Track to Track Definition @ 44dot1kHz

[Bob's Mods]

The last nagging problem with digital recording for me is the track to track definition issue. I understand and have worked out solutions and methods for dealing with common problems that can arise in digital recording but track to track definition is one I must work around.
I am my own everything, writer, recording studio engineer, musician, janitor, singer, etc. My common method of recording is track layering. Although we may all have an unlimited track count, I have found this to be almost meaningless as piling on the tracks causes more and more loss of track mix definition. I have found that recording at 96kHz rather than 44dot1kHz reduces this problem greatly but I don't like the change in sonic character 96kHz gives me nor do I care for how much resources it uses. I've also tracked at 48kHz and that does help provide for better track to track definition in a mix but not as good as 96kHz. My preference is to work at 44dot1kHz and massage the mix to bring out the tracks.
My rules are to center the drums and bass. From there I try to isolate vocals by panning to one side and panning a single dominant instrument on the other. This appears to provide the greatest clarity for mixing. By mixing in a way that allows for a dominant track on the left and on the right channel, I find I can simulate a well recorded tune.
It is my belief that if I were able to record a full band in one take plugged into a mixing board mixed down to two tracks this problem would not exist. I suspect mixing in analog would provide for a better track to track definition than digital mixing does. I believe that analog mixing is probably superiour to the digital mixing I am forced to do.
Has anyone else been fustrated by this annoying problem? It is not impossible to work around but you must adapt your writing to make best use of the digital medium compromises that must be made.
Are there others out there who layer tracks like myself and have discovered this to be a problem too? Or....is mixing that much of a craft? There is a lot of newgroup buzz on the black art of mastering but I have found mixing to be more of issue mastering. I believe too much is made of mastering buts its really mixing where the bullet hits bone for the home recordist. If your mix is right, then mastering only adds that extra few percent.

Bob the Mod guy

 

[Track Rat]

I think the root of this type of problem (folks tracking by themselves with only one signal chain) is that the signal chain is identical on every track (same mic, same preamp, etc). As you add tracks, they all have the same sonic signature so this problem snowballs. I do feel mixing inanalog (as I do) also helps.

 

[Farview]

Mixing is that much of a craft. Without carful attention to the individual tones, panning, and dynamics of each track, the mix won't come together.
Without hearing what you are talking about, I can't say what is causing your problem. The biggest mistake that I have noticed is: People have an 'ideal' eq curve that they gravitate toward. That is fine if you apply it to the mix as a whole, but if you apply it to all the individual tracks, all the instruments will be fighting for the same sonic space. You have to fit the instruments together like a puzzle. You might not like 400hz, but some instrument in the mix needs to have some.

 

[Autist]

Many consider mixing to be the most creative part of the process. I can spend more time on a mix than I did on tracking.

Having done both, I am not sure that there is any real advantage to mixing in analogue as compared to digital. Of course analogue has a quality that many people prefer over digital, but digital is very wildly used commercially and can produce excellent results and be a wonderful format to mix in.

If you are layering instruments that all fall in the same frequency range there will be overlap. If you have two instruments that both take up a lot of space in the midrange (say around 300Hz) those instruments will constantly be fighting each other for that space. EQ is important here because you can reduce frequencies on one instrument that may be conflicting with another instrument. If a lot of instruments seem to overlap in a certain range, then that range will definitely be more effected and you will need to do some EQ work on tracks to balance things out. Don't try to EQ the tracks individually so that they sound good solo. EQ them as part of the mix so that they blend. A track that is EQed well for a certain mix may sound too thin, too tinny, too bassy, whatever, if soloed, when in the context of the mix it sounds just fine. You may have a sharp peak at one frequency during certain parts of the song, which is not apparent when soloing the instruments. However, when you put them together it becomes clear that that frequency needs to be reduced on one, or more of instruments. Decide which tone you want to stand out there and which one you can sacrifice without effecting the quality of the material.

Each instrument needs it's own space in the mix. Like having guests over for dinner, you don't sit everyone in a single chair and if there are a lot of guests you may not sit everyone at the same table, things just get too cluttered and people start fighting for "elbow room." The instruments need room to breathe in a mix. There is a difference between the frequencies interacting and the frequencies competing. Also, sometimes during the mixing process a decision has to be made to remove certain instruments from the mix at certain parts, or to reduce their volume dramatically.

When tracking you should always be aware of the frequency range of the instruments and think if the sound you are using will conflict with other sounds in the mix.

 

[Kainz]

Hmh.. I don't want to be rude or anything (really), but to me this sounds more like an issue with your mixing skills than any digital/analog/44.1khz/96khz problem.

Mixing _IS_ an art. As stated above, it could be compared to a puzzle. When all the pieces are lying around in a mess, you really don't see the big picture there. Only after you start to build it up, piece by piece, the big picture starts to form.

Without a good mix the song is just intruments and frequencies fighting each other; plain mess without definition. You need to create a balance. Tweak volumes, frequencies, stereo field (left&right, up&down and far&front) and so on. And to know how and where and when to tweak those.. that is the hard part. It requires good monitors and most important _good ears_. You need to learn to listen where those clashes are and how to fix them. Also you of course need to know how to use basic mixing tools like EQ, compressor, delays, reverbs and so on..

Trust me, 95% of your problems are in the mixing stage (and possibly in the tracking stage too; bad mics, mic pres, AD conversion, instruments). Also recording whole band at once won't help the clashing at all, no matter if it is digital or analog.

EDIT: Oh and regarding to that "piling of tracks". In every studio they record everything to their own tracks. There is no studio that records everything to single stereo tracks when recording the takes. This is because this way we have way more control in the mixing stage to create that "track to track definition" you are after.

 

[Autist]

EDIT: Oh and regarding to that "piling of tracks". In every studio they record everything to their own tracks. There is no studio that records everything to single stereo tracks when recording the takes. This is because this way we have way more control in the mixing stage to create that "track to track definition" you are after.

I personally thought he just meant he was recording and trying to mix many tracks of somewhat redundant material, perhaps to try to get a fuller sound, which of course is not the right approach anyhow.

 

[FALKEN]

Bob,

these pros might not know what you are talking about, but I do, cuz I've been there. You are right that the more tracks you layer, the worse your problem gets. and They are right that a lot of the problem is in the mixing. But it just seem so Easy! WRONG. it really does take years to learn how to do just like an instrument. However, there are some things you can do now to make an improvement.

-The first has already been mentioned...use different recording channels on different instruments. it helps to have a few different mics and at least 2 different pre's. that why each track will have its own "color", not the same color. then each track will start to stand out on its own without you have to actually DO anything.

-the second thing I would take advantage of, since you are mixing "In The Box" so to speak, is eq. almost every track can have most of the lows cut out and some of the highs too. with digital recording, you could see a huge improvement by cutting everything below 100 or 250 Hz on guitars and vocals. cut everything above 15kHz on bass drum and bass. its real easy, and now (a) your instruments aren't stepping on each other so much, and (b) you have removed a lot of "mud". mud usually occurs below 250 or 300. yeah a guitars fundamental note is 80, but trust me, try it. its done on commercial recordings all the time. on bass and bass drum you can also cut everything below 50, because you can't hear that crap anyway, but your speakers are trying to re-create it, if its there.

-you can also use eq to boost or cut certain frequencies on instruments, and there are a lot of threads on how to eq, but almost none of them mention that if you do more than 3 db in digital, it sounds like arse (unless you are using a high or low shelf as described above). also boosting frequencies too much will make them jump out of the mix, which is bad, because then your mix will be much quieter overall.

-stay away from too many plugins.

-you might want to lightly compress the lead track to bring it out instead of panning it left or right. although what you are doing can be pretty cool too. retro!

-record in 24 bit. 44.1 is fine.

-dont record levels too hot because your soundcard doesn't work as well at the highest Dbs, your optimum point is around -12 db, in 24 bit. it will be a little higher in 16 or 20 bit.

 

[chessrock]

with digital recording, you could see a huge improvement by cutting everything below 100 or 250 Hz on guitars and vocals.

And how is making everything sound really thin going to make a huge improvement?

 

[Autist]

One of the first things I do is cut out frequencies from instruments that either don't add anything at all to the quality of the sound, or that I know will conflict with frequencies that I have chosen to cover with other instruments. If the sonic quality of a guitar piece is intended to be emphasized in the upper mids, then I may cut some of the bottom out of it to free up space, since it isn't the sound at the bottom I am after. However, there has to be a limit to what you cut out so that you are not making the instrument sound bad when soloed, or in the context of the mix. That's why ears are used to make EQ determinations.

I don't follow any set rules of what to roll off on any instrument, I treat each as a whole new entity. If an instrument has accompanyment it may get EQed so that if I solo it then it may sound thin, but in the mix it sounds perfect. If the instrument solos in the song and frequencies need to be adjusted when there is accompanyment, I will adjust the accompanyment so that the solo instrument sounds fine either way. You have to work with the EQ to achieve a balance between sounding natural and fitting well in the mix

 

[Bob's Mods]

Great Stuff

These are all super replies and is the stuff of what great mixes are made of. Superior mixing is a subject I don't see a much of on this newsboard but mastering the mix process is a key ingredient in the recording arts. At least I am not alone on this. I have done some of the things mentioned but feel I need to take it to the next level now.

Falken, I use different mics but I only have one pre (which I made). Its squeeky clean and...using it is probaby not unlike a quality studio that uses a good mixing console to record the entire band. I do do lo-cuts but yes, I should look into more hi-cuts per track as well. I definatly stay away from too many plug-ins as I try to get what I'm looking for before the mic rather than after it. As for compression, I find I don't like it very much at all (at least the plug-in kind). It changes the track in a funny way I don't care for. I use compression as a last resort when no other method will keep the dynamics under control. I've been happier with the mixes I've done since I've gotten over the compression fix. I do record at 24 bit 44dot1kHz. You make an excellant point regarding levels too. I record my tracks at -18 dB ave BUT normalize them after. I will not do this next time. I'll leave my tracks peaking at most -12 dB and see how that works out. I have wondered about this before but never tried it as something in the back of my mind was whispering "there is a difference" but I never paid attention to it.

I realize now that I have to improve my eq'ing methods within the contexts of the mix. It's easy (relatively) getting a mix with one or two tracks. As the track count increases, your mixing skills have to be up to the task.

Bob the Mod guy

 

[Farview]

Your tracks should not be peaking at -12dbfs, they should be averaging at -12dbfs.
Think of setting the recording levels on a cassette deck (or any other analog machine) Remeber, the idea was toget the needle to bounce around 0. That 0dbvu on an analog machine is equal to -12dbfs on a digital one. You can go above -12dbfs, but that is only where the transients should live. That is your headroom.

 

[FALKEN]

And how is making everything sound really thin going to make a huge improvement?

HA! I make no guarantees.

 

[Blue Bear Sound]

Mixing 101

 

[Bob's Mods]

Blue, thats a most excellant article on mixing. It should be req'd reading. I've never really used automation for volume, panning and eq'ing. I'll be starting to use those tools now to help my mixes out.

Bob

 

[SouthSIDE Glen]

Blue, thats a most excellant article on mixing. It should be req'd reading. I've never really used automation for volume, panning and eq'ing. I'll be starting to use those tools now to help my mixes out.

And another one sees the light and learns that "mixing" is a verb! Welcome to the real world, Bob. I think you'll like it

G.

 

[kylen]

My common method of recording is track layering...piling on the tracks causes more and more loss of track mix definition.

Hi Bob - It's hard to fit everything into a post and this may not be anything but I'll ask it anyway. This is the only comment I see about your track recording - is it mono close-micing, meaning 12" or less (debatable I know) or do you get some of the room reflections in there too, maybe a more distant mic to get room reverberation?

My rules are to center the drums and bass. From there I try to isolate vocals by panning to one side and panning a single dominant instrument on the other.

Likewise this is all I can gleen on your mixing style. [Assuming] you are panning mono close-mic'd tracks then regardless of the eq and dynamic balance that you've had some fine responses about - the tracks will all be piled up 'in your face' from a 3D spatial point of view, even though they are mono panned across 2 speakers. In other words no depth or spatial stereo field.

There's some folks here more articulate and experienced than I that can explain the use of delay and reverb to simulate room reflections and reverberation but I was wondering if that is part of what's missing also?

 

[FALKEN]

Blue, thats a most excellant article on mixing. It should be req'd reading. I've never really used automation for volume, panning and eq'ing. I'll be starting to use those tools now to help my mixes out.

And another one sees the light and learns that "mixing" is a verb! Welcome to the real world, Bob. I think you'll like it

umm...automation? whats that? can I use it with tape? heh heh...

I thought rocking the faders would make mixing a "verb".

 

[Bob's Mods]

This is the only comment I see about your track recording - is it mono close-micing, meaning 12" or less (debatable I know) or do you get some of the room reflections in there too, maybe a more distant mic to get room reverberation?

My room is a small neutral sounding second bedroom. All the tracks are recorded as mono with mic distance about 3 or 4 feet for my electric to pick up what little there is of the room. Acoustic and vocals I record 1 to 2 feet away. I found that close micing everything in a questionable room causes serious flat sounding mixes.

There's some folks here more articulate and experienced than I that can explain the use of delay and reverb to simulate room reflections and reverberation but I was wondering if that is part of what's missing also?

I'm careful with reverb as not all are created equal. I usually use just a touch of Glaceverb or FreeVerb. I generally don't care for a wash of reverb as these freebees don't sound all that good if you over do them. But in small doses they work naturally. For my electric I'll go for the reverb in the amp first before I'll use a software reverb.

One thing I do do which helps alot is separate mains for different groupings. My Main A is instruments, Main B is drums, Main C is vocals and Main D is bass. I do not use Aux busses at all because I have discovered they can cause subtle phasing problems. Anyone can prove this. Just turn on any track Aux buss without any plugin installed. Then push up the aux buss gain and listen. I was rather surprised by this discovery as it makes the Aux buss unusable. I use Sonar 2.2 but maybe other seq's don't exhibit this problem. As the gain goes higher you can hear the phase distortion that is caused. What ever plugins I use are loaded into any one of the main output busses.

Bob

 

[Autist]

I forgot where I read it, but an article on mixing said that the more complex your arrangement, the more time you have to spend working with EQ and compression. It summed up my experiences fairly well...

 

[Farview]

If you are in a small room and layering, you are going to have to go with close mic'ing. The lack of clarity you are experiencing is because the sound of the room is building up on your tracks. Mic'ing an amp from 4 feet away (unless it is cranked up really loud) requires a really dead room. Otherwise, the midrange will get smeared by the 'ambiance' of the room. It isn't a big deal with one or two tracks, but once you get 10 or 12 tracks with the same room sound, you've got problems that eq won't fix.

Try using 2 mics on you guitar to get a stereo representation of the guitar in the room. You won't need as many different performances to make the big sound that you want. That will cut the number of 'rooms' in the over all mix.

You should also try to do your vocals in a different room than the guitars. That will help because the other room will have a different sound to it. You could also get closer to the mic.