Compress or Limit? Or both?

I am kinda with Bit on this one... kinda<G>.
 
I love the sound of a wide dynamic range - it is not always practical, but when it is...
 
Out of habit more than anything else I will use a limiter on the final stereo mix, but not on the 2-mix itself, unless I am aiming for that SSL sound. I usually apply some limiting in Wavelab or Sound Forge when I am polishing the final stereo wave file of the mix. I suppose I would end up with approximately the same result if  I applied it to the 2-mix, but that is not a habit I have to break<G>!
 
I use compression as an effect more than anything else when I am working with sample libraries. I use it more for it's intended purpose on live tracks, especially electric bass and fingerstyle guitar, and sometimes on vocals. But I've learned to use it after the fact, non-destructively, and just manage levels coming in.
 
So that's my take - dynamics processing is not as necessary to overcome shortcomings of FM exciters, magnetic tape or vinyl anymore. But it remains a very useful effect!

Thanks for your take on the subject Bill.
 
One thing I can say for sure is that most of my stuff, how ever compressed or limited, does still have a lot of dynamic range. I've experimented with harsh compression to get that large blocky waveform, but I've never been keen on the results.
 
Main use now is to tame those odd peaks here and there. Even though I rarely record real instruments, soft synths, when summed together, can produce clipping peaks here and there. So, as I have alluded to above, it's me being lazy, rather than taking time to tone down the offending peak during mixing. I could even create a volume envelope to do the same (which is what I had to do on my latest upload to SoundCloud).
 
cheers 
 

+1 for Kotelnikov! On the rare occasions that I do use compression on the master bus, it'll be Kotelnikov. It's the only place I don't automatically put Pro-C on as the first choice.
 

The problem with limiter -> compressor is that the compressor could cause overs that the limiter couldn't protect you from. It's OK to use the compressor's makeup gain to bring up overall levels and then have little to no gain in the limiter.

Compressor > Limiter it is then peeps.
 
Many thanks for your time and patience 
 
And of course, I will still try stuff out anyway 

clintmartin
Look out for those inter sample peaks. If your going export a mix use Pro-L's over sampling and isp detection. It will matter on mp3s.

Hi again Clint, yeah, I spotted that when I watched the Dan Worrall tutorial on Pro-L. And it really does matter, like you say. So ISP switch goes on automatically now 

clintmartin
Pro-L is a very, very nice limiter. Everything Fab Filter does is Fab...and pretty!

Aye... And they have Dan Worrall doing their tutorial vids 

I had seen that before and the comment of being proudly digital, emulating nothing, and taking advantage of all digital has to offer is rather catchy.

Dan does rock. I enjoy listening to his videos.

synkrotron
 
Main use now is to tame those odd peaks here and there. Even though I rarely record real instruments, soft synths, when summed together, can produce clipping peaks here and there. So, as I have alluded to above, it's me being lazy, rather than taking time to tone down the offending peak during mixing. I could even create a volume envelope to do the same (which is what I had to do on my latest upload to SoundCloud).
 
cheers 
 

Hi synkrotron,
 
Sorry I'm a bit late to the party here, but in my personal opinion, I'd like to see you control the peaks instead of trying to tame peaks with a limiter or compressor. You'd be astonished at how little processing you need to do when you do all the right things. I don't use much limiting until I master. Even there, I'm not going crazy and always maintain dynamics. You want to be able to totally enjoy a mix with barely nothing on it. That's when you know you got it right. I'll give you a few examples as to what I mean.
 
Sometimes transient attacks on a note on a particular instrument can cause the peak. Instead of just limiting that or compressing it away, see what may be causing it. If a bass guitar has a spot where the note seems to lash out, A compressor to smooth it or even a transient designer type plug can remedy the problem. It's better to soften the blow over possible over-processing.
 
Granted, we may have had to use a compressor in that situation, but by adjusting the attack of that compressor or transient designer, you've done very little to the sound other than to trim the transient that may be the cause for the peak. Other times you can split the clip at the transient and lightly fade it so it's not as apparent. Though the simple way would be to throw a limiter on, this is how we can degrade our audio if we're not careful. I have never used a limiter on an individual instrument track. The reason being? I can always control what is going on without one. I have found in all my years doing this....that less is more in a good way. That's not to say limiting is bad or any advice you received above is bad. I'm saying in my opinion, it is better to find out WHY we have something peaking and it's best to fix it without loading limiters or compressors up to where they can negatively affect the audio.
 
Sometimes a snare drum may have too much rim shot or "crack" to it. This ramps up peaks like crazy. The transient shaper can help this with one simple move to lessen the hit....and again, a compressor can do the same thing. Now that's two times I've mentioned those tools. Let me try to further explain the method to my madness.
 
When you get a bit more advanced in this field, you learn the many different ways you can use these tools. Compression for me is used quite a few ways, but the two most popular are for actual compression of an instrument, and then for effecting an instrument.
 
Compressing is just that. We keep the instrument in line without artifacts so it doesn't go over a certain threshold. The problem with this is people don't use it to compress the right way and end up with bad artifacts. Compression used as an actual effect is another animal. I mentioned the transient shaper. Compression as an effect was doing what the TS does now since the compressor came with attack and release options. With the right attack and release, you can literally make an instrument have more pop...crack...and transient attack. Like a snare drum cracking harder. I actually put some videos out showing how transient shapers and compression can be used in the same way. Transient designer type plugins just concentrate on how hard or how subtle a note or drum hit can be heard. It's basically the attack and release of a compressor without the compression part. So it's not as easy to mess your material up. Try the transient shaper in Sonar. It's quite good and can control peaks.
 
But a compressor can also remove transient attacks too...which is another way we can use it as an effect. This controls peaks in those nasty snare drums or even if a bassist uses a pick or pops and slaps a bit too hard. To me it's better to control or lessen the transient that hits instead of using a limiter....or even a compressor. But most of the time it's because the compressor or limiter is used to the extreme. Like there are a few people on here that know how to use a compressor in all of its glory. I'd never try to talk them out of it. It's the people that don't know the power of the compressor or limiter that need to be careful. With the tools we have today, it's easy to slap something on and get some sort of fix. But at times, it's really not fixing anything. It's masking a problem and can degrade the audio.
 
Other times we can remedy peaks by using eq. At times a transient or peaked instrument may be out of line due to an eq setting being too extreme. This usually happens when someone is not familiar with how eq really works and they just tweak until something sounds good by itself when soloed. When they allow the instrument to enter the field of other players, it sounds bad and they wonder why. A bass guitar that blasts out too much low end on a particular note can be fixed by adjusting the eq on that part only or by automating the sections that may need less of the low end he is putting out due to the pull on the string or the string note itself lashing out. It's better than compressing the heck out of it or using a multi-band compressor. However, there are times when a multi-band compressor can work wonders policing special circumstances to where you:
 
a) may have allowed a recording that should be re-recorded but you may not have that luxury
 
b) have a player/client that just wants to sound the way they want to sound while not understanding the instrument may be offending/disrupting the entire mix.
 
Sometimes we need to process using the tools we have....but mostly this is due to performance issues with players or even the instrument they use. Some drum sets just sound bad when mic'd up so we have to process them to compensate or even sample replace....some electric guitars and bass guitars just sound bad because of how they are made, how they are set up. (or not set up) So we have to do our best to get the best sounds out of the instruments. Sometimes some good old manipulation is the key. Other times using a compressor or limiter may be the answer.
 
But with electronic instruments or samples, you really shouldn't be running into any massive peak problems that need the attention of a limiter unless you are going for a specific sound. Again, I'm not saying don't use a limiter....I'm saying if you don't have to use something, try your best not to and find out where the source of the problem is coming from, and try to remedy it without the possibility of over-processing. If you know how to use the tools I've mentioned, they can definitely make a difference for the better. But we always want to know the source of the issue before we just put a band-aide on it. This helps us to fix things....and more importantly, teaches us when we should use a sound...and when we may want to choose another. :) Best of luck!
 
-Danny
 
(edited for spelling and added a few additional things)

So Danny inquiring minds want to know; What comp are you using on the 2 buss?
 
Tom

How Danny,
 
Thanks a lot for you advice. There's quite a lot to read and take in there, so I'm going to take some time to digest before I say more... I've got to go to bed now... Been slaving over a hot DAW all day (sometimes being out of work has its advantages).
 
cheers
 
andy

Jeff Evans
+1 to Danny re using an editor to sort things out BEFORE you go slamming dynamic effects over stuff.  I do it all the time.  A program like Adobe Audition (Cool Edit Pro 2) is excellent for doing this sort of stuff.
 
The great thing too is when you tame the peak heights down to the rest of the stuff you can often add rms gain to a whole mix too for free.  What I like about the way Audition handles a peak is that it leaves everything down the bottom alone and just limits the height of the peak.  And nicely too.  You certainly won’t hear it.
 
Over tracks you may have a track that varies a bit in level all the way through eg a vocal track or a bass part.  The worst thing you can do I believe is to slam some form of dynamics processor over the whole track to sort it all out.  The problem with that is the processor will be working hard on some bits and hardly at all on others so it becomes too inconsistent.
 
A much better thing to do is open a vocal track in an editor say and I put a VU meter over the whole thing and just edit the louder bits down and the softer bits up.  Then you can put a compressor over the vocal track but now you can set it so it is not working hard at all and with a low ratio.   Low ratios produce a bigger sound. And hence you can turn those parts down and still hear them nice in a mix.  In this mode you will find after editing the compressor will be nice and consistent over the whole track now.  (same in mastering too)
 
Another reason why I like the fact the waveforms in Studio One reflects clip gain adjustments.  You can split tracks and adjust a lot of this by eye and it way fast.  Another way of doing it.
 
And hello to Danny too.  Have not heard from you in a while.  I have been very busy these days over in the Presonus forums going on about V3.

Well said Jeff. :) Hello to you too! Hope you're doing ok brother. Doing great here. Been busy also. Got that Midas all set up, built my new room and have been having a blast with all my new toys. Amazing how good those Midas pre's are. If you ever start thinking about one....don't procrastinate. That console was one of the best purchases I ever made. I'll have to show you some pics of the new room. It came out so good! Talk soon.
 
-Danny

Great stuff...what tools are included with Splat to evaluate and correct...aside from Comp/Lim.
 
In other words what is a "volume envelope", is it included with Splat?
 
You know what I mean...I think.
 
I have Alloy2 and Ozone5...and I think they have analytical SW on there...that is if that is what you are speaking to.
 
Thanks.

Andy,
 
Wow...I am such an idot. Of course I know what a Volume Envelope is...I just know it as Volume Automation...DOH!
 
Are you kidding me. my tracks are riddled with those.
 
I don't have one track in any of my projects that don't have them. I control all spikes using that tool.
 
I even use them to mute parts where string noise from myself and the bass player are present, as a result of changing neck position on our instruments.
 
I will mute parts of drums tracks where there is bleed from other sources, or even on the overheads to control bleeds into parts of a song that need to be quiet at a specific point and quickly too.
 
Volume Envelopes are one of my biggest editing tool and one that requires an enormous amount of my time.
 
I am still learning about comp/lim subtleties. I mean, I don't use them aggressively...I just want to understand them better and therefore enable the ability to use them more effectively to achieve what I want.
 
Thanks