MUFF WIGGLER Forum Index
 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

What level do you aim for when recording hardware into daw?
MUFF WIGGLER Forum Index -> Production Techniques  
Author What level do you aim for when recording hardware into daw?
rekko
Hello everybody. As the title says, what sort of volume level do you aim for when recording hardware into your DAW?
matthewjuran
-10dB to have distance between the signal and noise but to leave room so digital clipping is comfortably avoided. The ideal level probably depends on the audio interface and DAW though and maybe I'll have a different general answer once more experience is gained.
electrik noize
Line Level (0dBVU) is calibrated to -18dBFS for me. So I record in at -18dBFS as that is what range my preamps and line level audio gear work best. Plenty of signal, no noise, nice and clean.
Hermetech Mastering
All calibrated here, from converters to monitors to SPL level in the room. I usually aim for peaks around -14 dBFS on tonal instruments, bit higher for percussive stuff.
rekko
electrik noize wrote:
Line Level (0dBVU) is calibrated to -18dBFS for me. So I record in at -18dBFS as that is what range my preamps and line level audio gear work best. Plenty of signal, no noise, nice and clean.


Many thanks for the reply. Would you be able to briefly explain what it means when you say 'line level is calibrated...'?
slumberjack
due to older ad convertors, i have to record as hot as possible.
proteus-ix
electrik noize wrote:
Line Level (0dBVU) is calibrated to -18dBFS for me. So I record in at -18dBFS as that is what range my preamps and line level audio gear work best. Plenty of signal, no noise, nice and clean.


References for how to calibrate such?
Jean Luc Cougar
I have older converters that max out at 48k so I record with peaks around -8 dB
PompeiiRuler
record at -12db, mix to -6db, master to -0.3db ceiling. just doing what someone else told me years ago lol
commodorejohn
I just record it as loud as I can get it without actually hitting peak at any point on the track, though admittedly this only works for sequenced multi-tracking, which is all I do.
Koekepan
In a perfect world, you'd record everything so that it peaks at 0dB, and never clips, so that you get maximum signal-to-noise ratio, before then mixing things so that your mix peaks at 0dB without ever clipping, and you'd do it all in 32bits, 384kHz.

In reality, try to record as hot as reasonably feasible, with as many bits and as high a sample rate as feasible, and mix things down as needed.

Honestly, you can probably get away with recording in 16/44.1 peaking at -18dB, and some dude listening to a 128kbps mp3 on his Beats headphones will never know the difference.
Hermetech Mastering
There's absolutely no need to try and record everything as loud as you can before clipping, this will in fact sound worse. That hasn't been the case since analogue recording and 16 bit digital went the way of the dinosaur. If you are recording at 24 bit you will achieve much cleaner results with the peaks and meat of the sound well away from digital zero. At 24 bit you can record with peaks at -24dBFS or even lower, and still have way better S/N ratio than analogue or 16 bit recording.

The reason it sounds better to record lower is that many DACs and ADCs don't have a lot of analogue headroom, and can sound pinched and crappy as they approach zero. Have heard it myself, especially with consumer/prosumer level stuff.

You have over 120dB of dynamic range and insanely good S/N ratios with digital recording these days, why not use them?
Hermetech Mastering
Regarding a calibrated recording and monitoring chain, this is the best article I've read so far, and what I used to set up my own studio:

https://www.soundonsound.com/techniques/establishing-project-studio-re ference-monitoring-levels
thetwlo
Hermetech Mastering wrote:
There's absolutely no need to try and record everything as loud as you can before clipping, this will in fact sound worse. That hasn't been the case since analogue recording and 16 bit digital went the way of the dinosaur. If you are recording at 24 bit you will achieve much cleaner results with the peaks and meat of the sound well away from digital zero. At 24 bit you can record with peaks at -24dBFS or even lower, and still have way better S/N ratio than analogue or 16 bit recording.

The reason it sounds better to record lower is that many DACs and ADCs don't have a lot of analogue headroom, and can sound pinched and crappy as they approach zero. Have heard it myself, especially with consumer/prosumer level stuff.

You have over 120dB of dynamic range and insanely good S/N ratios with digital recording these days, why not use them?

thumbs up yup.
electrik noize
proteus-ix wrote:
electrik noize wrote:
Line Level (0dBVU) is calibrated to -18dBFS for me. So I record in at -18dBFS as that is what range my preamps and line level audio gear work best. Plenty of signal, no noise, nice and clean.


References for how to calibrate such?


Well I have a UAD original Apollo that came factory calibrated. Most of the UA plugins (besides the tape sims I think) are also cal'd to line level equaling -18dBFS. So very easy to work with.

However, the article Hermetech Mastering linked to is spot on, should you want to do a deep dive into setting up your system.
Koekepan
Hermetech Mastering wrote:
There's absolutely no need to try and record everything as loud as you can before clipping, this will in fact sound worse. That hasn't been the case since analogue recording and 16 bit digital went the way of the dinosaur. If you are recording at 24 bit you will achieve much cleaner results with the peaks and meat of the sound well away from digital zero. At 24 bit you can record with peaks at -24dBFS or even lower, and still have way better S/N ratio than analogue or 16 bit recording.

The reason it sounds better to record lower is that many DACs and ADCs don't have a lot of analogue headroom, and can sound pinched and crappy as they approach zero. Have heard it myself, especially with consumer/prosumer level stuff.

You have over 120dB of dynamic range and insanely good S/N ratios with digital recording these days, why not use them?


In other words: if you buy gear that sounds crappy when you use it within spec, it will sound crappy when you use it within spec.

In a PERFECT WORLD, as referenced above, one wouldn't have the crappy gear in the first place ...
Hermetech Mastering
Koekepan wrote:
In other words: if you buy gear that sounds crappy when you use it within spec, it will sound crappy when you use it within spec.


Not really, just be aware of the issue and do your own tests to confirm or deny it.
ignatius
Jean Luc Cougar wrote:
I have older converters that max out at 48k so I record with peaks around -8 dB


it's the bit depth that matters most when recording. 24bit recording is ample for decent signal level w/low noise floor.

i still record at 24/44 and have no complaints. 96k isn't relevant all the time. sometimes it's worth it to record at high sample rates but typically the difference is negligible.

if you're recording orchestras and solo piano then higher sample rates might be worth it but only just.

clocks in many (not all) converters are still shitty. less so than previously but once you get over 44.1k the clock is just inserting errors.

but do what sounds best to you of course. everyone's set up is different.

and +1 to hermetech's post up there about things sounding crappy due to shitty analogue front end that doesn't handle things well when pushed to zero
Jean Luc Cougar
Thanks for the tip! I do use 24bit, so I guess I am good. Should I lower my input volume?
proteus-ix
Hermetech Mastering wrote:
Regarding a calibrated recording and monitoring chain, this is the best article I've read so far, and what I used to set up my own studio:

https://www.soundonsound.com/techniques/establishing-project-studio-re ference-monitoring-levels


Thank you! I shoulda just searched SoS. I actually found a quick and dirty guide in a Presonus monitor manual. Having been doing this for almost 3 years now, I'm still shocked at how much I don't know that wasn't included in manuals for products I bought from supposedly pro audio companies. You'd think they'd want to give people the minimal info needed to get good results with their products...
Futuresound
proteus-ix

I’ve been at this 20+ years, and I still learn stuff pretty much every day.

I think it’s great.

Can’t argue that manuals often leave much to be desired though
Panason
More good stuff here:

https://www.soundonsound.com/techniques/gain-staging-your-daw-software
umma gumma
excellent stuff, thx for the links all!
naturligfunktion
Hermetech Mastering wrote:
There's absolutely no need to try and record everything as loud as you can before clipping, this will in fact sound worse. That hasn't been the case since analogue recording and 16 bit digital went the way of the dinosaur. If you are recording at 24 bit you will achieve much cleaner results with the peaks and meat of the sound well away from digital zero. At 24 bit you can record with peaks at -24dBFS or even lower, and still have way better S/N ratio than analogue or 16 bit recording.

The reason it sounds better to record lower is that many DACs and ADCs don't have a lot of analogue headroom, and can sound pinched and crappy as they approach zero. Have heard it myself, especially with consumer/prosumer level stuff.

You have over 120dB of dynamic range and insanely good S/N ratios with digital recording these days, why not use them?


I had no idea, but I'll try this from now! Thanks smile Usually I did record as hot as possible Dead Banana but since I while back I have tried a different approach, and my later music sounds as good as the previous one (if not better).
Hermetech Mastering
Cool. I've heard a lot of people say that when they back off the levels (depends on the source of course, and if you are talking peak or RMS etc., but donlt be scared to have peaks as low as -12dBFS for steady state sources, could be higher for drums) their mixes suddenly sound way more open and clear.

Of course is more pertinent if you are recording acoustic and analogue sources, if you are doing everything ITB there may be less to gain.
BenA718
rekko wrote:
Hello everybody. As the title says, what sort of volume level do you aim for when recording hardware into your DAW?

I am typically looking for an average of -9dBFS on my master bus.

To achieve this I get each track to average at -14dBFS. (My nearfield monitors are calibrated to 70dB at -14dBFS with pink noise)

Tracks are sent to buses for individual processing and then those buses are bused together. (example: I have separate buses for kick and snare but those feed into a single bus for all drums). This gain staging keeps my levels consistent.

Each of those main buses then go to four final ‘mix bus’ buses. Each bus is crossed over at a different frequency and have slightly different processing. I have one for low frequencies, low mids, upper mids, and high frequencies. Thsese mix buses then go into my master bus. Typically this is also where I add Waves NLS for coloration.

Gain staging is very important to maintaining the dynamics of your mix, but level is only a small part. EQ, dynamics processing, and automation make that static mix come alive.
dubonaire
BenA718 wrote:
Typically this is also where I add Waves NLS for coloration.


Your post was interesting and I'm not in anyway trying to be smart, but I really do find the idea of an analogue summing plugin to be amusing.

I do wonder though, if I'm using analogue-modeled plugins on my individual tracks and busses, what exactly is it that an analogue summing plugin will add, given that I already have 'analogue warmth' modelled on my individual tracks and busses?

I also wonder how much analogue modelling a signal can take. I guess everything needs to be dialled in fairly subtly.
BenA718
dubonaire wrote:
BenA718 wrote:
Typically this is also where I add Waves NLS for coloration.


Your post was interesting and I'm not in anyway trying to be smart, but I really do find the idea of an analogue summing plugin to be amusing.

I do wonder though, if I'm using analogue-modeled plugins on my individual tracks and busses, what exactly is it that an analogue summing plugin will add, given that I already have 'analogue warmth' modelled on my individual tracks and busses?

I also wonder how much analogue modelling a signal can take. I guess everything needs to be dialled in fairly subtly.

Horses for courses. Grab a free demo and try it, I suppose! smile

The NLS is subtle, which is why I like it. The way I have it set up in my template is each mix bus has its own console chosen for that frequency band; the EMI handles low mids differently from the SSL, for example.

There is no right or wrong and having channel strips on the instrument buses would probably have the same effect, but a) that’s a lot more CPU overhead just for running audio and b) the NLS has several desks modeled so I don’t need to purchase multiple plugins. Plus, to my ears, I prefer the sound of all of the buses being summed together instead of being processed separately. But your mileage may vary, and other cliches!
dubonaire
BenA718 wrote:
dubonaire wrote:
BenA718 wrote:
Typically this is also where I add Waves NLS for coloration.


Your post was interesting and I'm not in anyway trying to be smart, but I really do find the idea of an analogue summing plugin to be amusing.

I do wonder though, if I'm using analogue-modeled plugins on my individual tracks and busses, what exactly is it that an analogue summing plugin will add, given that I already have 'analogue warmth' modelled on my individual tracks and busses?

I also wonder how much analogue modelling a signal can take. I guess everything needs to be dialled in fairly subtly.

Horses for courses. Grab a free demo and try it, I suppose! smile

The NLS is subtle, which is why I like it. The way I have it set up in my template is each mix bus has its own console chosen for that frequency band; the EMI handles low mids differently from the SSL, for example.

There is no right or wrong and having channel strips on the instrument buses would probably have the same effect, but a) that’s a lot more CPU overhead just for running audio and b) the NLS has several desks modeled so I don’t need to purchase multiple plugins. Plus, to my ears, I prefer the sound of all of the buses being summed together instead of being processed separately. But your mileage may vary, and other cliches!


I sum through an analogue desk, but it's not special and I might not always do that. I've contemplated the Neve summing desk. I really appreciate your post and might try that path.
BenA718
I have a 16 channel analogue desk as well, it's an old Mackie, and it definitely works great for summing but the amount of cabling required is nuts! lol

I did an entirely analogue mix a few years ago and there were things I liked about it and things I didn't. Overall, my impression was that it's a fun skill to have but not worth it for the compromises it imposes, at least in my case. If I had a huge desk I might feel different.

For mixing synths, I find that reamping the synth tracks through a mic'd amp and blending those tracks into the direct tracks adds much more than analogue summing does.

One thing that an outboard analogue desk does add is the trace amounts of noise which I feel is probably the real Secret Sauce here ... I don't like dead silent tracks, and a bit of noise really glues together a mix!

This is something I mixed last year using the method I described above. WARNING: this is straight up neoprog stuff so if you don't like late 60s/early 70s English prog you will probably find it very boring and pretentious! smile

[s]https://soundcloud.com/as-follows/purple-prose[/s]
Zerius
@BenA718 is that SoundCloud track, the stereo output recording of your mackie mixer ?
BenA718
Zerius wrote:
@BenA718 is that SoundCloud track, the stereo output recording of your mackie mixer ?

No, that mix was done ITB.

This is one of the analogue mixes I did through the Mackie.

[s]https://soundcloud.com/as-follows/the-eye-of-fire-and-fear-part[/s]
dubonaire
BenA718 wrote:
This is something I mixed last year using the method I described above. WARNING: this is straight up neoprog stuff so if you don't like late 60s/early 70s English prog you will probably find it very boring and pretentious! smile


Not a prog fan but I don't find it boring or pretentious, or at least I don't care if it is pretentious.

I've only listened to it on computer speakers and it sounds OK on those so my guess is it's really well mixed.
Hermetech Mastering
Sounds great on the ATCs here too. Good stuff.
BenA718
Hermetech Mastering wrote:
Sounds great on the ATCs here too. Good stuff.

Thanks! I saw the equipment list on your site, very impressive!
Hermetech Mastering
BenA718 wrote:
Hermetech Mastering wrote:
Sounds great on the ATCs here too. Good stuff.

Thanks! I saw the equipment list on your site, very impressive!


Cheers, yeah I had to sell my modular to obtain a lot of that. smile
leeski
Thanks for the Tips you lovely Peeps nanners
DickMarker
I aim for about -18dbfs but don't sweat it if peaks get a little hotter now and then depending on the material.
Personally I just find things come together a little better mixing-wise when RMS is around that level. No idea why, just what works best for me after plenty of trial and error.
calaveras
that whole thing about recording hot to 'get all the bits' is trash. I heard that from some Guitar Center dork decades ago and it screwed up my first few years of digital recording.
You want to leave headroom for transients that are too fast for the meters to catch. Think of the strike of a pick on guitar strings. Stick on cymbals especially ! But also even vocals and other stuff you would not consider having a transient attack do in fact have quick bits of detail that fly by too fast to register on the peak reading, VU calibrated meters.

Also, every time you double your number of tracks you lose 3db of headroom. Then you have to pull down the faders to avoid clipping the mix if you recorded too hot! It will sound better in the long run if it all just goes though at 0 gain 0 attenuation.

I aim for -18dbfs as this works best with my external gear and internal plugins. Though ironically some plugins seem to have been engineered for -3dbfs signals! They don't respond until you are clobbering them with level.
(like Logic's native compressor plugin!)
MUFF WIGGLER Forum Index -> Production Techniques  
Page 1 of 2
Powered by phpBB © phpBB Group