Hello KVR, I thought you guys might be able to help me.
Ok. So... I'm getting ready to mix, but I may have done (A LOT OF) things wrong, and I need to know where to go from here? I feel lost.
I've got several songs that are finished. Each song is mostly virtual instruments, a small bit of audio here and there, and vocals.
I have several questions about what my next steps are or should be. (In Cubase if that's relevant).
First I'll tell you what I THINK I have to do, please tell me if I'm wrong and what I might be missing.
1. BOUNCING DOWN AS IT PERTAINS TO PANNING
I've always been taught to mix split channel audio tracks, so that I can place different things at different spaces in the mix.
When looking at the interfaces on my vst instruments however, IF they have a pan setting, it seems to be one dial for the entire stereo file, not two separate controls for each independent channel.
Therefore, don't I have to bounce all of my MIDI tracks down to split channel audio files so that I can set the panning on each channel?
(Side question. Do vst instrument makers pan all instrument patches hard L&R? I ask this because the few tests exports I've made of different instrument tracks, my two split audio files don't sound like what I had on the instrument track until I pan the left 100% L and the right channel 100% R.)
2. BOUNCING DOWN AS IT PERTAINS TO CLIPPING/HEADROOM
On all my vst instrument tracks, if there is any red in the track meters, don't I have to go in to the insturment interface and lower the volume until the red is gone, before I bounce to audio?
Where should the top volume of all of my instrument tracks be hitting in the DAW meters before I bounce down, -6 or -0?
Also, do I want to monitor the DAW meters at “Input”, “Poster Fader” or “Post Panner”?
Do I worry about the individual track meters at all, or do I just focus on the final stereo out meter?
These are my big questions. What else am I overlooking? Any suggestions would help.
Thanks!
Getting Read To Mix. What Do I Do Next?
-
- KVRAF
- 20941 posts since 22 Nov, 2000 from Southern California
Leave the synth's pan in the middle. In Cubase, use the Stereo Combined Panner, then hold down Alt and drag left or right.Domlun wrote: Mon Aug 22, 2022 5:34 am 1. BOUNCING DOWN AS IT PERTAINS TO PANNING
I've always been taught to mix split channel audio tracks, so that I can place different things at different spaces in the mix.
When looking at the interfaces on my vst instruments however, IF they have a pan setting, it seems to be one dial for the entire stereo file, not two separate controls for each independent channel.
Therefore, don't I have to bounce all of my MIDI tracks down to split channel audio files so that I can set the panning on each channel?
2. BOUNCING DOWN AS IT PERTAINS TO CLIPPING/HEADROOM
On all my vst instrument tracks, if there is any red in the track meters, don't I have to go in to the insturment interface and lower the volume until the red is gone, before I bounce to audio?
Where should the top volume of all of my instrument tracks be hitting in the DAW meters before I bounce down, -6 or -0?
Also, do I want to monitor the DAW meters at “Input”, “Poster Fader” or “Post Panner”?
Most people would say turn it down. It's not supposed to matter with floating point but I'm sure someone with more up-to-date knowledge than me can chime in about it. I would just say don't overthink this part too much, just keep your levels low everywhere so you have headroom to compress the master.
-
- KVRAF
- 3506 posts since 12 May, 2011
So far as panning goes, "what you've always been taught" seems to refer either to ProTools, which can have mono channels or to real-world mixing desks, whose channels are all mono. Follow Uncle E's advice.
Headroom - the "rule" is to keep the meters at around -12dB, this allows you some flexibility to raise the volume on the quiet bits, instead of having to turn every other track down.
The only time you have to worry about "going into the red" is at the final output (where the audio hits the hardware converters) or if you are using so-called "analogue" plugin emulations of vintage gear - these can distort badly (by design!) if the input or output gain is too high.
Good luck!
Headroom - the "rule" is to keep the meters at around -12dB, this allows you some flexibility to raise the volume on the quiet bits, instead of having to turn every other track down.
The only time you have to worry about "going into the red" is at the final output (where the audio hits the hardware converters) or if you are using so-called "analogue" plugin emulations of vintage gear - these can distort badly (by design!) if the input or output gain is too high.
Good luck!
-
- KVRist
- Topic Starter
- 59 posts since 22 Aug, 2022
Ok. I never knew about that stereo combined panner. I just checked it out. Cool.
But this info has me with more questions. What about drums? I'm talking about in something like EZDrummer. The kit is already panned within the application. So... I'm assuming I dissolve the midi drum file to separate tracks/instruments first, but what do I do with the panning? Aren't they already panned within the app? Do I set all the panning to zero in the application before bouncing the individual dissolved tracks to audio? I mean, if I bounce the dissolved tracks to audio before setting all panning to zero, then whatever way EZDrummer has the panning on each drum will be printed to that bounced audio, right? That doesn't sound right.
I mean, I may have twenty five tracks of EZDrummer on one song. Do I dissolve every single track, turn all the panning to zero, bounce, and then reset the panning to what the application had each drum at? That will take months to do just the drums.
What do most people do with things like EZDrummer tracks?
Also, when do I normalize anything? The internet tells me I MUST NORMALIZE ALL TRACKS. When and how do I do this?
But this info has me with more questions. What about drums? I'm talking about in something like EZDrummer. The kit is already panned within the application. So... I'm assuming I dissolve the midi drum file to separate tracks/instruments first, but what do I do with the panning? Aren't they already panned within the app? Do I set all the panning to zero in the application before bouncing the individual dissolved tracks to audio? I mean, if I bounce the dissolved tracks to audio before setting all panning to zero, then whatever way EZDrummer has the panning on each drum will be printed to that bounced audio, right? That doesn't sound right.
I mean, I may have twenty five tracks of EZDrummer on one song. Do I dissolve every single track, turn all the panning to zero, bounce, and then reset the panning to what the application had each drum at? That will take months to do just the drums.
What do most people do with things like EZDrummer tracks?
Also, when do I normalize anything? The internet tells me I MUST NORMALIZE ALL TRACKS. When and how do I do this?
-
- KVRAF
- 3506 posts since 12 May, 2011
Why are you bouncing everything to audio?
-
- KVRAF
- 20941 posts since 22 Nov, 2000 from Southern California
Can you assign the EZdrummer parts to different outputs? If so, you can bounce them down all at once.
Honestly, most EZdrummer users just use whatever panning/level is set in the kit.
Honestly, most EZdrummer users just use whatever panning/level is set in the kit.
-
- KVRist
- Topic Starter
- 59 posts since 22 Aug, 2022
I'm not sure myself except that everyone on the internet is all like "you must bounce down all midi tracks to audio before mixing". Then they go on to talk about normalizing all your audio tracks before mixing.
It's just talked about like normal protocol. Seems like a lot of extra work is you ask me, but that's what they say. I thought maybe there was some sonic benefit or cpu benefit. Maybe it's just for archival purposes. I don't know.
-
- KVRAF
- 20941 posts since 22 Nov, 2000 from Southern California
Normalizing everything (to -12 average) can be cool because it allows you to create mix templates that you can apply to new mixes. I personally wouldn't normalize EZdrummer kits because they are pre-mixed.
-
- KVRAF
- 8749 posts since 24 May, 2002 from Tutukaka, New Zealand
You absolutely don't have to bounce everything down to audio. But...personally I do it quite a lot. The main reason is that I don't have a bang up-to-date supercomputer with limitless power, so it is definitely conceivable that I run out of power if everything's live.
Normalising everything once rendered is likely just old habits. There's no absolute need to do it, though you get some small benefits of knowing by sight how mix faders etc relate to each other when mixing. For example if you have a really quiet audio channel next to a normalised one, the fader for the quiet one will have to be higher than the normalised one - which gets confusing for a mix when there are loads of channels. I used to do that but not nowadays - and I record hot anyway, as you're supposed to do. i.e. set the recording level gain until you get very occasional red peaks then bring it down until they only just disappear. Then most recorded channels are at very similar levels to start with - just makes mixing easier. (You want recording hot to reduce the noise floor).
It makes sense to do that with VSTi also just so your channels are similar level at the starting point. I won't say you shouldn't normalise but be aware that as mentioned already - you might be using analogue emulation plugins, distortion etc that will increase the levels quite a bit. You can get some nasty surprises if you apply those FX to an already hot audio channel. Byy normalising, it means you're at max already. Though again, as mentioned zero is not an absolute maximum in a floating point DAW. I do think it's a good habit to get gains close though - if ever you get into using h/w there are some rules like these that definitely apply and you'll f**k up your recording if you treat it like digital in a DAW.
It's one of the nice things about digital recording in a DAW though - you can do it so many ways. Whatever works for you.
As for EZ Drummer - I never used it so don't know. Personally I like having drums all on their own channel just for things like Eqing. Though for dynamics I group them together in a group channel. If EZ Drummer is anything like Groove Agent in Cubase (what I use a fair bit) then it's easy to get each drum on separate channels to record them. Or you can play them as they come and you can still tweak panning etc even if the whole thing is sent to a stereo channel. I do think it's a good idea to have kick on a separate channel, but that's just my preference.
Normalising everything once rendered is likely just old habits. There's no absolute need to do it, though you get some small benefits of knowing by sight how mix faders etc relate to each other when mixing. For example if you have a really quiet audio channel next to a normalised one, the fader for the quiet one will have to be higher than the normalised one - which gets confusing for a mix when there are loads of channels. I used to do that but not nowadays - and I record hot anyway, as you're supposed to do. i.e. set the recording level gain until you get very occasional red peaks then bring it down until they only just disappear. Then most recorded channels are at very similar levels to start with - just makes mixing easier. (You want recording hot to reduce the noise floor).
It makes sense to do that with VSTi also just so your channels are similar level at the starting point. I won't say you shouldn't normalise but be aware that as mentioned already - you might be using analogue emulation plugins, distortion etc that will increase the levels quite a bit. You can get some nasty surprises if you apply those FX to an already hot audio channel. Byy normalising, it means you're at max already. Though again, as mentioned zero is not an absolute maximum in a floating point DAW. I do think it's a good habit to get gains close though - if ever you get into using h/w there are some rules like these that definitely apply and you'll f**k up your recording if you treat it like digital in a DAW.
It's one of the nice things about digital recording in a DAW though - you can do it so many ways. Whatever works for you.
As for EZ Drummer - I never used it so don't know. Personally I like having drums all on their own channel just for things like Eqing. Though for dynamics I group them together in a group channel. If EZ Drummer is anything like Groove Agent in Cubase (what I use a fair bit) then it's easy to get each drum on separate channels to record them. Or you can play them as they come and you can still tweak panning etc even if the whole thing is sent to a stereo channel. I do think it's a good idea to have kick on a separate channel, but that's just my preference.