dbfs levels for rhythm guitar
- Thread starter Nitsuj
- Start date
Record as hot as possible without clipping, -3dBFS peak is smart, you want a hot signal because you get better dynamics since there are 6dB of dynamic range added for every bit.
ie:24 bit=144dB of dynamic range, since there are 24 quantization points per sample, more than the human ear would ever know, but it's more accurate.
16 bit=96dB of dynamic range, since there are 16 quantization points per sample, again more than the human ear would ever notice but accuracy on the front end will assure and ensure a better sounding mix at Red Book CD quality.
If you record a low level and normalize later, then you'll have less quantization points as you expand the dynamics of the audio making it a more square wave volume shift, which may be noticed later but it seriously depends on what you doing as audio is different for every project.
What do you guys think? I might be off.
ie:24 bit=144dB of dynamic range, since there are 24 quantization points per sample, more than the human ear would ever know, but it's more accurate.
16 bit=96dB of dynamic range, since there are 16 quantization points per sample, again more than the human ear would ever notice but accuracy on the front end will assure and ensure a better sounding mix at Red Book CD quality.
If you record a low level and normalize later, then you'll have less quantization points as you expand the dynamics of the audio making it a more square wave volume shift, which may be noticed later but it seriously depends on what you doing as audio is different for every project.
What do you guys think? I might be off.
Force666
Member
- Sep 12, 2007
- 1,096
- 0
- 36
recording as hot as possible without clipping is an old analog thing....now a myth at 24 bit. If you record 10 drums tracks, 2 guitars, bass and vocals you're going to have to turn down the faders to get them to sit on the master.
2nd.....in the digital realm your AD converters are probably (read your literature) optimized for 0dbvu....which translated to digital on average is about -18dbfs rms, which means the body of your signal should be sitting around -18dbfs and shouldn't really clip higher than -6, anything higher is just eating up into your head room.
Like I said these are averages, your gear might be calibrated differently so read your manuals.
2nd.....in the digital realm your AD converters are probably (read your literature) optimized for 0dbvu....which translated to digital on average is about -18dbfs rms, which means the body of your signal should be sitting around -18dbfs and shouldn't really clip higher than -6, anything higher is just eating up into your head room.
Like I said these are averages, your gear might be calibrated differently so read your manuals.
Zombietakeover
Member
I have never read on a subject were so many ppl do so many things diffrently.....now im starting to re-think all of my stuff and how im recording IN....i got some research to do on my A/D
to those who do the whole "hot as possible without clipping" thing, try tracking some stuff that averages out to around -16 or so, and compare it to stuff that was tracked hotter than that
Thanks for all the responses. I have been tracking at various levels for the past few days and I must say that at -18dbfs, even though the mix is not that hot, the dynamics and headroom is more apparent. Everything is jumping about, its much easier to mix and I'm glad to have achieved a more musical result than before.
I use C4 as the master buss compressor. It seems that there's so much headroom left that I can actually bump it up +10db on C4 before it starts to clip. Does this mean that I can afford to track hotter and use less of the master buss compression?
Track quiet. Leave as much headroom as you can for the mastering engineer. Pulling down a fader doesn't fix tracks were recorded too hot or clipped when recorded to hard disk.
There have been many tests done by the REP community at PSW, and the general consensus is tracking softer in the digital domain is sonically superior to tracking with hot levels digitally.
Just an addition to what I was saying.....
Recording hot to tape was for a few reasons, but mainly for...signal to noise and the "saturation" from hitting tape hard. In the digital enviroment you don't have tape hiss and hitting a DAW hard sounds like ass. In the digital enviroment it's all about headroom.
Remember that 0dbVU is approx -20dbfs. So in the analog enviroment when they were pegging the VU meters they were really only at about -20 in relation to a DAW.
Loudness comes later. Mix for balance and musicality.
Djabthrash
Member
Sorry for super old thread resurrection, but i'm trying to get to know how my RME FireFace UC is calibrated and where the body of my signal should be when i track, but i can't seem to find any info on that in the user manual...
Is there a specific term i should look out for ?
Even in the technical specifications section i can't seem to find anything that gives me any relevant info for that.
Here is the user manual link : http://www.rme-audio.de/download/fface_uc_e.pdf
Here are the mic input channel specs (didn't find any converters info) :
"30. Technical Specifications
30.1 Analog
Microphone/Line 1-2, front
• as AD Line In 5-8, but:
• Input: Neutrik XLR/TRS Combo jack, electronically balanced
• Input impedance: XLR 2 kOhm, TRS 8 kOhm balanced
• Low Roll Off –0.5 dB: 18 Hz, -1 dB: 12 Hz
• Gain range: 0 dB, +10 up to +65 dB
• Maximum input level XLR, Gain 0 dB: +10 dBu
• Maximum input level XLR, Gain 65 dB: -55 dBu
• Maximum input level TRS, Gain +0 dB: +21 dBu
• Maximum input level TRS, Gain +65 dB: -44 dBu
• CLIP LED: -2 dBFS
• SIG LED: -65 dBFS"
Trevoire520
Member
