| Author
|
Bouncing to Audio FAQ
|
willsanquil
IsraTrance Full Member
Started Topics :
93
Posts :
2822
Posted : Aug 13, 2010 00:38:07
|
Couple questions regarding bouncing something to audio, notably relating to mixing volume and whatnot. I am using Ableton as my DAW so while I'm sure some of the questions are applicable to all DAWs, some might be Ableton specific.
I use two methods for bouncing, one if there are no returns/sends on the track and one if there is. If there are no returns, then I just freeze the MIDI track and copy the waveform down to another audio track.
If there are returns (which freeze will not capture), then I set an audio track to Resampling, solo the track I want to bounce and hit record, so I get all those lovely delays and reverbs recorded as well.
These both work well and good, but I'm wondering if the level at which the track is set prior to bouncing matters in terms of the quality of the sound? (I don't look at levels, I usually have everything set to 0 db and bounce from there, and do volume mixing at a later stage) I can only assume that with field recording you can't scream too loud into the microphone or else you will get clipping, right? Does this apply to in-the-box mixing as well?
I ask, because I have noticed that a lot of my songs have a shitload of different channels in them, and I would love to compress that down by bouncing all the little parts and when possible putting them on one or more audio tracks, this way I can cut and chop and manipulate audio without scrolling through pages of MIDI data.
However - mixing these bits. It is unfortunately not a perfect world so I'm certain that I will need to adjust volume between the different parts that I bounce, and obviously adjusting the global volume on the audio track that contains all the bits is not going to be effective...I figured i'd just adjust the volume at the clip-level?
I know this is kind of a stream of consciousness rant and not a particular question, but any thoughts or comments on...any of it I'm sure would be helpful.
cheers!
  If you want to make an apple pie from scratch...you must first invent the universe
www.soundcloud.com/tasp
www.soundcloud.com/kinematic-records |
|
|
|
supergroover
IsraTrance Junior Member
Started Topics :
39
Posts :
1505
Posted : Aug 13, 2010 13:44
|
I would recommend to bounce down to 24 bits 44 khz. You can also go for 16 bit if you don't have much space.
It is best to bounce down with the volume high but not clipping of course. The higher the volume the more you use the information that can be stored in the 24 bits if i am not mistaken. So it will sound the best. that is also the reason why alot of sample cd's have high volume. Maybe Colin can add some info about this or someone else.
Afterwards you can easily lower the volume of the sample.
soundcloud.com/supergroover |
|
|
|
Magus
IsraTrance Junior Member
Started Topics :
22
Posts :
61
Posted : Aug 20, 2010 21:38
|
When do u guys normally bounce to audio? I work alot with the midi files as i go along and i think that since im not sure i want to keep what ive got or if i want to work some new patterns or melodies after bouncing, it will all be mess. Then again, im quite ignorant and unexperienced  |
|
|
|
willsanquil
IsraTrance Full Member
Started Topics :
93
Posts :
2822
Posted : Aug 20, 2010 23:26
|
I have this thing with bouncing where I want to keep everything in MIDI just so that I can adjust it later, but I find when I actually make that leap to bounce it down then I can actually progress and get shit done.
To remedy the horror of potentially losing out on tweakable data, I just save what I was working on patch wise or MIDI wise into a preset or folder to grab later if I need to - or just save the project, then re-save a copy as 'songtitle Bounce' and then bounce *everything* down and delete the MIDI data, then start chopping and audio editing.
If you want to make an apple pie from scratch...you must first invent the universe
www.soundcloud.com/tasp
www.soundcloud.com/kinematic-records |
|
|
|
orgytime
IsraTrance Full Member
Started Topics :
120
Posts :
1703
Posted : Aug 21, 2010 02:54
|
|
Axis Mundi
Axis Mundi
Started Topics :
75
Posts :
1848
Posted : Aug 21, 2010 03:08
|
In my ableton studio setup, I run separate MIDI channels for certain kinds of CC messages per VSTi. In the VSTi track itself I put the effects racks, and set up a number of audio channels per VSTi which accept their inputs from the VSTi channel, armed, and disabled. I always render in real time, by triggering clips. That way I can render multiple versions of the VSTi output including realtime tweaks and still preserve all the MIDI data... not only that, but as each MIDI channel accepts/stores different kinds of MIDI inputs, I can preserve the automation data inside of dummy or pseudo-dummy clips for mix-matching with other clips.
I try to capture audio between -10 and 0 db, and if I ever notice any of my resampled audio clipping, I'll throw out the clip or cut out the part that goes above 0 db and try to salvage the rest.
If you are using an external synth with standard MIDI implementation, such as a Virus (like me), I can map my controller (in my case, SL MKII via Automap) to send automation data through a dedicated MIDI channel into the MIDI clip of the VSTi track. Thus, all of my automation is captured within the clip envelope and not necessarily dependent on Arrangement View automation.
|
|
|
|
Magus
IsraTrance Junior Member
Started Topics :
22
Posts :
61
Posted : Aug 21, 2010 15:39
|
|
Working in cubase and by looking at mdma's free tutorials i discovered a great trick for real time bounce in cubase. Add a group channel track and then an audio track. Route the output of what u want to the group track and then set the new audio tracks input to the group channel. That way u get real time bouncing! |
|
|
|
-=Mandari=-
Mandari
Started Topics :
28
Posts :
655
Posted : Aug 21, 2010 15:54
|
well,
i tend to set the level of things where they sound best. so if i wanna bounce down, they already fit the mix somehow. imo it´s only important you avoid clipping. if there´s no clipping at any stage of processing, you simply dont need to do anything. sure you can adjust level after, no worrys. thing you got if you scream into a mic is not given here. and also theres no big changes and difference in the audio material i guess (like there is if you speak or scream into a mic).
according to the wordlength n stuff i have to say i did that for long time like already mentioned here. 24bits/44khz. anyways i found later on that ableton is always working on 32bit, so it will dither down bits if you bounce 24bit or less. so i came to the point that it makes no sense to leave the decision to ableton. cause i dunno how ableton dithers down at the end. so to avoid this i just stopped bouncing to 24bit and leave everything 32bits.
that way imo i got the most out of ableton cause im able to decide on my own how to dither down and which algorithm to choose. it sounds far better to me since im working that way and far more space left now.
so i dont care about space on my disk, i rather go buyin a bigger one. the best sonic solution in ableton is to kep the 32bits its workin on.
just my few thoughts about that, hope kinda helped 
cheers: stephan
FUCK GENRES, LOVE MUSIC!!!!
http://soundcloud.com/mandarimedia
http://banyan-records.com |
|
|
|
Scolopendra
Started Topics :
9
Posts :
64
Posted : Aug 21, 2010 16:14
|
Quote:
|
On 2010-08-21 03:08, Axis Mundi wrote:
In my ableton studio setup, I run separate MIDI channels for certain kinds of CC messages per VSTi. In the VSTi track itself I put the effects racks, and set up a number of audio channels per VSTi which accept their inputs from the VSTi channel, armed, and disabled. I always render in real time, by triggering clips. That way I can render multiple versions of the VSTi output including realtime tweaks and still preserve all the MIDI data... not only that, but as each MIDI channel accepts/stores different kinds of MIDI inputs, I can preserve the automation data inside of dummy or pseudo-dummy clips for mix-matching with other clips.
I try to capture audio between -10 and 0 db, and if I ever notice any of my resampled audio clipping, I'll throw out the clip or cut out the part that goes above 0 db and try to salvage the rest.
If you are using an external synth with standard MIDI implementation, such as a Virus (like me), I can map my controller (in my case, SL MKII via Automap) to send automation data through a dedicated MIDI channel into the MIDI clip of the VSTi track. Thus, all of my automation is captured within the clip envelope and not necessarily dependent on Arrangement View automation.
|
|
I've read through this a couple of times and I have some trouble understanding the whole procedure! But it sounds like something I've been wanting to learn for a long time, since I am also in Live, and I really want to grasp more of the audio-editing/bouncing possibilities. =P Could you try to explain the procedure in another way pleaaase? PM is more than welcome!=)
dancing and singing |
|
|
|
|
