Limiter Shootout

[Kingston]

You want to end up with something that a label will see as an acceptable master...it must be very loud, but it must not be totally breaking down...you are pushing to the edge of an extemely fine tipping point.

Although I see where you're coming from, labels generally seem to have no idea of what is totally broken. Far too many records have proven that there is simply no control over "acceptable". The loudness war sometime a go progressed past the point of "anything goes as long as it's loud".

Basically, I could take your processed example, strip a bit more high RMS low/lowmid content and clip it further about 2 dB. It would still get okay'ed, sad but true.

[kilroy]

You want to end up with something that a label will see as an acceptable master...it must be very loud, but it must not be totally breaking down...you are pushing to the edge of an extemely fine tipping point.

Although I see where you're coming from, labels generally seem to have no idea of what is totally broken. Far too many records have proven that there is simply no control over "acceptable". The loudness war sometime a go progressed past the point of "anything goes as long as it's loud".

Certainly you make a good point. When you start to compromise on signal quality all of a sudden you have not only moved the bar by which this is measured, you have in fact changed the entire bar. Remember that one particular Smashing Pumpkins record? When I heard the distortion on that (all through it) I was stunned. Apparently that was Billy Corgan's idea, but it *was* accepted nevertheless. So as you say, if something that bad is OKed then one must consider whether a compromise of any sort is even necessary, or in point of fact, worth spending the effort to achieve at all...just shove the fader up to the DACs and giv'er till the meters are filled right up.

I guess in the context of this thread what I was interested in seeing a sampling of how far folks could get "to the edge" using the tools they have, with the mind that this might indeed be somewhat louder than what they would ordinarily accept without any outside influence, but not at the complete expense of rendering the track a totally unlistenable mess.

Err...a kind of high quality smashed track of sorts.

[Kingston]

Err...a kind of high quality smashed track of sorts.

yeh, the optimal compromise. I'm tempted to clip me tube preamp output transformers now, as they're slightly less intrusive (one might even say pleasing) than clipping the converters, or voxengo elephant.

Best successes I've had with both limiting and clipping together.

[kilroy]

That would actually be quite interesting to hear.

The big thing to watch on Bmanic's original wave is the snare sound. That element will display a marked tendancy to fold right up under heavy limiting...sink in, get mushy, lose snap. I consider such behavior as being a generally unacceptable limiting by product since that particular sound can contribute significantly to the overall perception of loudness.

I will be fairly impressed if a VST limiter can match the example I have posted here, but of course I'm entirely open to being impressed.

I should note that absolutely no hardware, or EQing of the original file was used in the harming of this song.

[Kingston]

Right now I'm in the middle of fixing the damn preamp. It's DIY and the specs didn't initially quite match my hi-end expectations. It's otherwise great, but there's still a +20dB 50hz gremlin spike hum above the noise floor. Urgh.. somebody get me lots of tin/mu shielding! Now!

Anyways, as soon as I find the culprit, I'll post a clip here as an "academic" example.

As far as VST limiters, I admit bmanic is far more qualified on the matter (I'm a mix guy, not ME), and I'm sure he's up for the optimisation challenge.

I also noted the transient happy snappy snare, most subjectible to nasty foldback/alias/harmonic distortions as a side effect from VST limiters/clippers.

[bmanic]

That would actually be quite interesting to hear.

The big thing to watch on Bmanic's original wave is the snare sound. That element will display a marked tendancy to fold right up under heavy limiting...sink in, get mushy, lose snap. I consider such behavior as being a generally unacceptable limiting by product since that particular sound can contribute significantly to the overall perception of loudness.

I will be fairly impressed if a VST limiter can match the example I have posted here, but of course I'm entirely open to being impressed.

I should note that absolutely no hardware, or EQing of the original file was used in the harming of this song.

One has to remeber that this little clip isn't mastered before going to the limiters. One of the first things I would make sure is to "tame" the spiky dynamics pre-limiting to get an overall more smooth/distortion free compromise. Also, the frequency response of that clip is not nearly optimal yet so fixing that would again enable me to take it quite a bit further. My guess is that one could get it to around -8dB RMS levels without the average john doe complaining too much.

Btw. your clip has some very pleasant pumping going on there. It's a bit too much for this clip but I'd love to get my hands on that for the bass+drum buss in a hiphop track!! Awesome!

Cheers!
bManic

[bmanic]

so it's no longer a limiter-shootout, but a limiter-shoot-to-kill.

How about somebody develops a shoot-to-kill limiter then?

- bManic

[ddummer]

ohh! Can I try two:

http://www.tinbrooketales.com/hosted/Li ... ard-el.wav
http://www.tinbrooketales.com/hosted/Li ... oft-el.wav

(right click and save if stuff isn't happening)

edit: re-did the samples so the level is equal to the original


//Daniel

[kilroy]

One has to remeber that this little clip isn't mastered before going to the limiters. One of the first things I would make sure is to "tame" the spiky dynamics pre-limiting to get an overall more smooth/distortion free compromise. Also, the frequency response of that clip is not nearly optimal yet so fixing that would again enable me to take it quite a bit further. My guess is that one could get it to around -8dB RMS levels without the average john doe complaining too much.

I figured you were not done tweaking this clip, bmanic...pretty much sure of it really. It was fine as is was for my general purposes so I left everything alone. I was shooting for a fairly hot track with more or less acceptable residual maximizing artifacts as an example of the possibilities for software dynamics processing. Your clip, with it's prominent bass and snare, was a perfect platform to demonstrate this. These two critical elements still retain fair definition and snap even after a significant level boost. In general, it seems difficult to find a software limiting solution that does not collapse the mix before reaching levels similar to this.

Btw. your clip has some very pleasant pumping going on there.

Actually...that's the tricky bit.

It's a bit too much for this clip but I'd love to get my hands on that for the bass+drum buss in a hiphop track!! Awesome!

Aye, well...here's another go, just for a lark...slightly different response on the detectors.

more merciless butchery

[kilroy]

ohh! Can I try two:

http://www.tinbrooketales.com/hosted/Li ... ard-el.wav
http://www.tinbrooketales.com/hosted/Li ... oft-el.wav

Ooooo...summat new in the works, then Daniel?




Jolly good news, that.

[BLUE-S-MAN]

Maybe this

[kilroy]

You know, it has to be said...Daniel's GUIs are just a fabulous balance of form and function. That thing just looks so...


so...


...beautiful and efficient.

[ddummer]

Sorry to let you down in the future then...

//Daniel

[kilroy]

Sorry to let you down in the future then...

Pffft...I'll get over it mate.



Just do what you do.