Can someone give me a brief explanation of the recommended loudness values for a simple mixdown (for a amateur music CD or MP3’s)? I’ve had a good read of the manual and got a bit confused with how the broadcast scale standards and Alignment level standards etc etc relate to making a simple CD recording. (After a bit of googling I have a vague grasp n the terminology).
All I want is a simple ‘baseline’ idea of what I should be aiming for. So far, I’ve just been using Control Room Meter, set on Digital Scale and just keeping the peak about -2dB. I’m not even sure where I got the idea to aim for that!
I read (in the manual) that ‘The recommended value for the integrated loudness is -23 LUFS’, but if I work on that value, the Digital Scale value is about -12dB, and obviously significant quieter. Also, I don’t know if that is referring to home stereo volumes, TV, Radio or Film scoring volumes (or anything in between!).
I’m assuming audio devices (home stereos, mp3 players etc) have been designed go handle a certain level of input dB - I just want to make sure I am working wisely within those parameters, without having my songs significantly quieter than something I’d download off of iTunes.
Bottom line:
If you end the day with a mixdown that doesn’t clip when bounced to a 44.1khz 16bit .wav file, you’re fine. You can maximize, normalize, compress, expand, limit, whatever the frak you want as long as the end result doesn’t clip.
Should you? I dunno. That is your decision based on the goal of the artist, your technical limitations, and the target audience.
If you want an in-depth explanation of this issue, feel free to PM me and I’ll send you my Skype name so we can chat about it in real time. Typing out a full explanation of this issue would be so much work I’d only do it if I was on the clock.
I think when you feel you’ve got some idea of whats going on that is the time to listen to the professionals. Without context, what they’re saying won’t be very helpful.
In my opinion, the time you’d spend doing that would be better spent working in your DAW and finding out what happens when you do things to the dynamics of your track.
We’re in a digital world now – you can duplicate tracks without signal degradation, click “undo,” re-amp, rewire, and learn by doing.
Then you can shoot files to Soundcloud, Dropbox, your phone, your laptop, iPod, a CD, DVD, etc, and learn by listening to what you learned by doing.
Thanks all for the info! Yes, the scope of this all gets a bit overwhelming, but I like what these standards are trying to achieve - I’m not a fan of the mega-compressed stuff. Hurts my brain.
I’ve been reading lots of articles, from stuff about K-sytems & calibrated monitors to others explaining the tech talk (one here). In that article it said ‘the max value of normalized program material according to the EBU R128 standard is -1 dBTP (dB True-Peak)’. So, theoretically, as these are from the same standard (R128), I would assume that ‘-1dBTP’ is similar to ‘-23LUFS’ - but it doesn’t seem to be from my experiments with C7. Obviously there are differences with the way the figures are achieved, but they don’t even seem close. Makes it all a bit confusing really!
So, I’ve been doing a comparative listen to music from different era/genres while looking at their loudness data, waves and EQ curves using C7 and the inconsistency of volume levels is dazzling - it does indeed make it incredibly hard to know where to start!
For now, as a new-to-Cubase hobby recording artist (mixing/amateur-mastering myself) I think I’ll work on about ‘-2dBTP’ - and try to keep a decent amount of dynamic range - how does that idea sound?
Ok, bit more experimenting . . . . bit more researching . . still got lots to learn.
K-14 meter and -23LUFS integrated loudness seem to be a reasonably comparative gauge of things in C7, measuring the average loudness or perceived loudness. This makes sense to me, especially with music with high dynamic ranges.
But, do mastering engineers (lets say latest top40 hits music) adhere to any of this? If I import a .wav directly off of a CD in to Cubase, they mostly aren’t even close (way higher) than these standards. I a bit confused with the inconsistency of it all!
As I previously posted, I think I might just use a true peak measure as a limit reference, but stick with just using my ears to keep the musicality of my recordings under that peak as sweet as possible.
Until I understand all the above . . . . . . . and then can utilize that information as an improvement!
Don’t worry about dBTP values, or any values frankly; just make sure none of your meters are lighting up red.
I can’t express enough how important it is not to get fixated on the amount of data being thrown at you when you work in a DAW or on a digital mixing board. It can be overwhelming, but at the end of the day you want something that sounds good, not something that meets a target data value.
Try to remember, this is the only data feedback a lot of studio engineers have used to define the overall dynamic range of a lot of really stunning sounding albums:
Side note: do you have a pair of MDR-7506’s yet? If not, you’re going to want to order a pair. You need brutally honest cans to really digest dynamic ranges and tricks to increase perceived loudness.
a) Human ears do not act as linear response transducers
b) The human brain’s “how loud?” processing system can be tricked
c) Most normal people have bad sound systems, bad headphones, or bad ears.
d) When broadcast over the radio, the track will be brick-wall compressed anyways
e) Top40 style music sounds good when its loud. Jay-Z’s Black Album is compressed to hell and back, but damn if that isn’t great to pump through a nice stereo.
Basically, you’re going to find the “standard” values on most top40 is the maximum dynamic range of a 16bit wav.
BTdubs, you should study The Joshua Tree. Make sure you have lossless rips of it, and literally study the thing. It is a masterclass in mastering, especially dynamics.
You only really need to leave quite a bit of head room if its going to be mastered through analog gear dude.make sure you have set up your gain stage before you start throwing plugins on your tracks and keep your levels nice and conservative throughout your mix your master/end product will be a whole lot easier from the bat.
February 2014 Sound on Sound magazine. Article on that and how it’s currently changing.
Try to keep everything at -6 to -10db on the masters through the Project and at mixdown / export time put the faders up to 0db. Should get you there. Though you might have to go lower to -12 / -14 at times.
Andyath. You do have to watch the headroom as in the end SPEAKERS are not “digital”. SPEAKERS are definitely analog. So are buds and headphones. So are EARS.
Sorry where did I say you didn’t have to leave or watch ANY headroom?
I said that you should leave good headroom for analog mastering situation.
if its going to stay in cubase sure you still need to leave some headroom.being floating point just not as much.then maybe some compression and a limit lets say that set to -2 As long as the monitoring levels is turned down how will this effect ears, speakers, headphones?
andy, it will affect speakers/ears/headphones because once a transducer has to turn the signal into sound it will be limited by the ability of the transducer to faithfully replicate all the frequencies at their given amplitude that your final mix is telling the transducer to replicate, and will be limited by the transducers in your skull (ears) ability to tell your brain you’re hearing all those frequencies at the intended amplitude of the final mix. It is complicated.
I suspect this is probably the tipping point in this thread where it is going to turn into bickering and a mess of opinion and fact so I’ll be stepping out. Be civil folks!
No bickering here. Sorry. Andy. I thought it might have been taken the wrong way by some and I felt it needed clarifying. Only mentioned you because that’s the quote I referenced. The audio world has been maxed out by “no headroom” with digital for a long time now and it is changing apparently.
In the day of digital (and 32-bit float), what you do in the chain doesn’t really matter.
That said, many plugins expect a certain input volume to work as intended, especially simulations of analog effects.
While still a newbie in the whole production scene, I must say my mixes sound better when everything other than the master channel is peaking at around -8db or even less rather than near the 0. It also gives you a lot more room to futz with with the faders. I pretty much drop all faders to around -8db when I start, rather than unity.
Also read a bit on the plugins you use, because some act a bit differently than others.
For example, my instructor just told me that the V-series from Waves expects much lower input levels since it’s simulating vintage effects that used to get lower input levels at that time, which changes the signal quite drastically. It’s all written in the manuals.
In the master channel, I have a bus compressor and, since I master for myself, finally a maximizer which I truly do not want to push too hard, because it clearly distorts the sound when I do. The final track has true peak at a bit under -1.0dbfs, with the maximizer only activating rarely. The sound stays dynamic, and for me, loud enough.
Yeah, I read recently too about certain Waves VST’s (simulators) working best at lower inputs (just like the analogue originals) - it’s something I never considered!
I’ve now got a better gist of what I should be aiming for (for mixing and hobby-mastering) - now it’s time to just get the music as good as I can in the first place!
I came to offer the SOS but I was beaten! Good supporting article to the other information out there.
Since I was beaten, I’ll add this.
DON’T mix TOO low. On the other end of the spectrum (leaving aside Fletcher Munson issues with too low monitoring) you can create problems with (lack of) good compression, artificial noise coming from these “analog” plugins, metering, referencing, and not being able to trigger proper amounts of saturation/distortion because your sweet spot is too low.
A faux limiter on the master to turn on and off is not the same. Find a sweet spot to mix, where you have staged you compressors and level based processing before arranging/writing/mixing. VSTi (if you use them) velocity levels, please sus that stuff out along with compression. Determine your highest high and the lowest low from as deep as the modulation matrix inside a synth all the way to the grouped parallel busses. Make sure you’re triggering healthy but not over compression.
When it’s too late, it is too late to do this. This is my experience, after findings of working too low and having to make gross overcompensations to get back, but losing tremendous balance and feel.
