lines up with my gut instinct, especially given what non-linear phase crossovers can do to a song if you aren't careful. I don't love the way you phrased things from 4:30 to 4:40 though, I've never thought of the sound as being in either/or frequency and time domain. Rather, I think of examining the frequency or time domain of a signal as two different ways of looking at the same signal at the same time, and the processing that I add to that signal might have more impact on one of those domains or the other (though often the reality is a little of at least one to go with the other in the case of something like EQ - whether that be linear phase with pre/post ringing or minimum phase with it's phase shifts). The way you explained it in this segment though, it sounds like you work on frequency, then time in a serial manner, and I don't think that reflects reality.

Very goood

Thanks for this! It all seemed a bit too good to be true...

can you please make a dedicated video on soothe2 and bloom from oeksound. I feel like with your expertise such a video could be so useful

Why didn’t you check the Adaptive Phase checkbox? The RX manual clearly states that the function now doesn’t have any impact on the sound. Are they wrong? And when unchecked, how would you know what amount of phase to use? Could you do the same test with Adaptive Phase? Also with adaptive phase checked: On a whole track, what do your professional ears say? Is there a difference/degradation?

@panorama_mastering

25 күн бұрын

It does say that, you're right. I think context matters. The use cases of using adaptive phase to offset dialogue or non-complex signals this would ring most true. The same test with adaptive phase, yielded the same results. The manual gives a few more hints as to how adaptive phase can produce some artefacts on musical material - docs.izotope.com/rx11/en/phase.html For my own analysis I'll leave this here RE Adaptive Phase Non-adaptive will offset the entire waveform (IN THE TIME DOMAIN) form and rotate it with one filter. Adaptive phase rotation, works the exact same except it's looking at the wave form in a series of "windows(Across the time domain still)" so it's still rotates the signal albeit adjusting rotation based on the offset in each window of time, so it won't alleviate the symptoms/side-effects of transient smearing.

It’s often bass and kick tracks that have lopsided wave forms as well. Have you tested adaptive phase rotation vs non adaptive?

@panorama_mastering

25 күн бұрын

Good question RE: adaptive phase rotation. Non-adaptive will offset the entire waveform (IN THE TIME DOMAIN) form and rotate it with one filter. Adaptive phase rotation, works the exact same except it's looking at the wave form in a series of "windows(Across the time domain still)" so it's still rotates the signal albeit adjusting rotation based on the offset in each window of time, so it won't alleviate the symptoms/side-effects of transient smearing.

@warpacademy

25 күн бұрын

If you have kick and bass with lopsided waveforms, you would want to first clarify whether it's DC offset or truly asymmetrical. Seeing as kick and bass take up the most headroom in the mix, I would carefully process them individually to center the waveforms if you need a loud mix. Otherwise, leave them be. It's not an issue necessarily for a waveform to be asymmetrical. Have you read the manual regarding adaptive phase rotation? iZotope says that it's intended for vocals and not to use it on a complex material as it causes artifacts. A great way to ruin a mix. Stay clear.

@panorama_mastering

25 күн бұрын

@@warpacademy Spot on! I've mentioned this in a later comment. This whole discussion for me is centered around mastering practices.

One concern I would have would be my transient information getting messed with, and the attack for things like kick, snare and other quick attack perc hits etc. getting messed up by blindly applying this to an entire mix!

The phase rotation you did caused pre and post ringing, is it not possible to do non linear phase rotation, as it’s linear phase that causes pre and post ringing?

I’d love to blind test this bit of pre-ringing vs extra limiting or clipping to decide what the actual net gains or losses are for what regular listeners would hear. Will pre-ringing be noticeably worse if it also means transients can stay slightly louder?

@panorama_mastering

25 күн бұрын

Fair question, yes please blind test this. But ultimately you could just blind test this in one sitting, it something that would need to be done on every use case of the tool. Ultimately as a mastering engineer, I look at my processes as what serves the music best, often, a process that splits hairs like this, and has inherent flaw/downside in the design won't pass my desk.

Does this also happen with oversampling? I feel like there’s a reason I avoid it a lot of the time and it sounds like some kind of smearing/softening effect

Putting aside the issue that many of the times times a peak will be caused in part by some or the other low frequency content, and partly defying the purpose of why we would rotate phase in the first place: Wouldn't it be an interesting concept to apply phase rotation only above a certain frequency and leave the problematic, audibly smeared content down low untouched?

@panorama_mastering

25 күн бұрын

Everything is relative, especially in the summation of a signal. To band-split, rotate phase and then resum would produce a non-linear band-split processor and have it's own set of artefacts.

More headroom? Take the premaster back a db lol

@panorama_mastering

12 күн бұрын

That won't change the crest factor.

This is semi related but I’m really curious how Sound ID Reference affects your listening setup in the time domain. They have normal and linear phase eq modes but those both must have a trade off and I’d venture to guess they’d smear all the transients coming out of your speakers. It would be really cool to see data and if getting a flat room response is worth the sacrifice.

@WorstViktorEU

15 күн бұрын

I used SoundID and also Dirac. I have no measurements or any scientific approach to the results, but I always felt SoundID only EQ'd the frequency response and Dirac did more processing than only EQ'ing the frequency response.

Dude, I'm so glad you did a video on this. Myself and so many other people have seen that questionable Adam video. I've had so many people say to me that they're doing this after seeing that video. It leads people astray by having people think that asymmetrical waveforms are universally bad and they need "fixing". Especially in the name of headroom. David Gnozzi from @MixbusTV also has a great video about this: kzread.info/dash/bejne/lnWuzNNvps_Km8Y.html. Spoiler alert: You guys are both on the same page. Don't use the Phase tool on a whole mix.

Of course nobody should do phase rotation across an entire mix, that's just ludicrous and anybody who proposes it doesn't know what they're talking about. At the same time anybody who doesn't consider phase rotation for individual tracks when there is an asymmetrical waveform driven by harmonics in a specific range of frequencies also doesn't know what they're talking about. I didn't hear where you indicated where phase rotation would be appropriate, do you know where it should be used, or could be used in an appropriate manner? If not, I would suggest that you look into devices that have done this for quite a long time like the Khan Symmetra, the Orban Optimod, and more recently the Kludge Rotating Limiter. Inducing group delay through a specific range of frequencies is incredibly effective on individual tracks and simple busses.

@panorama_mastering

25 күн бұрын

Specifically I was covering this in a mastering context, for the sake of gaining headroom across a mix.

@ThisGuyDude

24 күн бұрын

@@panorama_mastering Gotcha, thank you. I'm not a pro but to my understanding its not a good idea to do this broadly across the spectrum. Having said that, judicious selection of the center frequency of the all pass filter and the appropriate number of poles can focus the phase shift into a select range of frequencies (e.g. vocals) that produces group delay for that range and separates the fundamental from it's harmonics so they don't sum and pile up together, this providing more headroom specifically in that range. The Aphex 204 Aural Exciter Big Bottom does this in principle to the bass frequencies, as does the BBE Max.

@patrickalphenaar

23 күн бұрын

For sure i am one of those who doesn’t know were he talks about as I sometimes use “Phase Rotation” on a full mix to fix some issues with polarity to gain Headroom……. All problems caused by the Rotation Fix can be fixed again 😉. Just need to be creative and not just “too technical” all times…….. De-Clipping and proper Dynamic processing is a correct way solving any transient issues damaged by Phase Rotation. So if i will gain 3 or more DB of Headroom to do a proper Master, as the mix has some polarity issues i am glad to “fix” something after that process. But ofcourse like you said i am one of those “not knowing were they talk about”😁😎 Cheers and Happy Professionally Mastering! P.s there are no “absolute” rules within Mastering (if you send your mix to 100 different Mastering Engineers they all sound different). Being creative always might help a lot sometimes…….. Don’t limit yourself by rules which don’t excists……🎉🎉🎉

@ThisGuyDude

22 күн бұрын

@@patrickalphenaar Certainly. Did you read my second comment? I described use cases for a limited amount of phase rotation in a frequency dependent manner. Of course there are 'no rules' in audio, but there are some things that just don't sound good on 99.999% of use cases.

@patrickalphenaar

22 күн бұрын

@@ThisGuyDude Hi, sure i did read that. Sometimes its just a matter of choosing Headroom over some small issues caused by Phase Rotation which i know how to deal with. I Understand your point not just do it as an standard…….. i also wouldn’t do that. Cheers 🍻

Phase rotation, as a process, is for broadcast, to symmetrize a voice waveform exclusively for advertising, for maximum gain before clipping. For music (and especially a mix) it's impossible to apply without damaging something that doesn't need it. A mostly positive going waveform (acoustic pressure) is meant be that way.