Compression setting

J

Jack Real

New member
Hi

I just bought a TL Audio 5050 (Mono Valve Preamp & Compressor) and I want to check if my setting is correct. For the compression, I used the following parameters:

Threshold : -15 DB
Ratio : 3:1
Attack: .5 ms
Release: 40 ms
Knee: Hard

This might look too aggressive but that was the only way I could get 3 DB reduction on the peaks. Any suggestions on what I might do wrong ?

Thanks,

Jack Real.
 
"If it sounds good, it is good" applies here.

These certainly don't look like extreme settings to me, for what its worth.

Most importantly, what are you compressing? If this is for recording vocals, I'd consider soft knee, and a lower ratio (1.5:1 or 2:1), and then applying another round of compression at mixtime if you think its justified. Just an option, of course, and what you need depends completely on what you're recording and what you want it to sound like after compression.
 
Jack Real said:
Hi

I just bought a TL Audio 5050 (Mono Valve Preamp & Compressor) and I want to check if my setting is correct. For the compression, I used the following parameters:

Threshold : -15 DB
Ratio : 3:1
Attack: .5 ms
Release: 40 ms
Knee: Hard

This might look too aggressive but that was the only way I could get 3 DB reduction on the peaks. Any suggestions on what I might do wrong ?

Thanks,

Jack Real.
Click to expand...

Get real, Jack! (sorry, someone had to say it)

Correct for what? What's the instrument? What's the peak level coming into the compressor? What's the average level? What are the lowest levels? what are you trying to do? If there was a correct setting for compression, they'd just put a button for compression instead of all those knobs. ;)

-RD
 
Hi

Sorry, I forgot to mention that I was recording vocals. The setting that Sonusman was recommending for vocals in his article was:

Ratio: 2:1
Attack: 15 ms
Release: 60 ms

but I find that this is not enough compression for my type of songs.

Thanks,

Jack Real.
 
This has to many variables even for vocals alone for anybody to say one setting or another is correct. The style of song, the vocal delivery, the consistency of the singer, etc.
For instance it could be a intimate song with slow, breathy phrases or it could be loud, barked, in your face vocals and you would be looking at drastically different settings.
 
Jack Real said:
This might look too aggressive but that was the only way I could get 3 DB reduction on the peaks. Any suggestions on what I might do wrong ?
Click to expand...

the first thing to look at here is what is your peak level before compression?
 
I see. Any recommendation that someone makes is just a starting point to get you in the ballpark. Keep in mind that the amount of compression you end up with, ignoring attack and release for the moment, is dependent on both the ratio, and the relationship between the input siganl level and the threshhold. If you set the ratio to 4:1 or even 8:1, but most of the program level is under the threshold setting, then your not going to see or hear much compression going on. Only the peaks that exceed the threshold will be affected. Try your 3:1 ratio again, and then sweep the threshold from 0db, where you should be getting no compression if you have no overs, down until everything is getting compressed, and then back up again. Then bring it down till your getting the level of compression you want. Do the same with the ratio. Practice driving it off into the ditch, then back on the road again. Then you'll start to get a feel for what's right.

Regards,
RD
 
Also, no one's asked yet what KIND of vocal he's compressing. Metal? Jazz? Pop?

Threshold : -15 DB
Ratio : 3:1
Attack: .5 ms
Release: 40 ms
Knee: Hard
Click to expand...

These settings will actually produce a fairly gentle compression (despite the seemingly low threshold and the hard knee).

For a vocal, 3:1 ratio is...well, it's not too gentle, but I've certainly seen harder ratios before. I think what's really keeping you from getting much compression is your timings. 40 ms release is almost nothing, which means that your signal is not staying inside the compressor for very long. Of course, sometimes this short a release time is exactly what is called for. Nonetheless, trying upping it, and it probably wouldn't hurt to increase your attack time as well in order to smooth out the transient response. Vocals don't hit instantaneously like drums do, so you might try an attack of something more like 10 ms, or even more if you want that slight "pump" that can sometimes be cool. For release, start with 100 ms, but that's actually still fairly conservative. I often use as much as 200 ms release for vocals (yes, 1/5 of a second). Of course, that only works if you keep your ratio low, otherwise you'll hear too much pumping (by low, I mean 4:1 or less).

It's also worth noting that such principals are doubly true if you use your compressor for mastering. For mastering, I've always gotten the best results out of using an extremely low ratio (like even less than 2:1 sometimes) and accomodate with a VERY low threshold (like -30 or even lower). Basically, this means you get the signal into the compressor really early, and don't have to compress as much as a result.

My two cents. How you compress is entirely up to the style you're trying to record, and even more, personal taste.
 
Hi

Thanks for the answers and suggestions. The peak level is at 0 DB because I set the input gain of the preamp to have that (that's what the manual recommands). The vocal sounds good, it's just that we can hear the compression. I will continue experimenting, maybe increase the ratio and use a soft knee or maybe I will compress less and reserve another pass of compression for the mixing phase.

Thanks,

Jack Real.
 
Justus Johnston said:
Also, no one's asked yet what KIND of vocal he's compressing. Metal? Jazz? Pop?



These settings will actually produce a fairly gentle compression (despite the seemingly low threshold and the hard knee).

For a vocal, 3:1 ratio is...well, it's not too gentle, but I've certainly seen harder ratios before. I think what's really keeping you from getting much compression is your timings. 40 ms release is almost nothing, which means that your signal is not staying inside the compressor for very long. Of course, sometimes this short a release time is exactly what is called for. Nonetheless, trying upping it, and it probably wouldn't hurt to increase your attack time as well in order to smooth out the transient response. Vocals don't hit instantaneously like drums do, so you might try an attack of something more like 10 ms, or even more if you want that slight "pump" that can sometimes be cool. For release, start with 100 ms, but that's actually still fairly conservative. I often use as much as 200 ms release for vocals (yes, 1/5 of a second). Of course, that only works if you keep your ratio low, otherwise you'll hear too much pumping (by low, I mean 4:1 or less).

It's also worth noting that such principals are doubly true if you use your compressor for mastering. For mastering, I've always gotten the best results out of using an extremely low ratio (like even less than 2:1 sometimes) and accomodate with a VERY low threshold (like -30 or even lower). Basically, this means you get the signal into the compressor really early, and don't have to compress as much as a result.

My two cents. How you compress is entirely up to the style you're trying to record, and even more, personal taste.
Click to expand...

Hi

Thanks for the answer. I do rock music and I usually have a lot of instruments so compressed vocals are easier to sit on those mixes. The compressor on the TL Audio 5050 has only 2 attack times (.5 ms and 20 ms) and 2 release times (40 ms and 2 sec). So, if I want more a elaborated setting, I will have to use my Lexicon which has 8 settings for those times instead of 2. I will try the slow release to see how it goes.

Thanks,

Jack Real.
 
Jack Real said:
Hi

Thanks for the answers and suggestions. The peak level is at 0 DB because I set the input gain of the preamp to have that (that's what the manual recommands). The vocal sounds good, it's just that we can hear the compression. I will continue experimenting, maybe increase the ratio and use a soft knee or maybe I will compress less and reserve another pass of compression for the mixing phase.

Thanks,

Jack Real.
Click to expand...

Jack,
it's always a good idea to go light on compression during tracking. A lot of engineers will set up more for limiting than compression while tracking, with the limiter preferably doing nothing, just catching the one or two peaks that would have gone over. You just never really know what the vocal track is going to need until it's sitting in the fully developed mix, and then it's too late to un-compress it. (you can try an expander, but it doesn't get you back to a clean track). As you get to know this new, and probably excellent compressor better, you'll probably get a better feel for when and how much you can compress on the way in, but at first be careful with committing it to tape (or disk).

-RD
 
Robert D said:
Jack,
it's always a good idea to go light on compression during tracking. A lot of engineers will set up more for limiting than compression while tracking, with the limiter preferably doing nothing, just catching the one or two peaks that would have gone over. You just never really know what the vocal track is going to need until it's sitting in the fully developed mix, and then it's too late to un-compress it. (you can try an expander, but it doesn't get you back to a clean track). As you get to know this new, and probably excellent compressor better, you'll probably get a better feel for when and how much you can compress on the way in, but at first be careful with committing it to tape (or disk).

-RD
Click to expand...

Hi Robert

I agree with you. I will keep the compression to the minimum while tracking and then do a second pass during the mixing phase to make the vocal sit on the mix.

/Jack Real.
 
the sousman settings are a good start, but don't be afraid to really dig deep with the threshold setting.

there really are no transients to speak of in vox, so probably you'll be fine letting loose on the attack time some.

the problem being, you're trying to regulate the amount of compression with the attack time............ that's something you should do (at least in this situation) with the threshold.
 
If your peak level really is at 0, than a threshold of -15 should impart a pretty fair amount of compression, even at lower ratios.
 
You must log in or register to reply here.
Back
Top