multiband compressor recommendations - as in i need some

[AdamJay]

yea, and its wierd that all that latency is in the Loudness maximizer..

Possibly some sort of look-ahead limiting, the buffer size of which is dependant on the program material?

yea its definitely look ahead... i'm just surprised the Multiband Dynamics section doesnt induce nearly as much latency.

[randyh]

yea, and its wierd that all that latency is in the Loudness maximizer..

Possibly some sort of look-ahead limiting, the buffer size of which is dependant on the program material?

yea its definitely look ahead... i'm just surprised the Multiband Dynamics section doesnt induce nearly as much latency.

The multiband dynamics section only does "traditional" compression, expansion and limiting i.e. it's totally reactive. I wouldn't expect it to intorduce any latency.

-r

[AdamJay]

plenty of "traditional" compressors have look ahead, even ones that aren't multiband. Just look at Ableton's own CompressorII 0ms, 1ms, or 10ms.

I guess i should have said more specifically that i am surprised the difference in latency between RMS and Peak detection is non-existant.
Its interesting how one aspect of the Ozone is entirely reactive and the other is so drastically unreactive it creates so much latency.

just surprised thats all.
and pleasantly surprised as well.

[randyh]

plenty of "traditional" compressors have look ahead, even ones that aren't multiband. Just look at Ableton's own CompressorII 0ms, 1ms, or 10ms.

I really can't think of any traditional comps (theshold ratio etc) that have look ahead other than compressor II.

By traditional I meant the typical analog modelling thing, and since analog doesn't have look-ahead... Actually come to think of it, PSP VW has a sizeable buffer, but then it's trying to model a tape machine not just do compression. But most others (crunchessor, rcomp, gco, sonalksis, etc) don't have any look ahead.

I guess i should have said more specifically that i am surprised the difference in latency between RMS and Peak detection is non-existant.
Its interesting how one aspect of the Ozone is entirely reactive and the other is so drastically unreactive it creates so much latency.

I don't use ozone so i'm not sure what you mean about the rms and peak detection thing.

just surprised thats all.
and pleasantly surprised as well.

It's cool, i was just wondering why.

-r

[AdamJay]

I really can't think of any traditional comps (theshold ratio etc) that have look ahead other than compressor II.

they're out there... for a while i really liked the sound i was getting with CompressorII about a year ago. and i narrowed it down to the lookahead. So i looked for other traditional comps with it...

some that i found were these:
http://www.anwida.com/cx1v.asp
http://www.elementalaudio.com/products/ ... index.html (looks complex, but thats just the GUI, its actually quite simplistic, but powerful)

it is typically only found in Limiters, and i like limiters so thats probably why i was so keen to find other plugs with it. and since the Ozone's multiband dynamic had a sort of limiter character to it... i was amazed at its low latency.

[blakbeltjonez]

By traditional I meant the typical analog modelling thing, and since analog doesn't have look-ahead...

there are analog "look ahead" compressors/limiters - mostly used for broadcast or record cutting on variable pitch lathes where the lathe will cut tighter grooves with quieter passages to fit more music on a lacquer. they use a delay line (very small) for the compressed audio and the undelayed audio is the "look ahead" component.

there were a few feed-forward comp designs that were used in pro audio that had a "look ahead" function as well, i have an old A&H in my closet that does, i think API and SSL had models on the market too.

[Martyn]

I've been getting really good results by just using Live's compressor2 set to zero lookahead, sometimes 'tuned' to the kickdrum frequency, then following that with the excellent George Yohng W1 limiter. http://www.yohng.com/w1limit.html If you've not tried this one yet give it a whirl. Supposedly it has the same measurable response as the Waves L1, I've not tested that theory myself but it certainly works for me.

I recon getting your mix as balanced as poss without a compressor first is the real trick, then applying it sparingly to tighten everything up.

I've not got much joy out of multibands either. The best advice i've ever had about mastering is to get someone else to do it, or at least get a second pair of ears and monitors involved in some way.

[feyshay]

Did you find any latency with Neodynium?

[hambone1]

I've just started the tedious task of trying to match the levels & freqs of my library.. everything from 60s R&B/pop to up-to-date hip-hop. When DJing, I want the levels to remain constant.. not having to adjust the le els & EQ when going from relatively low-fi low-volume Supremes to hi-fi high-volume Nelly. It's basic private/corporate function VJing, so I'm not too concerned about losing dynamic range.

I had planned on remastering every track in T-Racks or something similar.

Would a multiband compressor do the same thing? I want everything to be more or less the same volume, and if a track is missing treble or bass, the plug-in or external box to add it on the fly.

For the older tracks that have no bass or treble, I'm warp-markering the track and dropping in percussion samples (hi-hat, cabasa, kick drum etc) to give it the bass or treble the original track doesn't have so it sounds hi-fi without destroying the original track. But I don't want to have to do this for every track.

Any help appreciated!

[randyh]

I've just started the tedious task of trying to match the levels & freqs of my library.. everything from 60s R&B/pop to up-to-date hip-hop. When DJing, I want the levels to remain constant.. not having to adjust the le els & EQ when going from relatively low-fi low-volume Supremes to hi-fi high-volume Nelly. It's basic private/corporate function VJing, so I'm not too concerned about losing dynamic range.

I had planned on remastering every track in T-Racks or something similar.

Would a multiband compressor do the same thing? I want everything to be more or less the same volume, and if a track is missing treble or bass, the plug-in or external box to add it on the fly.

For the older tracks that have no bass or treble, I'm warp-markering the track and dropping in percussion samples (hi-hat, cabasa, kick drum etc) to give it the bass or treble the original track doesn't have so it sounds hi-fi without destroying the original track. But I don't want to have to do this for every track.

Any help appreciated!

For getting the levels the same, us a mastering limiter like Voxengo Elephant, Waves L2, or the free L1 clone mentioned previously (it's free and good enough if you're not uber picky). Run your quiet tracks through it and make sure you're getting a little bit of gain reduction (like -1-3db). The modern hip-hop is probably squashed all to hell already so you probably do not want to do it on that stuff. Just look when you load it into your editor. If it looks like a big long rectangle that fills the audio window, you don't need any more limiting.

One of the cool things Elephant does is it shows you the RMS level of what you're processing. Thats more or less the actual loudness of the track. You could use this feature to get them all very very close. You may not want to bring the supremes to Nelly volumes though.

The limiting should be the last step. Proir to that you can try to match the frequency resposes of your stuff. The easiest way would be to use one of the spectrum-matching programs like halbar. You can also use a plugin like voxengo curveEq that allows you to capture a response curve and apply it to something else. You won't really have to think doing it this way.

The best way to match the frequency responses is to eq them manually, but that'll take a lot of time.

I suggest only messing with the old, low quality recordings. Just remember, some of the allure of those old recordings are their crappy sound. You might think twice about re-eqing the supremes. If you make everything sound to samish, it'll bore the listeners, consciously or not.

Again, I don't recommend a multiband compressor for this task. It will take a lot of tweeking to get right. You risk doing more harm than good. I think the method I described above will get you better results in a shorter amount of time.

Multiband compressors are one of those tools that people shouldn't use until they know enough that it occurs to them naturally when it's appropriate.

-r

[hambone1]

Randyh-

Thanks for the tips!

You're right.. I got caught up in brickwall/hi-fi mindset. It might make everything a bit samy and boring.

I added some additional percussion to an old Sly track.. not sure whether I liked it or not. Even though it was warped and matched up perfectly, it just didn't sound right.

Maybe I should leave well enough alone!

[linzatti]

I know they are a bit pricey but I use them in almost all my tracks. T-Racks and Sonalksis Multiband Compander/Compressor. They both have a warm and cripsy analog sound to them. Download the demos and open your wallet

http://www.sonalksis.com
http://www.ikmultimedia.com[/url]

[hambone1]

The Sonalksis CQ1 & DQ1 look awesome!

Might just be what I'm looking for...

[feyshay]

Hambone,
Have you considered using Spectral matching (check out Firium by Elemental Audio) to try to approximate equalization?

[montrealbreaks]

For getting the levels the same, us a mastering limiter ...
Again, I don't recommend a multiband compressor for this task. It will take a lot of tweeking to get right. You risk doing more harm than good. I think the method I described above will get you better results in a shorter amount of time.

Multiband compressors are one of those tools that people shouldn't use until they know enough that it occurs to them naturally when it's appropriate.

-r

If I can offer $0.02 on this;

I think that a multi band compressor, set up as a multi band LIMITER (maxed out ratio, threshold close to zero, fast attack and release on all three bands) would serve the purpose well. Then you'll get more of a sonic balance on your final tracks without the loss of hign end clarity that can result from pushed limiting.

But that's just me.