Stereo Tool

Hehehehouuuuu!

Be prepared for my ...comeback in a few days with many many reports and tests ON AIR
for now i am swimming on the beaches of Crete island to the borders of Greece...near Africa ..sahara...

Here the temperature today was 23 degrees celcius and 100% sunny

off topic ...end /;

I'd like to mention again that "vibrating vocals" and the related "hole punching" may not have really been a "problem", or at least not what it was thought to be.

Why do I say that?

It's not to slag on / attack someone.

It is to have a discussion about the situation and come to a greater understanding. Perhaps I am the one that is mistaken, but with the songs I've been pointed to thus far, I think I see a misunderstanding made by others.

The "Mina" track (I'd like to know who she / they are...I did a search and could not figure it out), and Adele's "Set Fire to the Rain"...

There's a vocal technique that is both liked and disliked by people for centuries. That technique is called "vibrato". This is something that a vocalist can do, or a vocalist can try NOT to do. In the two tracks listed above, the vocalists did use vibrato. This means that the sound that is being heard is something that the vocalist did that was not edited out in the recording mastering process. It is a part of the actual track.

Now, when an effort is made to "fix" those sounds, is it really fixing something?

Some vocalists use vibrato, so the vocals will seem to vibrate. In my mind, this is not something to be "fixed", but if it is deemed as needing to be "fixed", then the "fixing" should be something that is optional, because if it is not optional, it means there is a forced alteration to the original input based on someone else's subjective dislike of vibrato.

@Brian: We're not talking about vibrato. What we are talking about is really a form of distortion, which disappears completely if you remove or reduce the bass before clipping. BUT I should add that I have difficulty hearing it myself in the latest ST versions. So it's probably not really an important issue - and I'm working on other things now. (I will take another peek at the code though when I have some time, because I have heard this Mina song on the Omnia 9 and there there's no distortion AT ALL, not even when listening with headphones.

Exactly my point - not really an important issue.

However, certain people seem to be pushing this up as an issue of greater importance than I personally feel it should be. I don't know if perhaps they were using an older version or if there is an actual misunderstanding and the buzz being heard from distorting bass and extreme clipping that induces distortion is being heard overlaying vocals in certain tracks and being misinterpreted as "vibrating vocals"...

wow, there is no distortion at all in latest beta.. move on..

It is important. And it's really vibrating vocals. I know what causes them, it's not a problem in the bass itself. It only happens on sounds with phase issues (sharp short spikes, mainly female voices and trumpets). Before Bojcha reported it to me the first time I had never even noticed it (probably because *all* the Dutch stations that I usually listen to have them too) - but once you do, they are really really annoying. (I've learned by now to never doubt Bojcha, lol). So I really want to get rid of them.

The only thing is that right now, at least as far as I know, the only processor that suffers less from this than Stereo Tool is the Omnia 9 - all others have more and often even much more voice vibration issues. The real weak spot in Stereo Tool is currently the multiband compressor (Bojcha did a test with BBP for multiband + ST for clipping and he said it sounded much better than either separately, and that makes sense). I've also made a new version for a completely different market (nobody here knows about that yet), they were raving about it - but they also pointed at one weak spot, the multiband compressor.

On top of this, I kinda hope that a new better multiband compressor could potentially even solve the whole issue.....


I'll still take a look at the code some time soon to see if there's anything I can still improve. But right now I think I should really focus on the multiband section. And for that I need new parameter handler code. So that's what I intend do first. Unless I suddenly get a new idea (I'll keep this in the back of my mind, that usually works much better than continuously looking at it).

Aliens?

Aliens are not smart enough !

Ah, ringtones without brickwall compression. Cool!

:sigh:

Excuse me while I whip this out...



I'll be explaining two issues here. The first dovetails into the second.

People were convinced that the burst in "The Pretender" was the bass, but it never was. It was the "crash" / "attack" of the percussion, along with the loud and sudden "wh" sound with the resumption of the chorus ("What if I say I'm not like the others..."). All of the frequencies involved with the spike in that track are mids / low highs.

The reason that happens is because the AGC goes up way, way, way too fast. That's why I put in the gating that I did. Even without your protection mechanism, if you gate the AGC to put a brake on how fast it increases the gain, you don't have much of a spike, and what is left you can handle with appropriate clipping (stronger clipping from 3-5.5K). To be honest, without the gating, I find it very annoying that the AGC dropped the volume down so aggressively, and then the volume rises back up so fast. This causes too much up and down jerkiness, not smooth gain control.

So, people were convinced / convinced you that the spike protection was a greatly needed feature...

Now we're talking about "vibrating voices". You're claiming now that they don't have anything to do with bass, but have to do with loud spikes. Hmmm. What was the spike prevention thing in the AGC again?

You're also now talking about phase issues. Well, the default behavior of the HPF is to use Phase Linear if there are no artifacts, otherwise do non-phase-linear. When you move to NPL, that introduces phasing issues....which you then say lead to vibrating voices. :scratches head:

Further, and I might be mistaken on this, as I may have sniffed glue again, but I think you said that Bass Boost uses some non-phase linear tricks. If I'm remembering correctly, then there's another thing that deals with bass that has phasing issues around it, but yet the vibrating vocals have to do with spikes (protection in AGC) and phasing, but not bass at all (HPF and bass boost may introduce phasing issues). We won't even mention the phase rotator... Oops, I mentioned it.

Now, you can get grumpy with the sarcastic / flippant tone here if you wish, but I've toned it this way to try to get you to see that you're talking in circles. Up is down and down is up. This is because you're taking a lot of direction from someone who appears to be firmly entrenched in maintaining the status quo in "The Loudness War". Yes, that's where a lot of money comes from (the broadcasting segment), but I think some caution needs to be taken to balance the status quo (volume controls to 11) with emerging ideas of softer and more dynamic content (1770 / EBU R128).

Table the work on loudness and the FM Transmitter sections for the moment. Seriously look at multiband. That's where the most gain (pardon the pun) can come from, particularly if you allow people to control compression/limiting parameters on a per-band basis. I already know what is causing my preset to lose vocal strength on certain tracks. Those tracks, like "The Pretender", have very forceful vocals, thus with a global setting of "compress", when those bins are looked at they are determined to need to be softened.

BTW, "female voices and trumpets". The trumpet issue is something I've read about in my Google travels, but can't remember specifics. I think it can be addressed by the extra clipping / reduction of EQ / reduction of level of that particular band (bin).

As for "female voices", vibrato is something that a vocalist can do if they choose to, and I still think that actual artistic use of vibrato is being confused for "vibrating voices" in how it is being presented to you.