mixing down

  • Thread starter Thread starter downonthestreet
  • Start date Start date
D

downonthestreet

New member
Just wondering if some experienced CEP users would please give me a few basic tips on mixing down.

I've posted a few times in the past about recording and now have my head around that, thanks to the patient and detailed explanations of some of you guys (like lpdeluxe and ozpeter). I'm now at the stage where I have recorded a CDs worth of tracks - mostly just vocals, drums, bass and guitars, using a max of 7 tracks, including sometimes a fix-up track or two where I've re-done flawed parts, applied Silence to the flawed section on the original track etc, so it sounds seamless and perfect(!!). I've set volume levels and panning for each track of each song, done NR and EQing etc on each track individually for each song, and am now ready to mix down.

I know I need to go to Edit> Mix Down To File > All Waves, and that I should cut the extraneous sections of silence off the songs at the beginning and end. That's about the extent of my current understanding. Specifically, I need to know the following, which I'll enumerate for clarity and ease of response:

1. I recorded in 16bit and I believe for best sound the mixed-down song should be in 32bit - correct? How do I change to 32bit and when (ie: before Edit> Mix Down To File > All Waves??)?

2. I want to have all the songs on the CD at the same volume level - what I mean is, when the volume setting of a hifi amp is constant, I want the sound level of each song to be the same (all the songs are full-on rock with full distortion...no quiet ballads). The songs were recorded at varying levels, so, to put it another way, how do I now ensure there is some constancy about the volumes of the songs on the CD?

3. What is the optimum volume level re 2 above for best sound quality? I have some CDs that are recorded so high that the hifi amp volume setting has to be turned down pretty low to avoid blowing the roof off the house, and although this music I am about to mix down and burn to CD is loud by nature, I want to have the CD volume at an upfront level, but not extremely loud.

4. Once I have done a mixdown, how do I ensure there are equal-length silent gaps between songs on the CD?

5. From analogue recording experience, I am sure I will not get the mix exactly right first burn. I assume that once I have done a mixdown, nothing is "fixed" and that I can simply make any necessary adjustments to balance between instruments, panning, effects etc in Multiview and Edit and do another mixdown using the original recordings of each track of each song?

Cheers and thanks in advance.
 
Thank you for the kind words. I can't give you detailed answers to all your questions, but here goes:

1. The issue of 32 bit should come up in the recording phase. I have my software set so that the default recording setup is 44.1KHz and 32 bit. Once I have completed all the work on a track mixdown I press F11, which brings up the conversion dialog box, and I dither everything down to 16 bit for CD burning. This sequence depends upon having a soundcard that works at 32 bits. The theory is simple: the higher the sample rate and bit depth, the closer to the original your music will be; but the CD standard calls for 16 bit. My solution is to delay conversion to the last minute.

2. I tend to experiment and burn CDs with different levels. You can't really use a single "magic" setting for normalization or compression because a song will sound louder, or softer, depending on the song it follows and which song follows it. I buy stacks of CDRs and expect to go through several before I get it right. Probably there are experienced engineers out there who no longer have to do this.

3. Massive Master has a lot of information about this: do a search. I record mostly acoustic music, and the overall level is probably lower than what you want.

4. Nero and CD Architect insert a 2-second gap by default between songs. There will be situations where you want a gapless seque from one song to the next, and there will be others where you want a longer gap (as when, for example, you follow a hard rocker with a ballad). The spacing is adjustable.

5. Exactly. The mixdown is saving a new file that incorporates the features you edited into the original tracks via the Pan and Volume Envelopes as well as reverbs and whatnot you have added in Edit View. You will retain the original tracks in the original session (with the qualification that, if you saved the effects, you can't go back after you've closed the session) and they can still be tinkered with. All you need to do then is to do another mixdown. If you want to save your original mix, save it as, say, R&R Deacon, and the first variant as R&R Deacon 1, and so on. If you save it with the same name as the first one, it will replace it. I do this when I KNOW I won't want to call up the mix I started with.

Hope this helps.
 
Congrads Bro,

This is the question of a dedicated and established CEP user...

In my opinion, you are eventually going to come to realize that you must tweak each mixdown respectively, keeping your eye on the meters.. Also, a good thing to look at is the frequency analysis..

You may about thing like the "Group waveform Normalize", but from what Ive heard, it isnt the best bet.*

Waves has a plugin called an "Ultramaximizer"...and I have no idea what it does.. :p ...im kidding ...

wait...no im not :(
 
Thanks for the replies, guys.

So, lpdeluxe, how much difference do you think it makes recording in 16 bit vs 32 bit? Is there a discernible compromise in sound quality such that I will need to start again and record everything in 32 bit (slap on forehead at the prospect)?

I should add that this is a bit of an unusual thing I'm doing: I have recorded (in 16 bit, to CEP) a finished mastertape already compressed, reverbed etc and archived on video mastertape, and have been adding bass parts and a few extra guitar parts that were never completed on the original mastertape recording. I have only applied NR (default setting) to the CEP stereo recording of the video mastertape, but have applied both NR and Normalisation to individual bass and added guitar tracks that I've recorded recently in CEP. When I mix down ready for burning to CD, given that I have recorded all this stuff in only 16 bit, will normalising the WAV file(s) of the completed song(s) create 'artifacts' (especially considering this will be a double-normalising of the bass and added guitar tracks, so to speak)??

On reflection, I think I should have left the stereo mastertape CEP track alone and not done any NR even, so that I could have done a NR to the completed WAV file only, then normalised it prior to burning to CD. Probably I should not have normalised the bass and added guitar tracks, either, so they wouldn't have been normalised twice. Would you agree that this would have been a better way to go?

Oh, sheisse, I think I sense I'll have to re-do the bloody lot!!! Nooooo!!
 
I recorded a lot of stuff in the beginning of my computer experience using 16 bit, just because I didn't know about sample rate conversion and dithering and so on. Think about it this way: if you like the way CDs sound, you're OK. They are all 16 bit.

32 bit recording is a luxury, but not essential. In some ways it's like burning premium gasoline in your Yugo, as opposed to putting regular in your 500HP Corvette. It's like upgrading your monitors: having mediocre monitors is not a problem when your ears are not developed, but once you (that's **you**, not someone posting on a discussion forum and, in the immortal phrase, clanging their balls together) can discern the difference it's time to move on.

I'd leave things as they are, for now. There are so many issues to master that bit depth is not going to be that important for now. You and I both know that there are people who obsess over such things, but I prefer to obsess over the music.;) It's like buying a $200 Porta-Potty and insisting on $200 cables to hook everything together. Get the basics down, then you'll appreciate the better methods. Training your ears is a slow, but sure process, and everybody starts from scratch. Another way of looking at it is that you don't want to be too precious about your recordings. Learn from them, and figure out how to make them better. After all, that's a lifetime project, and you'll never run out of ways to tweak your mixes or your mic technique or your gear or you name it.

This is not hands-on advice, I know. But to me, putting in the seat time in learning to mix and manipulate is 'way past bit depth in importance...but, of course, if you were recording at, say, 8 bits, I WOULD point out that you can do better!
 
Thanks for your advice, lpdeluxe - sage, as ever. I'll do the mixdowns, burn to CD and see how the songs sound. As you say, that's all that matters. If the CD doesn't pass my fairly picky listening test, I'll re-do the lot in 32 bit.

I agree that one shouldn't waste time being precious about nuances, but this is a project I've been working on for 6 months in conjunction with writing up and designing an accompanying website, and it's important to me that I get this recording as good as I possibly can, as it's going to be part of an "historical" compilation of originals dating back to 1977/78. Sorry - too much information already, no doubt. Suffice it to say, I do have perfectionist tendencies, but in this case I think I can sort of justify them!

Cheers, and thanks again for your help and patience.
 
That's a common disease here!

If your originals are on tape, it shouldn't take too long to re-record them. My first project, when I bought my recording computer 5 years ago, was to do exactly what you are doing, & burn CDs from 20-some years' worth of demos, rehearsals, and what have you.

The REAL problem I ran into was not so much sampling rate, as it was clarity. For the first time, I could hear the tape hiss, I could distinguish the subtle distortion, and I could hear noises that were never there on the tape! Over the next year or so I upgraded about everything I could get my hands on, and ended up with a digital mixer and fiber optics replacing the analog mixer and soundcard w/breakout box that fed unbalanced lines into the computer. I now have a much better system, so that's not a complaint.

A word to the wise: CDRs may deteriorate over time (and will certainly do so if a CDR is your only copy of something dear to you) so invest in an external hard drive or DVD burner to archive your goodies to, in addition to the CDRs. Keep the CDRs in jewel boxes or cases with opaque inserts, and keep them away from heat and light. As with most such things, I learned this the hard way...the upside being that, several years after saving "perfect sound forever" I was forced to repeat the process (with cassettes, 4-track cassettes, and reel-to-reels) but by then my gear was much closer to snuff.
 
lpdeluxe

Re: "If your originals are on tape, it shouldn't take too long to re-record them."

True, but once the original recordings are transferred to digital format in CEP, I'm adding bass and guitar parts - that's what's taking all the time.

Re: "Over the next year or so I upgraded about everything I could get my hands on, and ended up with a digital mixer and fiber optics replacing the analog mixer and soundcard w/breakout box that fed unbalanced lines into the computer."

Interesting comment. I am using an (anaolgue) Tascam 244 as a line amp for guitars, bass and vocals, and have the Creative Audigy 2 Platinum soundcard/module, and to my ears the transferrence of instruments and vocals to digital sounds quite ok with this set-up. My original mastertape was top quality video, and this seems to have transferred well direct to the soundcard (although my video player intermittently drops out momentarily at times, which means I sometimes have to re-record a song several times before I get a take without any static interference or dropouts - this is annoying!).

lp, would you mind detailing the specifics of your set-up? Which digital mixer are you using, and could you elaborate on the "fiber optics" part.

Also, I'd like to know, pls, whether I am feeding "unbalanced lines" into my computer, going by my description of my set-up above?

Cheers
 
Let's see if I can tackle all of this...

When I got my computer, I was gigging with an Alesis Studio 12R, a small rack-munt analog mixer. I set up my computer around a Lexicon Core 2 sound card, which featured a "break-out box" that accepted up to 4 1/4 unbalanced analog inputs. With this I could route my Tascam 103 cassette deck or one of 2 2-track reel-to-reel decks into the Alesis, then into the breakout box into the computer. "Unbalanced" means that the cables had one conductor w shield: for short runs there is no great disadvantage to doing this.

I also had a Tascam Porta-Studio 2, with 4 analog outputs, that I would plug directly into the breakout box. Right away I noticed high levels of noise and tape hiss, and I further noticed that, when I recorded something through the Alesis and played it back through the same mixer, there was serious additional fuzziness. I had been pretty unaware of that when I was performing live, and using my various tape machines, but the digital recordings brought out all the flaws.

Along the way I got a pair of halfway decent nearfield monitors, which made things worse! I could now hear LOTS of problems!

At the time, Tascam was blowing out its TM-D1000 digital mixer for $500, so I got one, and luckily scored a pair of the rack ears AND a very nice rack. The Lexicon sound card had digital ADAT Lightpipe I/O in addition to the analog breakout box. The Tascam had TDIF digital outputs; a TAD-IF converter from Tascam solved that, and converted the 8-channel I/O of the TDIF cable into a pair of fiber-optic cables: 1 8-channel In and another 8-channel Out to and from the sound card. While I was at it I bought a Lucid Word Clock distributing amp to keep everything synched up.

I am now a serious convert to the Lightpipe, and have added a fiber optic patchbay to my motley collection of gear. No noise, no degradation, no interference from radio stations or CBers or fluorescent lights, and I can record or play back 8 channels simultaneously.

A word about word clock: when you have more than one digital audio device, the sampling process needs to be synchronized between them, with one device designated the "master" and the other(s), "slave." As currently set up, the mixer is Master, and the sound card and Tascam US428 I use for a control surface are synched to it. My Lucid box is really overkill, because, once I got into things, I learned that the word clock stream is carried in the ADAT data stream. Still, if I expand to more digital gear in the future, I'm already set for the word clock synch.

With all these changes, noise went down to practically nothing (although there is always residual electronic noise in any piece of equipment). At the same time (remember, we're talking about 4 years of upgrading) I got a pair of M-Audio BX8s, and demoted my old monitors to merely playback duty.

And I still have to clean up the bottom end (up to about 100Hz) on non-bass parts due to the acoustic response of my room. If I don't clean up each track, the bottom end gets muddy.

It appears that your soundcard uses unbalanced inputs (which is not at all uncommon, and, as noted, no particular issue for short cable runs). I am currently using a Terratec EWS 88 D, which, in addition to the ADAT I/O, has an analog mini-stereo jack so I can play CDs back through the mixer.

Here is a website that has comprehensive information of balanced, unbalanced, S/PDIF, TDIF and you name it: http://www.rane.com/digi-dic.html

But the fact is, if you are happy with the results, then don't worry about it. The problem is that, as you get more experienced, your hearing becomes more acute, and sounds that were perfectly fine a couple of years ago become annoying. That's really just an aside, though: there's no cure!
 
You must log in or register to reply here.
Back
Top