Compression consult???

[cooliofranco]

I make all my tracks in Reason and Audacity and am trying to finalize some tracks for a vinyl release. I've read about compression, understand the basic concepts and parameters and am struggling a bit with solutions to a few basic and very common problems to someone new to mixing/mastering game.

 

If anyone out there is able to Skype with me for 30-40 min that would be best. That's a big favor though, so I'll outline the state of things here, and if anyone wants to chime in that would be great.

 

At some point I got the advice to lay off compression until the end. The argument is that compressing stuff as you go often leads to an arms race in which you compress individual elements to get them to sound good, and you inevitably come to an end product that is over compressed. Most of my drum machines have some compression on them on their way out, but it's minimal. I now know I really should be treating the kick/snare/metal all differently, and I think I sort of the do that with the 4 channel strategy, but yeah, in the future I'll be smarter. The synths: the basses and leads and pads mostly have no compression.

 

So here I am at the end, I'm running the main out through 4 channel parallel compression and then a 4 channel limiter. I'm finding I have 2 common and related problems

 

a. still too quiet

b. peaking transients

 

My sense is the second issue is affecting the first. If I look at the waveform of say, produk 29, I can see nice fat peaks, with nothing peaking out, and of course it sounds perfect. Also, the compression doesn't raise the lower stuff so that the song is nicely dynamic- there's no constant drone of jacked low sounds.

 

I understand it's hard to really help me without diving into the tracks because compression solutions, complex, context-dependent, and somewhat individual. But I have a few general questions about the solutions I've come up with.

 

1. What are some good guidelines for the parameters of a 4 channel parallel compression? For all of them I have very low thresholds, attack and releases at 40ms and 370 ms. For the first 3 bands I have ratios of around 2:1 but for the highest band I'm at 8:1 to push offending transients, which are mostly in that range, down. I basically like the sound I have uncompressed, so I don't feel like I need to add punch to individual elements. I really just wanna get this louder while preserving the uncompressed sound.

 

1. Is hard limiting OK? Many of the transient problems are from snares. I can help that by changing the attack on the snare sounds but I can also just hard limit the final waveform, and then compress the whole thing again lightly, at say 3:1 with high threshold, short attack and decay.

[TubularCorporation]

There's some decent basic information about mastering for vinyl here:

 

http://productionadvice.co.uk/vinyl-mastering/

http://www.resoundsound.com/mixing-for-vinyl-dont-fall-for-these-traps/

http://www.thevinylfactory.com/vinyl-factory-releases/how-to-master-a-record-noel-summerville-vinyl-dubplates-the-clash/

https://www.gearslutz.com/board/914005-post5.html

 

 

I'd still say if it's at all possible, send it to someone with experience mastering for vinyl AND experience with the kind of music you make. Having a fresh set of ears is really helpful (even if it's just someone else with some production experience whose judgement you trust giving you notes it's a huge help).

 

EDIT: I know this doesn't address your questions specifically but there's helpful stuff in there.

 

As far as specific tips I'm not the one to ask, I've mixed and mastered exactly two vinyl releases and both of them were long before I had any idea what I was doing.

 

I can say from that experience that excessive hard limiting really does sound a lot worse on vinyl.

[sheatheman]

Post a clip of the music here.

 

Nowadays I compress minimally. I find a bus compressor on the master suits my needs. Sometimes I compress drums before they get to the master and compress the master again, but lately I have found you can solve most of those problems by setting the levels of individual drum hits well.

 

Judicious setting of levels and the isolation of each hit in the frequency spectrum or the chronological spectrum with master compression will do what you need.

 

The more you mix, the less you have to compress.

[cooliofranco]

Post a clip of the music here.

 

Nowadays I compress minimally. I find a bus compressor on the master suits my needs. Sometimes I compress drums before they get to the master and compress the master again, but lately I have found you can solve most of those problems by setting the levels of individual drum hits well.

 

Judicious setting of levels and the isolation of each hit in the frequency spectrum or the chronological spectrum with master compression will do what you need.

 

The more you mix, the less you have to compress.

I've seen a few people say you HAVE to mono the low ends, but it seems like folks disagree on exactly where to draw the line. Some say everything under 500Hz and others just under 100Hz. Any ideas?

[sheatheman]

500hz is way too much. Those low mids sound lush with stereo chorus.

 

I don't do the mono/stereo split to the master tho, I just pick the thick sounds/tracks and mono them.

 

Truthfully, most of my tracks are always mono, minus one layer, plus maybe stereo send fx...

[cooliofranco]

yeah i've been messing around with where to put the cross over. i'm actually doing it in a kind of weird way: i'm just spltting the audio, in split A I EQ out everything over 90 and in split B I EQ everything out over 90 and check the spectrum to make sure it's working. then split A, <80, I mono, and split B >80 I keep stereo. it's good because it gives me a lot of control over where I mono and it seems to sound decent without getting rid of the low lush spreads, like you can still here the reverb on the bass because you get it form the stuff that's over 80. but, it's possible its doing something weird. ill send you some clips.

[sheatheman]

I think what you are doingdoubling the audio and processing each partis the main way of doing it, especially in a DAW, and 80hz is the most common point I have heard of, but some people might do 120hz if you don't spend much time in the true sub range.

[cooliofranco]

yeah i was trying to do it in reason and the splitter which split them at 80 didn't really work that well. it must not be a hard split because i could hear and see in the master EQ <80 stuff bleeding through and vice versa. I tried to add some EQing to the splits to clean it out but the EQ curves weren't sharp enough and i was losing too much. so now, i'm just giving audacity the final waveform, splitting that into two. and then doing the EQing there because i can draw sharp line at 80: giving the low 2 baths of lowering everything over 80 (after 1 iteration i can still hear high stuff) and 2 baths for the high, monoing the <80 track, then mixing 'em back together. sheesh.

[cooliofranco]

yeah i was trying to do it in reason and the splitter which split them at 80 didn't really work that well. it must not be a hard split because i could hear and see in the master EQ <80 stuff bleeding through and vice versa. I tried to add some EQing to the splits to clean it out but the EQ curves weren't sharp enough and i was losing too much. so now, i'm just giving audacity the final waveform, splitting that into two. and then doing the EQing there because i can draw sharp line at 80: giving the low 2 baths of lowering everything over 80 (after 1 iteration i can still hear high stuff) and 2 baths for the high, monoing the <80 track, then mixing 'em back together. sheesh.

cleaning out the <80 for the stereo section is tough actually. if i keep re-doing the filter it bring the level down in that area, but i have to repeat a lot. this can't be correct.

[nicklimegrove]

I make all my tracks in Reason and Audacity and am trying to finalize some tracks for a vinyl release.

 

so what you're saying is that, without no experience in using, for example, a compressor, you're doing the actual vinyl mastering yourself -- the very last step before the cutting lathe is engaged... right?

 

As someone who's grown up on punk rock and DIY ethics, I'd say: if you feel confident enough, by all means do it.

 

However.

 

I'm not sure if the two things you've learned by now -- (1) what a compressor does, and (2) that you're supposed to prevent major L/R differences below, say, 200 Hz -- will be enough to give you a result that you're pleased with. I'd even go as far as to say: vinyl mastering is the one step in production that you'll better leave to someone whose job is to do that job (and pretty much nothing else). It's not only good to have a second pair of experienced ears to listen to your material (and, if necessary, give you advice for revisions prior to entering the mastering stage) -- that goes for mastering for any medium. The main reason that you'll want to consider spending 50 to 100 bucks (per side, per track, YMMV) on getting a pro to master your material for vinyl is that there's a tremendous number of peculiarities of vinyl as an audio medium that need to be addressed. And, with all due respect, this isn't something that the vast majority of us is able to handle (or even has the required equipment for).

 

The good news is that, in case I got you wrong (i.e. in case what you're doing is merely preparing your material to hand it over to a dedicated mastering facility), some of the points that have been discussed here are points that you don't have to worry about at all.

 

You may, for example, make sure to keep everything < 200Hz mono as early as in the tracking or mixdown stage. But you may just as well not do it, and leave it to the mastering engineer to take care of it. You may even omit any (!) treatment of the mixdown. That is, once you've summed your material down to two tracks, don't do anything to it. No compressors, limiters, stereo enhancers, ›vitalizers‹ (in case you've treated your material to death). In fact, this is exactly the approach that most dedicated vinyl mastering facilities recommend. At least the first two that come to my mind do (both of them, I think it's fair to say, are well-respected and beyond any doubt regarding the know-how that they provide).

 

 

Should I limit or compress my mixes?

Many producers like to mix through a compressor, to achieve a certain sound. As a rule of thumb, however, we advise not to use a limiter on the mix bus – even if used minimally.

We use very high quality dynamic tools. When used properly in mastering, these tools guarantee the best possible sound quality, the shortest signal paths and minimal side-effects with highest levels.

To put it plainly: Even if for certain purposes you do some pre-mastering of your mixes, the files without the pre-mastering will make the best basis for mastering.

Anyone who has heard the difference will know what we are talking about!

http://www.calyx-mastering.com/en/faq.html

 

 

Should my master be pre-mastered?

Not necessarily. It can be useful, if the pre-mastering brings out your intentions. It is not useful as preparation for a vinyl cut. We have excellent equipment in the studio, and we prefer to see through the full process here. In-depth knowledge and long experience of the vinyl cutting process - together with test cuts - are vital to determine which treatments are really appropriate. We advise caution with compressing and limiting, especially with maximizing or optimizing plug-ins, as they may cause additional problems for the cut.

http://www.dubplates-mastering.com/faq.html

 

The more you've done to your mixdown already, the less leeway the mastering engineer will have for their own work.

 

So in a nutshell, what I suggest is: consider taking the following steps (I take it that it's a DIY release without a label doing most of those steps for you:

(1) pay someone who will listen to your material, give advice for revisions, and master and cut your vinyl

(2) carefully listen to the test pressing you'll probably receive and stay in touch with the mastering people concerning potential revisions you want from them.

(3) be happy with the results

(4) go through all the hassle of getting your vinyl master actually pressed (which is an entirely different subject and involves test pressings etc. again)

(5) one fine day, receive a pallet of 500 or so of your records at your doorstep

(6) go through all the hassle of getting those records out to your listeners

(7) be happy with the results

(8) be very happy that, for the time and energy you've needed in steps 4 to 6, you've saved so much in the mastering stage.

 

PS: in looking for a mastering facility, try to prioritize geographic proximity over reputation. This will give you the opportunity of getting an ›attended session‹, which can be very instructive.

[TechDiff]

I basically like the sound I have uncompressed, so I don't feel like I need to add punch to individual elements. I really just wanna get this louder while preserving the uncompressed sound.

 

I mean, if you like the sound, why change it?

 

The number of old tunes I have sat on a hard drive which I like, but basically ruined with excessive compression at one stage or another. I think I'm getting better with it now, but that's maybe because I try to use it as little as is possible/not at all.

 

Excessive filtering and splitting into ranges sounds like a very bad idea, I can imagine you running into all sorts of weird phase issues by doing that, which would only serve up more problems with final volume.

 

Similarly, if you're not super confident working with compression, diving in with a multiband jobby also might just confuse the whole thing further.

 

If you're really running into weird issues, struggling to get separation or whatever, then maybe this is all mix related stuff. Better to check all that first.

 

Maybe a personal thing, but with mixing I like to try and make any adjustments smaller the further along a signal path I get. So like, if it's one channel with one source on its own, you can afford to be a little creative with your effects chain, because you're only affecting that one thing. As you start to group onto buses, you should only try to make small changes, as everything you do is changing multiple sounds. Once you get to the master bus, only make tiny adjustments as everything you do affects the entire track!

 

It's also a good idea to make sure every signal is arriving at its next destination (bus/master/FX send) nice and cleanly. Make sure nothing is clipping in or out, and make sure there is plenty of headroom. Lots of signal processors respond very differently depending on input signal strength, most respond very poorly to overly hot signals

 

My issue with compression is that it can very quickly lead to an unnatural sound. I guess anything can really.. It's all processing, adding more processors into your signal path will make the result sound more processed

 

If that's what you're after, then no worries! Personally though, the mixes I've been most happy with are the ones with the least amount of processing, and more time spent nudging levels.

 

Anyway, if this is going on wax then tbh I think it's all a bit redundant. I would've thought the mastering engineer will want the cleanest, non-compressed or limited file you can give them, with plenty of headroom.

[cooliofranco]

i've temporarily abandoned finding a compression strategy (4-band, 8-band, parallel, various options), and am mostly going through and tweaking individual elements, trying to clean out channels ("why is this LEAD making stuff at 70 Hz??").

 

the main issue i have is that i have a lot of reverb and spreading on bassy stuff because i love that sound. it seems like a consensus that lots of low-end stereo effects should be avoided for vinyl so i'm getting rid of a lot that, which is sort of a bummer for me. after minimizing those effects in the track, i am doing that trick of doubling the final waveform, EQing away >100 in 1 copy, monoing that, and in the second copy EQING away <100 in another and leaving all the stereo in that, and mixing back together. i'm meeting with a real pro to talk about compression, but i think trying to recreate the fat transient-less peaks of 4 bit 9d api are not going to work for me at this stage

[hma]

regarding mastering/vinyl mastering this is really inspirational and informative video, highly recommended

[cooliofranco]

thanks for all the advice. for anyone looking at this after asking the same question, here's how i ended up going forward

1. took off almost all compression in Reason (after chatting with a pro and showing him the available compression options he said i'd be better off not using any of it, and sending mixdowned premasters o someone else who has better tools), except for some slight compression on drum machines only. if used, very minimal: ratios < 4:1, lowest attack and decay

2. 5-6db headspace for all tracks

3. all kick drums and most basses are mono or center-panned

4. if i felt i absolutely had to widen/chorus/reverb a bass synth, i split the signal in 2. In A I EQed out everything over 100Hz and sent the signal as mono, in B i EQed out everything under 100Hz and kept it as stereo (and with whatever effect)

5. tons of transient hunting: inspect mix down in audacity and go back into reason to fix any velocity that caused high peak. this took a lot of work and felt like whack-a-mole at times but i think was critical

6. put amp envelopes on kicks and snares- these just come in so hot and i was getting so many annoying transients. by setting attack on amp envelopes i could make the peaks thicker not as spiky, which allowed for everything else to be louder and more balanced. i'm a little worried about this because it's not something i've seen folks talk about but it sounds good to me and the engineers i've played this for haven't seem concerned.

7. eq every drum sample. most of the my drums are from tape recorded vintage drum machines and for whatever reason they often had a ton of signal in areas where i think they shouldn't (e.g. hi hat with signal at 150Hz??, EQ it out).

 

anyway I learned a ton these past few weeks and am really psyched for all this to come out on vinyl on Sleepers Records in a few months.